1 Introduction

The importance of MSMEs lies in the fact that they account for about 90% of all private sector firms (Purwandani & Michaud 2021; Wang et al. 2022), practically in every country in the world (Epede & Wang 2022). However, several of these small size businesses struggle to implement environmental management accounting systems (Javed et al. 2022), and environmentally conscious business practices (Purwandani & Michaud 2021), showing a weak progress and commitment to corporate sustainability (Ernst et al. 2022). In the case of rural MSMEs in developing countries, making a positive contribution to the mitigation and prevention of environmental impacts is an even greater challenge (Nulkar 2017; Vásquez et al. 2019).

The lack of skills (Lee & Cowling 2015) and access to environmental knowledge, a low innovation culture, and a misperception about CE practices considering them optional, more costly (Ernst et al. 2022), and hard-to-implement alternatives with no tangible benefits (Rizos et al. 2016), make very few MSMEs interested in the Circular Economy (CE) and its eco-innovations (Uvarova & Vitola 2019). In the same way, the scarce external support, and environmental programs with narrow approaches that ignore the specifics of MSMEs, limit their transition towards sustainable production models (Uvarova et al. 2020).

The aforementioned barriers keep MSMEs in a state of environmental passivity (Wang et al. 2022), undermining climate sustainability and preservation (Coldwell et al. 2022). MSMEs as a whole are responsible for about 70% of global emissions and the shared discharge of large volumes of pollutants (Javed et al. 2022). Hence, this research is concerned on the transition of MSMEs in developing countries towards circular production models. A case study applied to a coffee and pig farm allowed to analyze how the implementation of appropriate clean technologies, from a more holistic approach, influences the transition of MSMEs towards circularity. Following the Cleaner Production (CP) transfer principles, such transition was supported by learning-by-doing and learning-by-interacting methods that facilitate the acquisition and internalization of new CP knowledge (Van Hoof 2014; Rodriguez et al. 2020). Qualitative and quantitative data were collected through a variety of tools, including interviews, on-site observations, and multi-stakeholder dialogs.

Recognizing that agrifood industry is key to the socioeconomic development of developing countries (Cortés Marín, 2007; Food and Agriculture Organization of the United Nations—FAO, 2013), the case study focused on coffee and pig production, traditional productive sectors with a high potential for environmental transformation by being highly polluting (Gamarra et al. 2018; Cárdenas et al. 2019; Fernández et al. 2020). The agrifood chain is dominated by small businesses (Vincent Sebikari 2014; Chiciudean et al. 2014; Zaridis et al. 2021) that have proliferated in recent decades in Africa, Asia, and Latin America (Reardon et al. 2021), including food growing and processing, wholesaling, and logistics provision businesses (Saweda & Anthony 2021). Despite being considered a “missing middle”, agrifood MSMEs produce two-thirds of the food consumed in developing regions; they are relevant to food security, employment, economies, rural households and the environment (Reardon et al. 2021). Their decisive role in different aspects makes their transition to more sustainable production models absolutely necessary and urgent (Briamonte et al. 2021).

The applied case study, based on a systems approach, managed the transition of the farm-system towards circularity. Easy to apply and low-cost complementary clean technologies allowed an effective interaction between the different parts and sub-systems, increasing the environmental efficiency of the farm as a whole.

This led to 86% reduction in water consumption in coffee pulping stage, 100% use of pig raising waste for biogas generation, and 100% use of solid waste from coffee processing for bio fertilizer production. The research showed that when an appropriate selection and adoption of complementary clean technologies is performed in a productive unit, neither the condition of rurality nor the size of the firm should limit its effective incursion into circularity. It was found that the simpler the clean technology, the easier it was for MSME entrepreneurs to accept, understand and adopt it.

This paper begins with a literature review of the gaps and barriers faced by MSMEs during their transition to greener production processes, especially in developing countries. This review was conducted within the framework of CP, SCP, CE, and ST concepts and principles. Subsequently, the methodology used is described, corresponding to a simple case study with a participatory approach. Then, the development of the case is presented through three phases: I. Diagnosis and action plan; II. Implementation of the action plan; and III. Monitoring and follow-up. Finally, conclusions and discussion of results are provided.

1.1 Gaps in the transition of MSMEs towards green and circular growth

Different strategies have sought to guide industrial sectors towards green growth. In 1989, the concept of Cleaner Production (CP) was first introduced by the United Nations Environment Programme (UNEP), defined as “the continuous application of an integrated preventive environmental strategy to processes, products, and services to increase overall efficiency, and reduce risks to humans and the environment” (UNEP, 2007). It is a broad term, also referred to as pollution prevention (United States Environmental Protection Agency—EPA, 1995) and green productivity. Then, around 1994, the Sustainable Consumption and Production (SCP) strategy emerged, promoting the production and use of socio-environmentally friendly and economically viable products throughout their life cycle (LC). SCP is a holistic approach that encourages the efficient use of resources, especially those that are scarce or non-renewable (Michaelis 2014), from their extraction and manufacture to the marketing of products and their disposal at the end of their useful life (Zu 2013). In the same direction, in contrast to the “take-make-waste” linear model, the circular economy emerges as a system that is regenerative by design; it leaves behind the “end-of-life” concept and promotes restoration. It encourages the use of renewable energy, and the elimination of waste and the use of toxic chemicals by designing durable goods that are easy to repair and can be reused by others at the end of their service life (Valavanidis 2018). In this way, CE seeks to maintain and to share value along the time (Vence & Pereira 2019), decreasing the dependence on finite resources and increasing the resilience of the system (Ellen MacArthur Foundation 2013).

The three strategies mentioned (CP, SCP and CE) are not mutually exclusive; on the contrary, they support and complement each other. Each one, or their articulation, can guide businesses towards circularity. This is because they all promote the efficient use of resources, and thus produce more with less, becoming Win–Win strategies with tangible benefits for the environment and the firm's productivity. Despite their multiple benefits, their adoption by MSMEs has been very low and slow (UNEP, 2001; Nunes et al. 2019), lagging behind in the field of green business (Maniu et al. 2021), especially in emerging markets (Das et al. 2020). Simply stated, clean technologies and life cycle approach (LCA) business models are failing to make it on the agenda of small firms (Nulkar 2017). Some gaps or inappropriate approaches in the processes of transferring green and circular transition strategies might have limited their adoption by MSMEs, including:

1.2 CP, SCP and CE transfer processes prioritize end-of-pipe (EOP) approach and complex technologies

Eco-innovations, also called green innovations, are key to promote and lead the transition to a green economy (Jang et al. 2015). These correspond to products, processes, technologies or services, new or significantly improved, in order to neutralize, minimize (Pichlak & Szromek 2022), or reduce the negative impacts to the environment from human or industrial activities (Vence & Pereira 2019; Jędrzejczyk 2022). The latter generated by conventional practices and technologies (Castellacci & Lie 2017). As an integral part of eco-innovations, environmental technologies in the areas of green industry transformation and climate change mitigation, are receiving increasing attention (Urbaniec et al. 2021). Their role as enabler of the circular economy is gaining more and more importance and recognition (Vence & Pereira 2019).

Technological eco-innovations are usually framed in two broad categories: clean technologies and EOP technologies (Hammar & Löfgren 2010; Li et al. 2021). Clean technologies look inside the production process, introducing eco-innovations to manage resources more efficiently, and reduce amounts of waste, air pollution, and material resource usage throughout the production stages (Yurdakul & Kazan 2020). From an external view to the production process, end-of-pipe technologies are installed at the end of it without fundamentally affecting or modifying it (Li et al. 2021), in order to treat, handle or dispose emissions and wastes caused by the activities of the firm, to prevent the spread of pollution to the environment (Hammar & Löfgren 2010; Agnello et al. 2015).

Green growth strategies (CP, SCP and CE) coincide in the importance of addressing the corporate environmental transition from a pollution prevention at source approach, as provided by clean technologies, and complementing it with corrective EOP actions. This is considering that preventive measures primarily reduce the volume, but do not eliminate the totality of waste from the production stages. The end-of-pipe technology act then as an alternative technology for cleaner technology (Agnello et al. 2015), to manage the already reduced volumes of waste resulting from cleaner production processes. Thus, EOP solutions and clean technologies are not substitutes, they are complementary alternatives in which firms might invest (Hammar & Löfgren 2010), without undermining the importance of any of them.

However, while large companies tend to adopt comprehensive approaches concerned with both preventing and controlling pollution, in the MSME sector priority is given to corrective actions, namely a curative approach. Thus, large companies conceive their environmental transformation from the inside out, innovating in the reconversion of their processes with actions such as: replacement of hazardous inputs, substitution of primary resources for secondary resources, closed production processes, adoption of alternative resources such as renewable energy and rainwater, integration of new low consumption equipment and redesign of more environmentally friendly products (Rennings 2000; Hammar & Löfgren 2010). As a result, multiple benefits are obtained from savings due to reduced consumption of materials and resources, the use of increased recycled content, improved resource productivity, reduced environmental risks, and improved image with customers (Nulkar 2017).

In MSMEs, on the contrary, environmental transformation occurs mainly outside the production process, through innovations around recycling and final waste disposal (Rennings 2000; EC 2022). Paper and cardboard recycling, being an economical and easy-to-implement alternative, is one of their favorite environmental practices (Purwandani & Michaud 2021). However, very few MSMEs focus on preventing waste generation, reusing or recycling other materials, in their own processes or products (Valavanidis 2018). Although MSMEs are highly involved in various sectors, including textile, clothing, and the durable goods sector (electrical, furniture, etc.), it is the large companies that have mostly advanced towards CE with new solutions and innovations on reuse (Valavanidis 2018).

We are facing a scenario where the final disposal and external recycling of waste have been put in first place, ignoring or leaving in second place the efficient use of resources within MSMEs. Thus, efforts to promote their circularity have been based on an incomplete interpretation and application of the concepts of CP, SCP and CE, ignoring that: CP through the inclusion of water, material and energy saving technologies, primarily promotes the elimination of waste before it is created in order to systematically reduce post-production waste going to final disposal or external recycling (UNEP 2001; Pichlak & Szromek 2022); CE and resource efficiency are key elements of the sustainability transition that are closely related (European Commission 2022). CE promotes a system of resource utilization where both the reduction, reuse and recycling of materials prevail, reducing waste to a minimum (Valavanidis 2018).

This is a critical and unsustainable scenario, especially in developing countries, where reverse logistics and municipal recycling systems are non-existent or insufficient. In many Latin American and Caribbean countries, the infrastructure needed for waste management, recovery and final disposal is not growing at the same rate as waste generation. In 28 countries in this region, it was found that only 4% of the waste collected goes to recycling, and less than 1% to composting (de Miguel et al. 2021).

In addition, MSMEs are not able to store and condition high volumes of waste in their facilities to meet the quantity and quality criteria required by the few existing industrial waste collection systems. Thus, MSMEs are simply left out of the formal schemes for special waste recovery. Insisting on corrective environmental management, which generates lower profits with recycling and high costs for final disposal and maintenance of waste treatment technologies (Agnello et al. 2015), is unfeasible, unsustainable and demotivating in the MSME context. Being an investment with no return, the EOP emphasis accentuates the belief that environmental management does not pay, becoming one of the main constraints for CP and circular implementations in firms (Vásquez et al. 2019). When MSMEs do not find tangible benefits in environmental actions, they simply discard them (Nulkar 2017) as being a non-relevant expense that does not add status to the business (OECD 2016).

Instead, pollution prevention, resource saving and reuse innovative technologies (Yurdakul & Kazan 2020), which have a direct positive effect on production processes and products (Hammar and Löfgren 2010; Graf 2015), provide in return competitive and cost advantages (Ma et al. 2018; OECD 2021), and contribute to green economic growth (Vence and Pereira 2019; Li et al. 2021; Jędrzejczyk 2022), should be mainly promoted. Until this happens, large companies will continue to lead as Green Innovators, while MSMEs will remain as Green Technological Laggards (Castellacci & Lie 2017), feeling “locked in” at the bottom of their supply chain, without the capacity to fully implement a circular solution by themselves (Rizos et al. 2016). Low innovation capacity limits the adoption of Corporate social responsibility approaches in MSMEs (Zbuchea & Pinzaru 2017).

Acknowledging that green management innovation is mainly pushed by external knowledge providers, diverse stakeholders and knowledge networks could help removing barriers to identify, transform, use and disseminate new environmental knowledge within MSMEs (Rennings 2000; Ma et al. 2018). Laggards usually possess less absorptive capacity than leaders, being more difficult for them to absorb knowledge spillovers (Smeets & Bosker 2011). The constant interaction of large companies with knowledge management organizations, including universities, increases their absorptive capacity and internal research and development skills (Castellacci & Lie 2017). Thus, strong relationships with a variety of stakeholders motivate effective learning in the firms, and the development of sustainable, innovative and long-lasting processes and products (Ghassim & Bogers 2019).

The fact that the size of the firm significantly influences the level and type of environmental practices to be adopted in the firm (Maniu et al. 2021), seems to be often ignored by those stakeholders that interact with MSMEs for transferring circularity and green growth principles to them. It is not uncommon for these stakeholders to promote actions that exceed or underestimate the capacities of MSMEs for a green transition. Such a transition could be easily achieved by using technologies in line with MSMEs reality (Alayón et al. 2022), providing the necessary knowledge and time for its understanding and integration within the firm. The simpler and clearer the new technology is, the more easily it is understood and adopted by MSMEs. Conversely, technological complexity negatively impacts on adoption of green practices (Melegoda & Dissanayake 2021). It is necessary to implement new approaches that guide the environmental transformation of MSMEs from the inside out, starting from a low complexity preventive approach and moving towards circularity.

1.3 Environmental management is based on a fragmented vision that promotes isolated initiatives and favors the modification of processes rather than products

Since its launch, the concept of CP defines that pollution prevention can be applied to both products and production processes; both options increase the overall efficiency, and reduce risks to humans and the environment (UNEP 2007). However, there are important differences between eco-product and eco-process innovations that should not be overlooked (González-Moreno et al. 2019). Process innovations occur when a good or service can be produced with less input by optimizing process steps or integrating alternative components, such as substitution of hazardous inputs, heat recovery, and water reuse between production steps. Product innovations require improvements of existing goods (or services) or the development of new goods, through the integration of new product components, and changing parts of the product or the complete product (Rennings 2000).

The launch of the SCP and CE strategies sought to highlight the importance of approaching environmental management from product innovations, specifically eco-products. The development of the latter requires the integration of environmental considerations from product strategy and idea generation to reverse logistics and end-of-life management. That is, an LC perspective that ensures the design of efficient products that are easy to dismantle and recycle; its goal is to create a green product (or service) with minimal environmental impacts during its use, and that can be easily and properly disposed at the end of its useful life (Charter et al. 2001). Thus, eco-design is essential to identify, evaluate and reduce by more than half the environmental costs of products and services from an early design stage (Kamalakkannan & Kulatunga 2021; Dahmani et al. 2022).

However, MSMEs do not always have the environmental knowledge, awareness and capabilities to articulate with green suppliers and value chains, in order to jointly design products that demand minimal resources during their lifetime (Nulkar 2017; EC 2022). Most are focused on survival rather than greening themselves or their value chains (ADB 2020). Despite being the most significant contributor to greenhouse gas emissions in every region (Fahad et al. 2022), MSMEs focused on environmentally friendly products that decrease their carbon footprint in their entire lifecycle are rare to very rare (Quintás et al. 2018; Nasih et al. 2019). Some studies in developing countries have found that eco-design, eco-labeling, and LC analysis represent less than 10% of the environmental practices adopted by MSMEs (Quintás et al. 2018). For the most part, they opt for fragmented and informal environmental actions that do not seek to improve an entire product or process, or a transition towards CE principles (EC 2022).

Such fragmented vision, derived from a reductionist approach, limits the resolution of complex problems like industrial pollution; on the contrary, adopting a systems thinking approach helps to solve them (Monat & Gannon 2015). To this end, it is necessary to see the organization, in our case the MSME, beyond a collection of isolated elements, considering it as a system with the following properties: (i) Every system is delimited by a boundary, which allows it to differentiate it from its environment and the wider system of which it forms part; both exert an influence on the system. (Jenkins & Youle 1968; Bleicher & Müller-Stewens 1996); (ii) The firm-system is a set of different independent parts working together in interrelated manner with the aim of turning organizational vision into reality. For this, its different elements, including organizational structures and processes, technologies or equipment, and people, must be coordinated to achieve a common goal, aligned with greater efficiency, effectiveness and corporate growth. The system only works and has meaning through the interaction of the elements, which mutually influence each other (Bleicher & Müller-Stewens 1996; Chikere & Nwoka 2015). If a change arises in one, it will generate a change in another; (iii) Systems can be broken down into a number of sub-systems (Daryani & Amini 2016) to be determined according to the problem under study. In general, a sub-system in an organization may be considered as some process which transforms certain input flows of money, materials, energy, resources, data, into corresponding outputs (Banathy 1967); (iv) The outputs from a given sub-system provide the inputs for other sub-systems. Thus, the performance of one sub-system may influence the performance of another sub-system, and hence cannot be studied in isolation (Jenkins & Youle 1968; Iacovidou Eleni et al. 2021).

This implies that the efficiency of a production plant or company will depend on the correct functioning of all subsystems (Jenkins & Youle 1968), and the ability to anticipate the changes that may occur if one part of an interconnected system is affected (Stewart & Ayres 2001). In this sense, the environmental efficiency of a company will be affected by the environmental performance of each subsystem, and by the collective environmental management among subsystems. Only from a broader perspective, focused on identifying the different waste streams generated in the subsystems and their potential for recovery or reuse, will it be possible for companies to achieve their maximum waste minimization potential and move towards circularity (Musee et al. 2007). The idea behind a circular economy is to treat undesirable outputs from one subsystem as useful inputs for another subsystem, creating closed or semi-closed systems (Xu et al. 2009) where resource input, waste generation, and energy leakage are minimised by closing and narrowing material and energy loops through a lifecycle design (Guzzo et al. 2023).

From this perspective, companies from various industrial sectors have managed to reduce pollution throughout their production processes, and the consequent negative impacts on the natural environment. Integrated environmental actions in the firm-system contribute to face relevant environmental situations in the broader system that contains it. For example, considerable progress has been made in the electroplating industry to maximize the reuse of rinsewater in rinsing steps (Zhou et al. 2001). The textile industry has shown progress in developing combined water reuse treatments, including flocculation, sedimentation, sand filtration, ozonation, ultrafiltration and reverse osmosis units, to obtain quality water for reuse in industrial processes (Yin et al. 2019). Consequently, water consumption and the volume of industrial effluents are reduced, decreasing pressure on watersheds and the ecosystem services they provide. In the same vein, in the petroleum industry, steam pressure boosting technology is used to recover low-pressure steam, to increase energy efficiency by reducing conventional heating fuel consumption (Goodarzvand-Chegini et al. 2023). This technological eco-innovation, in addition to leading to significant savings, contributes to climate change mitigation. In construction projects, it is estimated in advance the amount of concrete waste that will be reused on-site in subsequent activities that will require clean fill material, and the amount of remaining waste that will be sent to off-site recycling facilities; thus, improving resource recovery and minimizing waste disposal in landfills (Guerra et al. 2020). It is therefore essential to recognize the enterprise as a living system in which the production, consumption and management parts act interdependently, therefore, all inputs, transformations, outputs, stocks or sinks, leakages and hidden flows must be analyzed by looking at the system as a whole. A wider view is absolutely necessary to identify systemic opportunities for waste and resources management (Iacovidou Eleni et al. 2021; Guzzo et al. 2023).

Several contributions and discussions have emerged on the importance of systems thinking to design alternatives for the transition to a circular economy. Iacovidou Eleni et al. (2021), in their attempt to understand the way resource recovery systems operate, argue that boundaries can be space-specific (e.g., city, state, organization); resource-specific (e.g., inputs, product, substance); process-specific (e.g., paper and glass waste reprocessing); or, a combination of these. They further state that processes, actors and values act as interconnected subsystems, whose behavior affects the entire resource recovery system. Thus, the resource flow pattern between one process and another is influenced by those who are in charge of the movement and processing of that flow, namely, key stakeholders within (e.g., firm´s workers) or outside the system boundaries (e.g., waste management industry, government). Similarly, the model created by Sohal et al. (2022) to analyze the progress of MSMEs towards CE, based on the Socio-Technical Systems (STS) theory, suggests that the support of financial, public and private institutions influences the adoption of reuse and recycling techniques and technologies, and the organizational culture around waste reduction processes. As for ‘values’, these relate mainly to the environmental, economic, social and technical impacts caused by the resource recovery system in its surrounding environment or “place” (Iacovidou Eleni et al. 2021). It is therefore key for MSMEs to adopt a placed-based systems perspective in transitioning towards a circular economy, if they are to generate multi-dimensional value in the firm itself, in the ecosystem and in society. Resources valorization and waste free production add value to the organization by increasing its efficiency, and to the ecosystem by protecting natural capital through cleaner production processes. When environmental transformation arises from collaboration among multiple local stakeholders, who co-create knowledge and share it with other stakeholders, then value is created for society (Howard et al. 2022).

From a systems-based approach, Guzzo et al. (2023) propose an innovation road map to guide business transformation towards circularity, drawing critical questions about socio-technical aspects along the way: (i) Time frame: When should we act? (ii) Societal needs: Why should we act? (iii) Innovation Adoptions: What should we do? What are the innovations to be developed and implemented? (iv) Experimentation: How can we facilitate the implementation of innovations? What is the validated learning and capabilities needed to sustain the innovation? The proposed circular innovation framework put the concept into practice, by envisioning, implementing and operationalizing circular solutions. The right combination of strategies and practices help to define and implement the resource circularization route (Guzzo et al. 2019).

Therefore, an organizational culture that is open to broader environmental approaches and committed to experimentation paves the way for the company's circular transformation (Guzzo et al. 2023). Inevitably, in this journey firms face barriers of different nature: human (e.g., the difficulty to understand system dynamics), organizational (a corporate culture with high resistance to change), and technical (e.g., low knowledge to apply system-engineering techniques) (Pesce et al. 2020). Focusing capacity building on overcoming individual and local barriers will enable firms to identify persistent linear trends and, in turn, accelerate the process of closing the material, component and product loops (Guzzo et al. 2023). However, there are still many organizations (in particular MSMEs) that do not manage their environmental impacts within a systems approach (Pesce et al. 2020), as they do not have the knowledge and skills to adopt it on their path towards circularity (Hillary 2000).

1.4 Environmental transformation initiatives promoted from a single dimension

Adopting an LC approach, necessary for the transition to a CE, implies making systemic changes and innovations (Vence & Pereira 2019). Therefore, eco-innovations should be approached from a holistic view, motivating the integration of social, economic and environmental aspects in all its stages (Castiglione et al 2021). Green actions implemented in firms, in response to the challenges of climate change, are then expected to produce different benefits (Nulkar 2017). According to Rennings (2000), eco-innovations should produce at least three types of changes towards sustainable development: technological, social and institutional. In other words, to create a circular system on a farm, it is important that the green innovation, in addition to producing positive environmental impacts, promotes an efficient use of time and generates profits. For Maja & Elliot to be profitable while also being environmentally and socially sustainable is one of the greatest challenges (Maja & Elliot 2022).

In this context, cost savings derived from the adoption of green business practices may be the reason and not the effect of these practices (Purwandani & Michaud 2021). For MSMEs, an economic gain is a key requirement for the survival; therefore, they will only invest in eco-innovations if they have evidence of their economic returns (Oliveira Neto et al. 2017), that is, unless they see value in them (Javed et al. 2022). Only when green innovations produce multiple benefits, including improvements in the work environment, increased prestige, and the creation of new business models, do owner-managers embark on ambitious environmental projects (Torres-Guevara et al. 2021). Multiple studies show that MSMEs to invest in green technologies do not consider environmental sustainability as a main value but as a way of achieving social and economic value (Saunila et al. 2019); however, there is still a marked tendency to promote EOP practices in MSMEs, which seek quick environmental impacts. This single-benefit approach discourages small entrepreneurs, becoming a barrier to their transition to circularity.

1.5 External green knowledge providers tend to ignore the barriers faced by MSMEs to integrate into circularity

Several barriers hinder the integration of CE principles into MSMEs' business models (Rizos et al. 2016), with economic and financial barriers being some of the biggest obstacles (Purwandani & Michaud 2021; EC 2022; Alayón et al. 2022). Those practices that suppose an extra investment in new technology and a slow return on investment, or simply inflict costs, are not widely accepted among MSMEs (Nulkar 2017; Hwang et al. 2018; de Oliveira Neto et al. 2022). On the contrary, those that require low investment, and in turn generate savings, are easily adopted by MSMEs. Understanding that financial issues may act as an enabler or a barrier toward the adoption of green business practices (Purwandani & Michaud 2021), green knowledge external providers cannot ignore their influence on the environmental transition of MSMEs.

On the other hand, low access to technical knowledge is one of the major bottlenecks that small firms encounter in their attempt to improve their environmental performance (Javed et al. 2022). Establishing bonds of trust and intense collaborations with various stakeholders is key for MSMEs to access specialized knowledge and green technologies that they do not have in-house (Alayón et al. 2022; González-Moreno et al. 2019). However, external suppliers, seeing themselves as temporary actors in the environmental transfer process, are often not interested in creating such bonds.

Furthermore, recognizing the multipurpose nature of green innovations (saving resources, reducing pollution, protecting health, etc.), external knowledge from diverse sources is required (González-Moreno et al. 2019). This implies establishing different kinds of relationships and interactions with a wide range of actors (Trischler 2020). Unfortunately, many green growth initiatives in developing countries still tend to include in their teams mainly professionals (particularly in the environmental area), limiting the interaction between individuals with diverse disciplines and empirical knowledge.

Equally important is to create active learning spaces where MSMEs develop skills while solving environmental shortcomings in their own business. Cleaner production is a process of learning by doing in real time, where experimentation facilitates the understanding of new technologies (Dieleman 2007). Although learning-by-doing is considered the leading factor in decreasing the intensity of industrial pollution (Wang et al. 2018), many government and research initiatives are still limited to conducting environmental diagnoses in MSMEs in developing countries, without achieving radical changes in their environmental behavior.

Finally, the literature review identified both gaps and enablers for the transfer of green and circular growth strategies to MSMEs, which became the basis for designing the case study applied in this research.

2 Methodology

The selected methodology was simple case study, which, due to its qualitative approach (Arzaluz 2005; Monroy 2009), allows understanding the differences and similarities between the theoretical information and that collected in the field, using diverse data sources (Baxter & Jack 2008; Yin 2009; Lamboglia et al. 2018). By following a system thinking approach, a farm-system with polluting and inefficient coffee and pig production activities, acted as the unit of analysis. The complexity of the case also required an interdisciplinary and integrated approach for addressing the problem, starting from identifying its causes to proposing alternatives for its transformation towards circularity (Pacheco et al. 2001). The interdisciplinary approach helps to solve problems that exceed the capacity of an isolated discipline, creating new knowledge (Fernández et al. 2007), while the integrated approach seeks to relate the socio-environmental and productive dimensions. The systemic approach, meanwhile, encourages the adoption of green technologies focused on the interaction between subsystems, in order to optimize the environmental performance of the farm-system as a whole.

The case study followed three phases generally suggested in CP processes (Fajardo-Fonseca 2017; Quishpe-López et al. 2020). Phase I. Diagnosis and action plan: It is an initial characterization of the farm in multiple dimensions, in order to propose scenarios to select the most economically and environmentally viable alternative (Garcia Lorenzo, & Slocombe, 2019; Carvajal-Padilla et al. 2021), which responds to the needs of a rural family. Phase II. Action plan implementation: it is the adoption of good environmental and agricultural practices and technology upgrade measures, recommended in the action plan stage, for improving production processes (Alpízar et al. 2011; Bernal et al. 2016; Fajardo-Fonseca 2017). Phase III. Monitoring and follow-up: final evaluation of the environmental strategies implemented, knowledge appropriation, and green skills acquired by small entrepreneurs to sustain their rural unit in an eco-efficient manner. Such appropriation allows the rationalization of resources and the optimization of processes, raising the competitiveness of the farm and product quality in all dimensions of sustainability (Garcia Lorenzo & Slocombe 2019). CP implies an eco-efficiency approach that results in environmental and economic benefits (Bernal et al. 2016); however, the success of its application lies in the sustainability of implementations, being necessary to develop specific green skills in key actors to achieve such sustainability (Varela 2003; Bernal et al. 2016).

For data collection, Phase l included field visits and semi-structured interviews whose flexibility allowed the questions to be adapted to the context of the study (Troncoso and Amaya 2016). In addition, theoretical-practical workshops were used to develop skills from the application of theoretical concepts in the resolution of real problems; ensuring a collaborative environment, avoiding individual perception without experience (García et al. 2013; Fernández Fastuca & Guevara 2017). Thus, different tools supported the development of a comprehensive diagnosis and a participatory action plan for the green transition of the farm. In phases II and III, priority was given to participatory and systematic direct observation supported by forms that allowed the collection of accurate data (Marroun & Young 2018); the latter, necessary to evaluate the proper functioning of the implementations according to pre-established observation aspects. All phases adopted the approaches of Learning by Doing (supported by on-site measurement) and Learning-by-Interacting (supported by multi-stakeholder dialog); the first, conceived as an active learning strategy, which allows making mistakes, inquiring and exchanging ideas, providing feedback and systematizing the transformation process towards a green industry (Gamboa & García 2012; Fernández Fastuca & Guevara 2017). The second, as a method that enables innovation through interaction among participants, recognizes the creative potential of all parties, and requires a high degree of trust to integrate it (Retondaro 2015; Rodriguez et al. 2020). The combination of tools and approaches enables the transfer of knowledge to identify, operate and sustain new clean technologies, leading to significant advances in the adoption of CE.

The study followed a mixed methods triangulation design to collect and analyse qualitative and quantitative data, in order to increase the validity and consistency of the findings (Benavides and Gómez-Restrepo 2005); the data collected separately were compared, complemented and validated with each other (Creswell & Plano 2006).

3 Case study development

The case was developed in a small rural farm of 1.8 hectares, where different actions were introduced to address those barriers that prevented its progress to circularity. The farm-system is solely operated by the owners (husband and wife), and consists of three subsystems: (I) Rural family house, (II) Coffee production process where the ripe beans are transformed into dry coffee through a conventional wet milling process (with the use of water), and (III) Small-scale pig raising activity. Priority was given to environmental technological reconversion in the last two subsystems, since these are where the greatest water and soil contamination problems are concentrated.

The farm´s transition to circularity was carried out in three phases, beginning with an environmental diagnosis and action plan, followed by the implementation of good practices and clean technologies, and ending with monitoring. Experts in sustainable production constantly interacted with the farm owners, thus combining empirical and technical knowledge from the beginning to the end of the environmental transformation, following the Learning-by-Doing and Learning-by-Interacting approaches. Results for each phase are described below.

3.1 Phase I: comprehensive participatory diagnosis

A baseline of the farm was elaborated in relation to the socio-environmental and productive dimensions. It was focused on the stages of the productive activities with the greatest negative impacts on the environment. In the case of conventional wet milling activity, the most polluting stages are: pulping, fermentation and washing.

3.1.1 Coffee processing diagnosis

3.1.1.1 Pulping of cherry coffee fruit

An obsolete pulping machine with high water consumption was used to remove the pulp from the fruit; this increased the volume of wastewater. On average, 0.35 kg of pulp (waste) is generated for each kg of cherry coffee. Pulp was composted without technical basis, being reflected in aspects such as: mixing with other unsuitable waste, lack of good practices like aeration and control of drying times, and direct discharge of pulp decomposition leachates to the ground. On the other hand, the manual coffee shaking process caused ergonomic problems to the female owner of the farm, associated with repetitive movements of the rotator cuff.

3.1.1.2 Coffee fruit fermentation and washing

The mucilage that covers the coffee bean was fermented naturally and then removed by washing with water in a rudimentary fermentation and washing tank. The lack of technical knowledge and deep-rooted beliefs about the amount of water required for this process led to high water use during each rinse (37L/kg coffee) and an excessive number of rinses (7). This situation significantly increased the volume of honey water with a high organic pollutant load (0.2 kg BDO5 /kg coffee and 0.1 kg TSS/kg coffee), which was discharged into the environment without any treatment. Honey water refers to the wastewater resulting from the contact of coffee mucilage with washing water.

On the other side, the female farmer suffered from ergonomic and health problems associated with posture, physical effort and inadequate tools during the manual washing process. This increased the operating time during this stage, affecting the productivity of the farm and the quality of the coffee.

“...When I wash the coffee I have to shake it a lot and that makes my shoulder hurt horribly, and I already suffer from rotator cuff, it really is exhausting, but I have to do it”.

Female farm owner.

3.1.2 Pig farming diagnosis

The farm raises pigs on a small scale (10 on average) in an 18m2 area. The absence of water-saving nozzles resulted in water wastage and discharge of wastewater directly to the ground. Approximately 7.5 kg/day of swine feces (solid manure) were collected dry and improperly disposed in the coffee pulp composting system, affecting its physical, chemical and microbiological characteristics, since it was not suitable for the treatment of animal waste.

The diagnosis showed knowledge gaps in terms of poor environmental practices, deep-rooted habits that encouraged the excessive use of natural resources, and critical aspects in the production stages that had a negative impact on water, soil and air resources (Fig. 1). From there, three possible scenarios for the environmental transformation of the farm were proposed, considering different types of clean technologies and best practices, investment costs, and scope of the implementations. These scenarios were jointly constructed with the landowners, taking into account their capacity to operate and maintain the technological alternatives.

Fig. 1
figure 1

Critical points of agro-industrial activities

Full size image

This exercise allowed the exchange of ideas among the different actors, increasing knowledge and motivation towards circularity. In particular, it allowed the owners to learn and understand the operation and purpose of clean technologies and to question themselves about beliefs and habits that accentuated inadequate environmental and productive behaviors.

“I didn’t know that putting the coffee pulp with the feces of the pigs was bad; I thought that it was necessary to use a lot of water to separate the good coffee from the regular coffee and so that the pulping machine would not damage it; I thought that if the coffee was not washed several times, it would not be clean”.

Female farm owner.

By consensus, scenario “A” was selected as the most suitable action plan to be adopted, according to the context and characteristics of the farm (Table 1).

Table 1 Possible scenarios for the environmental transformation of the farm
Full size table

Scenario A was chosen as the most suitable for orienting the farm-system towards circularity, as it allows a greater reduction in the consumption of resources per critical stage, and a greater use of waste between subsystems. This scenario fosters the transition of the farm towards a resource recovery system, based on the reconstruction of desired properties (Cavallo & Klir 1979); in particular, the strengthening of the interaction between subsystems in order to increase the environmental efficiency of the whole system, and reduce the negative impacts on the wider system (Guzzo et al. 2023). Such reconstructability adopts a combination of environmental practices and technologies, which help to redesign processes to increase the value and extend the life of waste within the system (Mikulcic et al. 2022).

3.2 Phase II: implementation of the action plan—Scenario A

3.2.1 Pulping of cherry coffee fruit

Technological conversion. According to the maximum production capacity of the farm (300 @ Coffee/year), a 2 ¾" semi-integrated equipment was installed, which eliminated water consumption, increased process efficiency and improved coffee bean quality. It is an electric equipment, easy to operate and maintain, with a vibrating sieve and a pulping capacity of 300 kg of coffee per hour. Good practices were adopted for its dry operation. New composting system. A manually operated, double compartment composting system for pulp decomposition was designed and built; both compartments have a 6m3 useful volume, covered by a roof to avoid the entry of rainwater into the system and the consequent increase of leachates. Compartment 1 is the wet compost bin, with a concrete floor covered with bamboo mat, and graded to the leachate storage tank to permit drainage of leachate. Compartment 2 is the dry compost bin where the bio-compost used in crops is finally obtained.

3.2.2 Coffee bean fermentation and washing

Different actions were carried out to Reduce water consumption: (i) Training and adoption of practices for the efficient use of water resources, adjusting the washing process to a maximum of four rinses, using a reduced but sufficient amount of water to guarantee a good wash. The latter, filling the tank until a 6 cm thick sheet of water covers the coffee, without the need to fill it completely. This guarantees a maximum consumption of 5L water/kg coffee at this stage; (ii) Installation of a 700L double compartment fermentation tank, sloping towards the drainage area, with smooth walls and rounded corners, characteristics that prevent the coffee beans from becoming encrusted and requiring a greater amount of water for cleaning.; (iii) Use of a PVC paddle to eliminate the need to wash with the hands; this lightweight tool, with holes that reduce the contact area, makes it possible to stir the coffee with less effort and to reach all areas of the washing tank more easily. By achieving a homogeneous mixture, the mucilage is easily removed, and less rinsing is required. At the same time, the smooth material of the paddle prevents the accumulation of residues from previous washes on its surface that can interfere with bean quality. Waste management focused on the construction of a distribution box to convey all the usable honey water (rinses 1 and 2) to a mixed biodigester. In addition, because of their low organic load, rinses 3 and 4 and wastewater from the cleaning of the semi-integrated equipment and the fermentation tank are discharged to the soil.

3.2.3 Pig farming

By installing nozzles on the hoses, water consumption in pig housing cleaning was reduced. For waste and by-product management, a piston-type mixed biodigester with continuous flow was installed for the treatment of coffee process effluents (rinses 1 and 2) and pig farming waste (solid and liquid swine manure). It is a flexible structure with a 13 m3 biodigestion bag and a 6 m3 reservoir for biogas storage. A shade cloth perimeter enclosure and a zinc tile roof maintain a warm temperature that favors the survival of microorganisms.

To accelerate the start-up of the biodigester, a microbial inoculum composed of microorganisms that help in the transformation of waste was added.; this inoculum contains sugarcane syrup, yogurt or fermented milk, yeast, and fresh beef or swine manure. All are low-cost ingredients that are easy to acquire locally. Biofertilizer is collected in a storage tank at the outlet of the system to feed the soil and crops in the farm. A pipeline network carries the biogas to the house for food preparation. Proper management and disposal of swine waste and proper pig housing cleaning helped control offensive odors.

As a crosscutting measure, roofing elements were adapted and drainage ditches were built for storm water management; this prevented rainwater from entering the fermentation tank, compost bins, equipment, and other components of the process. In addition, it prevented the dilution of coffee processing effluents with rainwater and the consequent increase in wastewater, the over hydration and washing of compost nutrients, and the equipment deterioration.

Figure 2 schematizes the environmental implementations carried out, stressing the waste reuse and recovery routes, demonstrating circularity within the system. Figure 3a shows the semi-integrated equipment, fermentation and washing tank, distribution box, and composting system implemented. Figure 3b shows the anaerobic biodigestion system implemented on the farm.

Fig. 2
figure 2

Distribution of environmental implementations following a resource recovery system approach

Full size image
Fig. 3
figure 3

a Cleaner coffee process, b Mixed biodigester with flexible bag

Full size image

3.3 Phase III: Monitoring and follow-up

Final socio-environmental and productive impacts of the implementations carried out in the farm were determined. A cleaner pulping stage of the cherry coffee fruit led to the elimination of water consumption and the recovery of 100% of pulp waste through an enhanced composting process. This improvement made it possible to increase the useful volume and height of the compartments, separate the wet pulp from the dry pulp, eliminate the bad practice of mixing pulp with dry pig manure, facilitate the manual aeration process, reduce the presence of vectors and offensive odors, and prevent the entry of rainwater; in addition, recirculate 100% of leachates back to the wet compost bin, thus recycling a nutrient-rich waste that should not be discharged into the environment due to its high acidity. The Guadua Angustifolia (also known as the Colombian timber bamboo), used for the construction of the compost bins, improved their aesthetic, functional and safety conditions.

In particular, with the installation of the semi-integrated equipment with vibrating sieve included, the pulping time was reduced by 75%; thus, in a harvest of 136 kg of cherry coffee, the pulping and sieving process was reduced from 40 to 10 min. Switching from manual to automatic shaking to classify the grain led to the elimination of repetitive manual movements for long periods of time. In addition, this new equipment protects the structure of the grain by avoiding splitting or fractioning during the pulping process, ensuring its quality.

“The project was very beneficial for me as the machine does the shaking and that's a plus because of my arm pain”.

Female farm owner.

A better understanding of the general characteristics and benefits of the semi-integrated equipment and composting system was achieved through multi-stakeholder dialogs between the farm owners and female experts in rural environmental management. These served as learning, reflection and joint decision-making spaces.

“We felt very good with the support of the female engineers because they listened to our opinions and we learned useful and easy things to apply in our farm. This encourages us to keep going and continue working.”.

Farm owners.

The active participation of the owners in the theoretical and practical workshops, and during the implementation of clean technologies, helped the transfer of concepts through their application in their rural unit. Systematic and participatory observation visits showed that the owners were able to recognize and apply the concepts of CP, SCP and CE, and to identify the characteristics of a more sustainable coffee process. They also showed skills to solve potential failures and properly operate the new semi-integrated equipment, which preserves the same operating principle as the old one, resulting in a technology that is both simple to adopt and innovative. Besides, they proved their ability to: recognize organoleptic and visual characteristics to determine whether or not the composting system is working properly, correctly apply the biofertilizer to the crops, and control critical variables such as temperature, pH and humidity.

“I didn't know I could reuse everything, now I manage the pulp first in the wet pit and then in the dry pit, using it to fertilize the coffee plants and the vegetable garden, being self-sustainable”.

Female farm owner.

On the other hand, the implementations in the fermentation and coffee bean washing stage (new tank and good practices) reduced water consumption by 80%, going from consuming 40L water/kg coffee to 5L water/kg coffee. Good washing practices, a reduced number of rinses, the use of the coffee washing paddle, and the elimination of manual emptying of the old washing and fermentation tank, reduced the operation time by 53%, decreasing from 75 to 40 min the time to wash 136 kg of coffee. Particularly, since the coffee washing paddle is lighter and easier to handle, it improved the operator's body position during the washing process.

In relation to pig breeding and raising, training sessions on the rational use of resources and installing nozzles on hoses, kept water consumption at 179L/day, and reduced pig housing cleaning time by 66%, dropping from 15 to 5 min. The mixed biodigester made it possible to treat and reuse 100% of the usable waste, treating approximately 7.5 kg of dry pig manure/day, and 529 L/day of combined effluent (350 L/day of coffee wastewater + 179 L/day of liquid pig manure). Three months after inoculation, 1m3 of biofertilizer/month and enough biogas for daily food preparation were obtained as by-products. Appropriate hydraulic retention times allowed the anaerobic microorganisms, necessary to transform the organic matter in the combined effluent, to work efficiently in the different phases of the biodigestion process (hydrolysis, acidogenesis and methanogenesis).

As for biogas production, it showed variations depending on aspects such as: (i) Weather conditions. Sunnier weather periods favored biogas production, resulting in a longer life of this cooking fuel; (ii) Number of pigs. A lower number of pigs than stipulated in the design reduced the generation of waste (swine manure) and therefore the production of biogas; (iii) Coffee production volume. In the two coffee harvests per year, the generation of usable wastewater is higher, increasing the production of biogas.

Actively participating in the biogester installation and in different trainings for its correct operation and maintenance, coffee farmers developed skills to: recognize when to leak test and add inoculum to increase system efficiency, select inputs for a chemical-free cleaning process, control critical variables such as pH, and apply by-products such as biofertilizer to crops. Furthermore, to identify visual and organoleptic characteristics of the biofertilizer that reflect failures in the system, including the presence of large pieces of manure without decomposition and a strong unpleasant odor. In turn, to recognize the blue appearance of the flame as an indicator of good biogas quality.

“This natural fertilizer is very good because we no longer buy so many chemicals and we also use it to fertilize fruit trees."

Farm owners.

As for economic benefits, the implementations generated annual savings of USD$950.00, approximately 2.8 legal minimum monthly salaries in Colombia, as shown in Table 1 (Online Appendix 1). Considering that the total cost of implementations was USD$3,380.00, the expected payback time is 3.75 years.

Finally, the farm achieved the transition from a linear model to circularity by modifying deep-rooted behaviors and beliefs at the economic, productive, social and environmental levels.

“We want our farm to be self-sustainable; we are treating wastewater and fertilizing the garden with organic fertilizer from the farm itself. We want to install solar panels in the future”.

Farm owners.

4 Conclusions and discussion of results

This applied study adds further evidence to the existing literature on Green and circular growth at MSMEs within the context of a developing country. Concerned on the transition of MSMEs towards circular production models, it contributes to the identification and validation of gaps, barriers, and motivators that limit such transition. A case study developed in a coffee and pork farm in Colombia (South America) demonstrated that using simple eco-innovations, and from a preventive, more holistic approach, the MSMEs' path towards circularity can be accelerated.

On the one hand, low-cost or moderate-cost and easy-to-apply clean technologies made it possible to minimize the consumption of water resources and optimize the use of waste derived from coffee and pork production; at the same time, production times were reduced, and the health of the female rural entrepreneur was positively impacted by eliminating the need to perform repetitive movements with her arm for the manual operation of the old coffee pulper. This proved that with lower-cost and easy implementable strategies it is possible to minimize negative impacts to the environment, and to obtain parallel gains in terms of time, health and savings (Oliveira Neto et al. 2017; Purwandani & Michaud 2021); thus, simple eco-efficient strategies that allow producing more with less effort and resources, with tangible benefits and a short return on investment, are considered the most convenient way for MSMEs in developing countries to move towards sustainability (Vásquez et al. 2019; Torres-Guevara et al. 2021).

The findings of the study found a strong connection with the Lean Manufacturing strategy, which aims to safeguard the companies’ resources by eliminating wastes. According to its philosophy, those inputs, processes and outputs that demand high consumption of resources and time in the operations are considered non-value activities, therefore they should be eliminated (Kafuku 2019), in order to increase productivity (Serrano & Costa 2018). To do so, companies can simultaneously adopt and combine Lean and Green strategies, increasing process efficiency and product quality, while reducing costs and risks. Waste Reduction Techniques is one of the main points where Lean and Green Strategies converge, to help companies become economically and environmentally more responsible (Fercoq et al. 2016).

On the other hand, MSMEs often lack personnel trained in technical and environmental fields. It is then favorable that the new technologies do not require highly specialized knowledge for their operation and share operating principles with the previous ones. It is argued that prior knowledge or memories may facilitate new learning and creativity through associations with pre-existing concepts, increasing the ability to recognize the value of new information, and to assimilate it and apply it (Cohen & Levinthal 1990; Volverda, et al. 2010; Schmidt 2010). In our case, the new technologies share the same operating bases as the old ones, but at the same time include elements or operational methods that promote environmental care on a larger scale; thus, the new equipment for cherry coffee pulping with a mechanical sieve, which operates in a similar way to the previous one, but without the need for water and with less physical effort, was easily integrated into the coffee production process. It is well known that the existing technological capabilities of the firm (which include knowledge embedded in people and learning behavior) (Volverda et al. 2010), are needed in the process of adoption of new technological knowledge (Bell & Albu 1999).

Regarding the systems thinking approach, it helped to rethink, redesign, and implement a resource recovery system, focusing on the efficient use of resources and the closing of material, product and process cycles (Iacovidou Eleni et al. 2021). Here, the right combination of good environmental practices and technological eco-innovations, which promote the reuse of waste between subsystems, drove the implementation of a resource circularization pathway, resulting in a deep transformational change in the system-farm (Guzzo et al. 2019; Eshun et al. 2012). In our case, solid and liquid waste (pulp and effluents), derived from the new ecological pulping and washing system, feed respectively an improved composting system and a new mixed biodigester where swine manure waste is also treated. From the composting process, biofertilizer is obtained for coffee crops. From the biodigester, biogas is recovered for cooking in the rural house. In this way, the useful life of waste is extended and the need for raw materials is reduced (Schröder et al. 2020). This translates into savings by reducing the dependence on and purchase of chemical fertilizers and propane gas, the latter of which entails high transportation costs to remote areas. It is evident that the emergence of green innovations requires a variety of knowledge coming from different sources; however, this variety should favor the implementation of related and complementary technologies (Demirel et al. 2019; Grazzi et al. 2019) leading to positive drastic changes in the existing production process or at different levels of the firm (del Río González 2005).

It is clear then that the adoption of preventive, simple, economic and complementary green technologies contributes to break technological, financial, organizational and knowledge barriers normally observed in MSMEs; these barriers coexist and affect each other (Rizos et al. 2016; Purwandani & Michaud 2021; Alayón et al. 2022). Therefore, for the circular transition in MSMEs, several barriers must be broken at the same time, avoiding accentuating or prolonging them over time; that is, promoting a technological reconversion with benefits beyond the environmental ones (Castiglione et al 2021; Nulkar 2017; Rennings 2000; de Oliveira Neto et al. 2022).

On the other hand, the environmental implementations derived from the learning-by-doing approach and their multiple benefits, removed false beliefs in the coffee farmers that hindered the environmental transformation of their rural unit. In turn, constant interaction with external green knowledge providers, through multi-stakeholder dialogs and participatory observation, encouraged technical and empirical knowledge sharing and continuous feedback, acting as spaces for collaborative learning. Network relationships have been recognized as facilitators of organizational learning around sustainable production (Vickers & Cordey-Hayes 1999; Bass 2007). Social learning theory posits that learning is a cognitive process that takes place in a social context where individuals learn through observation of or interaction with other individuals (Heyes 1994; Bandura 1963). Under the CE framework, actors whose competencies and skills help to narrow, slow, and close the resource loops in order to implement circular business models (Re & Magnani 2022).

In our case, a constant interaction was maintained between coffee farmers and green knowledge providers, from the environmental diagnosis phase to the implementation and monitoring of green practices and technologies. This interaction further developed the firm's capabilities to acquire, apply and assimilate new external environmental knowledge. In other words, it helped to develop the firm's absorptive capacity (AC), needed for a successful green innovations transfer (Vickers & Cordey- Hayes 1999; Aboelmaged & Hashem 2019), the adoption of CE practices, and the redesign of a business model in a more circular way (Marrucci et al. 2022). This collaboration was essential to consolidate concepts, develop skills for the operation and maintenance of new environmental technologies, and finally adopt a new production model on the farm based on resource conservation, reuse and waste utilization.

The final results of the study have implications for consultants, regulators, universities, or other organizations in charge of environmental projects in MSMEs in developing countries; they, as external knowledge providers, should make a careful selection of green innovations from a systemic approach, assuring to promote those that fit the MSME context, without exceeding or underestimating their capacity. At the same time, environmental projects with multiple benefits should be promoted to motivate MSMEs towards circularity, avoiding stagnation in the elaboration of environmental diagnoses without producing positive changes in the firms. Interdisciplinary project teams should be favored, which are willing to create green innovations together with MSMEs' staff, valuing their empirical knowledge. In particular, regulators should be trained and support MSMEs in preventive and complementary green technologies, seeking to go beyond an end-of-pipe approach imposed by an obsolete command and control role. Finally, the results of this research are not characterized by generalizability.