Exploring pathways for circular supply chain in SMEs: the role of trust, information sharing and organizational adaptability

Circular supply chains (CSS) have emerged as a promising solution, offering a way to reduce waste, optimize resource use, and promote sustainability by reusing, refurbishing, and recycling materials throughout the production cycle (Lin and Chu, 2024; Münch et al., 2022). Unlike traditional linear supply chains that follow a “take-make-dispose” model (Tang et al., 2021), CSS focuses on closing the loop, ensuring that products and materials are continuously repurposed rather than discarded. This approach not only minimizes environmental impact but also enhances economic efficiency by reducing dependence on raw materials and lowering production costs. Furthermore, circular supply chains encourage innovation in product design, logistics, and business models, fostering a more resilient and responsible economy. By embracing circularity, small and medium enterprises (SMEs) can not only contribute to environmental preservation but also enhance their long-term competitiveness (Faisal, 2023). However, achieving effective circular supply chains in SMEs requires overcoming significant barriers, such as limited access to information and trust among stakeholders (Zhang et al., 2025). This study fills an important literature gap by investigating how firms can capitalize on effective communication, information-sharing practices, and quick collaboration to help SMEs transition to circular models.

Existing studies argue that for firms to effectively adopt circular economy (CE) practices, they must develop the necessary capabilities for inter-organizational collaboration, creating synergies across the entire supply chain (Mishra et al., 2019; Schöggl et al., 2024). This collaboration is particularly crucial for small and medium enterprises which often face resource constraints and need external partnerships to drive sustainable practices. To successfully implement CSCs, firms must engage in mutual efforts that involve trust, shared goals, open communication, and the continuous exchange of information and knowledge (Koh et al., 2017; Nasir et al., 2017). Studies highlight the importance of these elements along with adaptability and collaborative learning in fostering strong relationships and improving CSC performance (Masi et al., 2017; Prieto-Sandoval et al., 2018). Moreover, the interconnectedness of these factors plays a critical role in advancing CSCs and supporting sustainable development. However, while existing research emphasizes these variables, it remains unclear which specific combinations of these factors are most crucial for SMEs’ successful transition to circular models, warranting further investigation to refine strategies for achieving high levels of collaboration and sustainability in SMEs. The lack of clarity regarding these combinations arises because prior studies have predominantly examined each variable trust, adaptability, or information sharing in isolation rather than as interdependent mechanisms. This piecemeal approach obscures the synergistic effects that occur when these factors interact. Empirical evidence from recent studies (e.g. Alcalde-Calonge et al., 2024; Schöggl et al., 2024) suggests that the success of CSC initiatives depends not on single factors but on the complementarities among them, such as how adaptability amplifies the benefits of trust or how risk perception moderates the impact of information sharing. Clarifying these interrelationships is essential for theory development and managerial practice because it enables SMEs to identify which specific configurations yield high circular performance. Without this understanding, SMEs risk misallocating scarce resources toward isolated improvements that fail to deliver systemic sustainability outcomes, thereby perpetuating fragmented and inefficient circular transitions.

In the context of the circular economy, these collaborative elements: trust, shared goals, open communication, and continuous knowledge exchange are foundational to CSCs. Trust reduces uncertainty and transaction costs, allowing firms to share sensitive sustainability data essential for closing material loops. Shared goals align partners toward achieving collective environmental and economic objectives, ensuring that circular initiatives such as product recovery or remanufacturing are mutually beneficial. Open communication strengthens inter-organizational transparency and facilitates the coordination of reverse logistics and resource recovery processes, which are central to circularity. Likewise, continuous information and knowledge exchange foster innovation and adaptive learning, enabling SMEs to refine processes, adopt cleaner technologies, and co-create solutions that sustain material circulation. Therefore, the interplay of these factors not only supports effective CSC implementation but also operationalizes the core principles of the circular economy through collaborative and knowledge-driven practices.

This study aims to fill this void by employing the fuzzy set qualitative comparative analysis (fsQCA) to examine the pertinent configurations of factors derived from Organizational Information Processing Theory (OIPT) and Swift Trust Theory (STT), such as trust, information sharing, adaptability, risk perception, collaborative learning, and proper communication channels, that can be leveraged to advance high levels of CSC in SMEs. STT provides insights into how temporary or rapidly formed partnerships build trust and collaboration under the uncertainty conditions that typify SMEs' short-term, dynamic relationships in CSC networks. OIPT complements this by explaining how firms process, share, and utilize information to reduce uncertainty and enhance decision-making efficiency across supply chains. Integrating these two theories provides a new lens for understanding how swift relational trust interacts with information-processing capabilities to drive sustainable collaboration, thereby offering a theoretical bridge between behavioral and structural perspectives on CSC development. Precisely, the study achieves its overall objective by answering the research questions below:

By integrating OIPT into the CSC framework, SMEs can enhance their information-sharing mechanisms, improve communication with partners, and ultimately reduce inefficiencies and waste. Alternatively, swift trust can enable effective collaborations, reducing the friction that might arise from a lack of long-term relationships. For SMEs, this is particularly relevant as they often need to engage in quick, collaborative efforts with new partners, including suppliers, customers, and recycling firms, to implement circular strategies.

We contribute to the literature as follows: First, this study expands knowledge on the essential role of OIPT and STT in fostering strong collaborations, which are considered crucial for advancing CSC.The integration of Swift Trust Theory (STT) and Organizational Information Processing Theory (OIPT) provides a complementary lens for understanding how SMEs overcome uncertainty and collaboration challenges in circular supply chains. While OIPT focuses on how firms process and utilize information to reduce uncertainty and improve decision-making efficiency, STT explains how temporary and resource-constrained organizations, such as SMEs, can quickly build trust with new partners to enable collaboration in circular initiatives. Combining these theories bridges a crucial gap in existing CSC literature, which often treats trust and information exchange as separate phenomena. This integration introduces a new explanatory pathway, showing that swift trust enhances the efficiency of information processing, while effective information systems reinforce trust formation. Together, they provide a multidimensional theoretical foundation for understanding how SMEs achieve rapid, adaptive, and trust-based collaboration in implementing circular supply chains.

Second, by integrating OIPT and STT to investigate the critical factors for advancing CSCs in SMEs, this study overcomes the current deficiency in the integration of theory and practice highlighted in the CSCM literature as posited by researchers (Lahane et al., 2020; Zhang et al., 2021). Third, the study broadens current understanding by employing a fuzzy set qualitative comparative analysis (fs/QCA) to provide empirical evidence of how various configurations of factors can be leveraged to foster effective collaborations for advancing CSC in SMEs. Finally, through a deeper understanding of these variables, this research contributes to the theoretical development of the CSCM domain by offering comprehensive insights into the collective impact of these constructs on CSC success, especially in the context of SMEs in a developing country (Ghana) context.

Following the introduction, section 2 presents the literature review and theoretical background. Section 3 presents the research methodology, while Section 4 presents data analysis and results. Section 5 presents the discussion and implications, conclusion, limitations, and future research directions.

The Swift Trust Theory (STT) explains how trust can rapidly develop in temporary teams or collaborations, particularly under conditions of limited interaction and high uncertainty, making it particularly relevant for CSC practices (Tatham and Kovács, 2010). The theory reveals the important factors that influence how trust develops among different partners with insufficient information and prior engagement to determine each other's trustworthiness (Leung et al., 2022). Studies show that trust is an essential collaboration mechanism for not only developing long-term relationships but also initial relationship-building with a plethora of supply chain actors and organizations across multiple supply chain networks for establishing CSCs (Veleva and Bodkin, 2018). Leveraging the STT, this study posits that organizational trust, shared goals and objectives, and positive risk perceptions are critical variables that can help SMEs develop swift trust and foster collaborations to advance CSC.

First, organizational trust is fundamental for enabling collaboration and risk-sharing among supply chain partners in CSCs. It helps reduce uncertainties and encourages openness in sharing information and resources, which is essential for circular initiatives (Batista et al., 2018; Farooque et al., 2019). A baseline trust can emerge from existing relationships, easing interactions in new collaborative settings (McLaren and Loosemore, 2019). When this is present, supply chain actors are more likely to engage in joint efforts, enhancing knowledge-sharing and innovation within CSCs (Masi et al., 2017).

Second, shared goals and objectives further foster swift trust by aligning the interests of partners, ensuring that all parties work toward common aims. The STT suggests that organizations can build quicker trust with collaborating partners based on shared goals and objectives to lay the foundation for cognitive sensemaking towards a more knowledge-based trust (Yu et al., 2022). Moreover, when stakeholders share objectives, trust builds more quickly as participants feel more confident in the collaborative effort and its outcomes (Leung et al., 2022). For CSCs, the alignment of strategic goals, such as sustainability and resource optimization, is crucial for successful collaboration (Bressanelli et al., 2019; Sudusinghe and Seuring, 2022). Misalignment of goals can hinder trust and collaboration, potentially derailing circular initiatives (Schöggl et al., 2024).

Positive risk perception is another determinant of swift trust, as it influences the willingness of organizations to engage in uncertain collaborations. Research shows that when actors perceive the risks involved in a partnership as manageable or outweighed by potential gains, trust develops more readily (Dubey et al., 2019; Yu et al., 2022). In the context of CSCs, firms with higher risk tolerance are more likely to pursue innovative circular practices despite uncertainties around product design, regulatory changes, or market shifts (Singh and Dwivedi, 2024). Together, these mechanisms: organizational trust, shared goals, and positive risk perceptions create the foundation for effective collaborations in CSCs, enabling sustainable circular practices across supply chains. From the above, we propose that:

The Organizational Information Processing Theory (OIPT) emphasizes how organizations manage uncertainty through the acquisition, processing, and dissemination of information to make informed decisions and enhance performance (Makkonen, 2021; Peng et al., 2023; Srinivasan and Swink, 2015). The theory posits that increasing uncertainty leads to a need for greater information processing capacity, allowing organizations to navigate complex tasks and improve decision-making (Cegielski et al., 2012; Xie et al., 2022). In the context of CSC practices, OIPT provides a framework to understand how effective information management can mitigate uncertainties in the transition to circular economies, fostering collaboration and innovation. Four key mechanisms under OIPT: (1) information sharing, (2) communication channels, (3) adaptability, and (4) collaborative learning are crucial for enhancing CSC practices.

Information sharing is essential for managing the flow of data across supply chain partners, enabling firms to track resources, coordinate reverse logistics, and adopt circular economy practices (Koh et al., 2017; Nasir et al., 2017). Effective information sharing reduces information asymmetry and facilitates the collaboration needed for circularity, ensuring firms stay informed about technological advancements, best practices, and industry trends (Batista et al., 2018; Peng et al., 2023). Moreover, information sharing enables organizations to remain informed about sustainable trends, technological innovations, and industry best practices, thereby advancing circular product designs and production processes that promote zero waste. Communication channels further enhance an organization’s ability to process information by facilitating the exchange of innovation-related insights across supply chains (Makkonen, 2021). Studies show that robust communication systems are vital for joint decision-making in CSC initiatives, reducing uncertainty by enabling rapid data acquisition, analysis, and sharing (Dubey et al., 2020; Xie et al., 2022). Recent technological innovations such as AI, IoT, and blockchain enhance communication, improving the efficiency and effectiveness of information sharing within supply chain networks (Schöggl et al., 2024). Next, adaptability which highlights a firm’s capacity to adjust strategies and processes based on the information received, making it vital for responding to the dynamic nature of CSC (Sehnem et al., 2019). The ability to adapt enables organizations to navigate emerging opportunities and challenges within circular supply chains, helping them develop new collaborative models and respond to market changes. A recent study by Alcalde-Calonge et al. (2024) found that strong adaptive capacity enables SMEs to broaden innovative activities and capitalize on emerging market opportunities, thereby potentially increasing their CE levels. Adaptability is also necessary for product and process design modifications required to ensure the successful implementation of CE practices within supply chain processes (De Angelis et al., 2018; Farooque et al., 2019).

Finally, collaborative learning fosters continuous knowledge sharing among supply chain actors, promoting innovation and resilience in CSC practices (Shi et al., 2023). By leveraging shared knowledge and best practices, firms enhance their capacity to implement circular initiatives effectively, transforming uncertainty into opportunities for innovation (Bag et al., 2022; Sehnem et al., 2019). Collaborative learning can help SMEs acquire and accumulate the information and knowledge to promote innovation and advance CSC practices. Based on the OIPT, this study asserts that collaborative learning can reduce the myriad of CSC-oriented uncertainties, facilitate decision-making processes on innovation, and transform information into new ideas to support the adoption of circular initiatives across supply chains hence we propose that:

Figure 1 shows a graphical representation of how our constructs influence a circular supply chain.

The research follows a sequential methodology incorporating qualitative insights from the literature on Organizational Information Processing Theory and Swift Trust Theory to inform the quantitative data collection and analysis. The combination of these approaches enables a robust exploration of complex interactions between variables influencing CSC levels. To ensure the study's rigor, we employed a two-phase approach. The initial phase entailed a comprehensive literature review and engagement with supply chain management experts to refine the constructs for this context. We then applied a fuzzy set qualitative comparative analysis (fs/QCA), following similar approaches in circular economy studies (Alcalde-Calonge et al., 2024). This methodological choice is appropriate given the study's aim to explore configurations of determinants that lead to either high or low CSC levels, accommodating the possibility of equifinal pathways. To help illustrate and clarify the research process, Figure 2 outlines the steps followed in this study, beginning with problem identification and leading up to data analysis.

The data collection process focused on small and medium-sized enterprises (SMEs) located in the Greater Accra and Ashanti regions of Ghana. These regions were selected as they represent a microcosm of Ghanaian society and host a significant proportion of the country's SMEs, as noted by Asante et al. (2021). Having identified SMEs from Ghana Business Directory and the Association of Ghana Industries (AGI) membership list, as well as through reputable online platforms that list registered firms in these two regions, the convenience sampling technique was applied to this study to select the sample. SMEs were selected based on their accessibility and willingness to participate. This approach facilitates efficient data collection from a diverse set of firms in these two regions. To refine the pool of SMEs, we applied two exclusive criteria: first, firms had to demonstrate active involvement in supply chain operations through documentation showing engagement in activities such as sourcing, production, distribution, or waste recovery processes. Secondly, firms need to have been in operation for at least three years to ensure sufficient experience with supply chain management. Following these criteria, 350 SME owners were identified and initially consulted through a combination of emails and face-to-face interactions to determine their willingness to participate. Of these, 290 firms expressed interest in the study. Subsequently, questionnaires were distributed to these 290 SMEs, with 255 responses received. After data cleaning, which involved eliminating incomplete responses, a final sample of 237 valid responses was obtained. Table 1 summarizes the basic description of the respondents.

The items in our survey were derived from established scales in the literature, ensuring that each variable was measured accurately and consistently. Table 3 outlines the variables, their measurement items, and literature sources for item extraction. Measurement items for each construct were developed by adapting established scales from prior studies on OIPT, STT, and CSC-related research, ensuring alignment with the study's theoretical framework and regional context. The constructs were assessed on a 7-point Likert scale, ranging from 1 (strongly disagree) to 7 (strongly agree). As highlighted above, STT was measured using three variables: organizational trust, shared goals and objectives, and information sharing). These reflect the SMEs' assessment of potential gains and losses, which informs their trust dynamics and willingness to collaborate. OIPT considered the variables: information sharing, communication channels, adaptability, and collaborative learning. Each of these variables has been constructed based on previous studies, as shown in Table 3. All the descriptive statistics of the measures used can be seen in Table 2.

To verify the reliability and validity of our constructs, we performed confirmatory factor analysis (CFA) using maximum likelihood (ML) estimation in AMOS version 29.0. This approach is standard in quantitative research (Twumasi-Ankrah et al., 2024) and is frequently used in fs/QCA studies (Pappas et al., 2016) to ensure that the variables are measured accurately and consistently. The model fit analysis showed strong results, with χ² = 302.971 (df = 247; p = 0.009); χ²/df = 1.227; GFI = 0.915; TLI = 0.981; IFI = 0.984; CFI = 0.984; RMSEA = 0.031; and SRMR = 0.0421 (Hu and Bentler, 1999). These metrics indicate that the model is a good fit for the data, thus confirming its reliability. To evaluate convergent validity and internal consistency, we examined standardized factor loadings and their corresponding t-values. The items used to measure our study variables generally showed strong loadings, with all standardized factor loadings being significant at p < 0.001 and above 0.7 as noted by previous studies (Mahmoud et al., 2024; Tuffour et al., 2023). However, two items (IOT2 and CC1) had loadings just below this threshold but above the minimum threshold of 0.4. However, including these items did not compromise our findings regarding convergent validity, internal consistency, or discriminant validity (See Table 3).

Table 4 below presents the composite reliability (CR), Cronbach's α, and average variance extracted (AVE) for each construct, all of which surpassed the recommended levels of 0.7 for internal consistency and 0.5 for convergent validity (Fornell and Larcker, 1981). To check for discriminant validity, we examined whether the square root of the AVE for each construct was higher than the correlations between that construct and the other constructs. As shown in Table 4, this requirement was fulfilled, providing evidence that all the constructs in our study are distinct from one another. Furthermore, we assessed the heterotrait–monotrait ratio of correlations (HTMT), which yielded values below the threshold of 0.85, further confirming the discriminant validity of the constructs (Henseler et al., 2015).

Fuzzy set qualitative comparative analysis (fsQCA) takes a different approach than traditional research methods that typically focus on straightforward cause-and-effect relationships (Abdallah and Opoku-mensah, 2026; Ampofo et al., 2023; Ling and Peng, 2024). Its core principles of equifinality, complexity, asymmetry, and causal asymmetry are particularly useful for understanding multifaceted causal relationships in CSCs, especially in SMEs, which may have limited resources (Depino-Besada et al., 2024; Woodside, 2014). In our study, we used fsQCA to investigate and provide a richer understanding of how different factor configurations influence CSCs.

To ensure the accuracy and reliability of our data, we calibrated them using latent variable scores, as recommended by Rasoolimanesh et al. (2021). We used a direct calibration method with fs/QCA version 4.1 (Pappas and Woodside, 2021), which sets specific thresholds to determine the membership levels in fuzzy sets. These thresholds were designed for full-set membership, full-set non-membership, and intermediate-set membership, with values set at 0.95, 0.5, and 0.05, following the guidance of (Fiss, 2011). In fs/QCA, cases that are exactly at the 0.5 crossover point are usually excluded, which can make it challenging to interpret intermediate memberships (Ragin, 2009). To address this challenge, we added a small constant of 0.001 to the causal conditions based on the suggestions of Fiss (2011) and Pappas and Woodside (2021).

To begin our analysis using fsQCA, we conducted a necessity analysis to pinpoint the conditions that are crucially linked to achieving the desired outcomes. According to Ragin (2009), a condition is deemed necessary if both its consistency and coverage values are above 0.9. However, for our outcome variable “High CSCs,” we found that no condition met this criterion. The condition with the highest consistency was “Inter-organizational Trust” (IOT), which had a consistency score of 0.788 and a coverage of 0.685. Similarly, when analyzing the outcome variable “Low CSCs,” we again did not find any conditions with consistency or coverage above 0.9. The condition that came closest was “Adaptability” (AD), with a consistency of 0.807 and coverage of 0.575. This means that none of the conditions we examined could independently meet the necessary criteria for influencing either high or low levels of CSCs in SMEs (See Table 5).

In this study, we created truth tables (see Tables 6 and 7) to explore all possible combinations of the conditions investigated. This aligns with the methods proposed by Fiss (2011) and (Ragin, 2012). With a sample of 237 cases, well exceeding the recommended minimum of 150 (Fiss, 2011; Ragin, 2009), we set a frequency threshold of three observations for each configuration to ensure that our findings were adequately represented Fiss (2011). To identify strong causal relationships, we applied a strict consistency threshold of 0.80 (Fiss, 2011). To present a well-rounded view of the results, we include both parsimonious and intermediate solutions, focusing particularly on the intermediate solution. This approach strikes a balance between simplicity and theoretical depth by considering plausible alternatives (Ragin, 2009). Additionally, we categorized the conditions into core and peripheral conditions, highlighting the core conditions that appeared in both the parsimonious and intermediate solutions. This helps clarify and detail the causal patterns related to achieving high and low levels of CSCs in SMEs. Finally, we present the outcomes in tables for better clarity and readability, following the guidance of (Fiss, 2011)

To evaluate these solutions, we first examined the overall consistency and coverage metrics. Consistency measures how well cases fit a specific configuration that leads to the desired outcome. A higher consistency score means that a particular configuration reliably leads to the outcome in most cases, indicating a strong causal relationship (Schneider and Wagemann, 2012). Coverage, on the other hand, examines the proportion of outcomes explained by a specific configuration or a set of configurations (Ragin Charles and Sean, 2016). The findings in Table 8 reveal that both solution consistency and coverage metrics exceed the recommended thresholds of 0.74 (for consistency) and 0.25 (for coverage) (Ragin Charles and Sean, 2016). Specifically, for high CSC levels, the solution achieved a consistency score of 0.877 and a coverage score of 0.313, demonstrating the effectiveness of the model in capturing the conditions that lead to high CSC levels. For low CSC levels, the solution produced a consistency score of 0.758 and coverage of 0.420, indicating a strong fit for cases with low CSC levels. These results confirm that the configurations we studied have significant relevance for predicting both high and low CSC levels in SMEs, reflecting a solid alignment between real-world cases and theoretical models (Ragin Charles and Sean, 2016; Schneider and Wagemann, 2012).

The fsQCA results in Table 8 (Solution 1–3) revealed three distinct configurations, suggesting that varying combinations of core and peripheral conditions provide multiple pathways through which SMEs can advance their CSC implementation. Across all three configurations, ‘adaptability’ and ‘communication channels' emerged as fundamental core conditions, suggesting these are essential prerequisites for achieving high CSC performance. This finding aligns with recent studies highlighting the critical role of adaptability (Alcalde-Calonge et al., 2024; Stumpf et al., 2023), and communication channels (Dey et al., 2020) in enabling SMEs to respond effectively to circular economy transitions.

Interestingly, each configuration presents unique combinations of conditions. Configuration 1 demonstrates that high CSC levels can be achieved when adaptability and communication channels are combined with collaborative learning as core conditions, while other factors remain peripheral or absent. This finding also aligns with literature emphasizing the importance of learning capabilities in CE transitions (Stumpf et al., 2023). Configuration 2 introduces an important variation by incorporating risk perception as a core condition alongside adaptability and communication channels, emphasizing that SMEs' ability to understand and manage risks is essential for CSC success. This aligns with previous research by Masi et al. (2017), which highlights the role of risk management in CSC transitions. Interestingly, this configuration achieves success without collaborative learning or information sharing, indicating that strong risk management capabilities may offset these factors in specific contexts. Configuration 3 reveals that achieving high CSC levels in SMEs requires a synergistic combination of adaptability, communication channels, inter-organizational trust, risk perception, and shared goals. The presence of inter-organizational trust alongside shared goals supports the emphasis of Prieto-Sandoval et al. (2018) on collective vision in circular initiatives, while the inclusion of risk perception aligns with the findings of Masi et al. (2017) on risk-sharing mechanisms in CSC success. The configuration also reinforces the theoretical framework's validity, as it shows how OIPT elements (communication channels and shared goals) work in concert with STT components (trust and risk perception) to enable SMEs to overcome their resource constraints and establish effective circular practices.

Notably, across all three configurations, information sharing appears as an absent condition, suggesting that, while necessary, it may not be a critical determinant for achieving high CSC levels in SMEs. This finding contrasts with some existing literature that emphasized information sharing as crucial for CSC success (Koh et al., 2017; Nasir et al., 2017), indicating that the relationship between information sharing and CSC performance may be more complex than previously thought.

From Table 8, the fsQCA analysis has revealed five distinct configurations leading to low CSC performance, supporting the principle of causal asymmetry that Fiss (2011) asserted that the pathways to low outcomes are not simply mirrored opposites of those leading to high CE adoption.

Primarily, Configuration 4 validates how, despite the presence of adaptability and information-sharing capabilities, SMEs encounter significant barriers to CSC implementation. The absence of collaborative learning in this configuration particularly resonates with dynamic capabilities theory (Ortiz-Avram et al., 2024), which emphasizes the importance of organizational learning in adapting to changing business environments. Moreover, the lack of risk perception aligns with studies (Marshall, 2020) that have explained why firms become risk-averse when facing uncertain outcomes in CE initiatives. According to Madanaguli et al. (2023), risk aversion limits firms' willingness to engage in innovative practices, causing them to focus more on avoiding potential adverse outcomes rather than exploring new sustainable opportunities.

Moving on to Configuration 5, we observed an interesting paradox: although adaptability and communication channels are present, the absence of several critical factors namely inter-organizational trust, information sharing, risk perception, and shared goals severely constrains SMEs' engagement in effective circular practices. Thus, the absence of inter-organizational trust severely hampers the development of social capital for CE initiatives (Stål et al., 2023). In the case of Configuration 6, while adaptability and commitment to communication channels are evident, the lack of collaborative learning, inter-organizational trust, information sharing, and shared goals creates a significant void, preventing the development of a cohesive network where partners can effectively share knowledge and align strategies. These missing elements hinder the development of a cohesive network where partners can effectively exchange knowledge and align strategies, ultimately weakening the collective capacity to achieve circularity goals (Dzhengiz, 2020).

Meanwhile, configuration 7 presents a unique scenario where, despite the presence of several positive factors, including adaptability, communication channel commitment, inter-organizational trust, and information sharing, the absence of collaborative learning and risk perception still leads to diminished CSC capabilities. Subsequently, this deficiency hampers proactive risk management and knowledge-sharing practices. The lack of collaborative learning restricts firms' ability to jointly develop innovative solutions and share insights critical to addressing supply chain challenges (Shi et al., 2023). Similarly, without risk perception, companies may fail to proactively identify and mitigate potential disruptions, undermining resilience and adaptability in the supply chain (Masi et al., 2017). Thus, these deficiencies hinder the development of a robust CSC, as proactive risk management and knowledge-sharing practices are essential for sustaining circularity.

Finally, Configuration 8 reveals an intriguing dynamic: although multiple elements are present (adaptability, communication channels, collaborative learning, risk perception, and shared goals), the absence of inter-organizational trust and information sharing impedes progress toward CSC implementation. These missing elements prevent the establishment of reliable partnerships and effective knowledge exchange, both of which are essential for synchronized actions and innovative problem-solving across the supply chain. Without trust and open information sharing, organizations struggle to align their objectives, reducing their ability to effectively implement CE practices (Sudusinghe and Seuring, 2022).

We adopted the method described by Pappas et al. (2016) to evaluate how well the fsQCA model predicted outcomes. We split the dataset into two parts: one for developing the configurations and another (holdout sample) for testing their predictive power. This approach helps to avoid overfitting and ensures that the findings can be generalized, as suggested by Gigerenzer and Brighton (2009) and Woodside (2014). Through fsQCA analysis, we identified key configurations of factors in the subsample (see Table 9) that influence the levels of CSC in SMEs. We then applied these configurations to the holdout sample to assess predictive validity by comparing the consistency and coverage values of both samples. According to Pappas and Woodside (2021), if we find a high consistency, we can interpret the corresponding value as the coverage score. The results supported the predictive validity for both high and low CSC levels, confirming that the model is reliable and applicable to different datasets.

Figures 3 and 4 illustrate the predictive performance of the fsQCA models for high and low levels of CSC, respectively, using data from the holdout sample and the results from Table 9. These fuzzy plots visually compare the consistency and coverage values of the identified configurations, demonstrating their reliability in predicting CSC outcomes across different datasets.

The findings of this study provide valuable insights into the critical factors that can influence the adoption and implementation of circular supply chains among small and medium enterprises. First, our results revealed that no single condition can independently drive high or low CSC adoption. Instead, the study identifies multiple configurations of factors that can lead to either outcome, highlighting the equifinal nature of the pathways (Fiss, 2011). This aligns with the principles of complexity theory, which emphasize that organizational phenomena emerge from the interactions of various interdependent factors (Kumar et al., 2022). Again, the findings underscore the necessity of synergistic linkages among various conditions for advancing CSC in SMEs, revealing specific pathways to both high and low levels of CSC performance.

For pathways that yield high CSC levels, our results consistently demonstrate that adaptability and effective communication channels are foundational. This reinforces existing literature that highlights the critical role of adaptability and communication in enabling SMEs to respond flexibly and effectively to circular economy transitions (Alcalde-Calonge et al., 2024; Dey et al., 2020). In the SME context, adaptability is particularly crucial, as resource constraints mean that SMEs often rely on agile responses to overcome barriers related to finance, technology, and capacity for implementing circular practices (Malhotra and Manzoor, 2025; Stumpf et al., 2023). Similarly, efficient communication channels facilitate rapid and clear exchanges of information, reducing misunderstandings and delays, which are essential for the time-sensitive collaboration required in CSC (Bag et al., 2022; Gattiker and Goodhue, 2005). Notably, information sharing was not critical for high CSC levels. This is intriguing, as it diverges from previous studies that position information sharing as pivotal for CSC success (Koh et al., 2017; Nasir et al., 2017).

In SMEs, this discrepancy could be explained by the relative simplicity and immediacy of direct communication channels within smaller organizations, which may reduce the dependency on formal information-sharing mechanisms. Moreover, it may reflect SMEs' reliance on more interpersonal trust rather than structured information sharing to achieve operational goals, as suggested by the presence of trust and shared goals in high-performance configurations. This supports Prieto-Sandoval et al. (2018), who emphasize the importance of shared vision in circular initiatives.

The pathways to low CSC levels highlight the limitations SMEs face when lacking critical collaborative components. For example, the absence of collaborative learning and risk perception was a recurrent theme among low-performing configurations, signaling that without a learning culture and proactive risk management, SMEs are prone to stagnation in CSC progress. The absence of collaborative learning restricts the continuous exchange of innovations and best practices necessary for addressing circular challenges (Stumpf et al., 2023). Furthermore, without risk perception, SMEs may struggle to anticipate and mitigate CSC risks, resulting in a lack of resilience and adaptability, as suggested by previous research (Leung et al., 2022).

A fascinating insight from low CSC configurations is the role of inter-organizational trust. When trust is absent, even with other positive conditions like adaptability and communication, SMEs experience substantial obstacles in forming effective circular collaborations. This reinforces previous assertions by Stål et al. (2023) that trust is foundational for building the social capital required for circular economy partnerships. For SMEs, trust becomes especially crucial because they often lack the formal contractual mechanisms larger organizations employ to manage partnerships, relying instead on trust to coordinate actions and manage risks.

Lastly, configurations exhibiting low CSC levels highlight the complex, non-linear nature of CSC determinants in SMEs. In contrast to larger firms, SMEs may not have the luxury of robust, multi-tiered strategies. Consequently, any absence of core elements like trust, collaborative learning, or risk perception disrupts the alignment of resources and objectives necessary for advancing circular practices, underscoring the importance of targeted support for SMEs to develop these capabilities. The diversity in these configurations supports the idea of causal asymmetry Fiss (2011) in achieving CSC levels, as different combinations of conditions lead to both high and low CSC outcomes.

Collectively, these findings underscore the complexity of CSC implementation in SMEs and suggest that policymakers and business leaders should focus on fostering core collaborative and adaptive capabilities to support SMEs in the transition to CSCs, particularly by strengthening trust and risk perception. This approach could optimize SMEs' contribution to the CE, particularly in developing countries like Ghana, where support for SME-driven sustainability efforts is crucial for achieving broader CE goals.

The study’s findings highlight the importance of a multi-dimensional approach for strengthening CSC initiatives in developing economies, emphasizing organizational adaptability, communication infrastructure, trust, and risk management. SME managers should prioritize developing flexible organizational structures and investing in communication technologies to enable real-time information exchange. Additionally, they should establish adaptable standard operating procedures to accommodate new circular economy practices, ensuring their organizations can quickly respond to market changes and emerging opportunities (Alcalde-Calonge et al., 2024).

Risk management also emerged as a key factor for successful CSC implementation. SME managers should create risk assessment frameworks tailored to circular initiatives and establish mitigation strategies for identified risks. Regular risk review sessions and clear protocols for evaluating partnerships and circular opportunities are essential for ensuring effective risk management, as emphasized by Masi et al. (2017). Furthermore, fostering collaborative learning environments through knowledge-sharing platforms and inter-organizational learning sessions can drive innovation and improve CSC practices. Policymakers can support these efforts by facilitating industry-specific forums and funding mechanisms for SMEs to engage in collaborative learning networks.

Trust-building among supply chain partners is another critical aspect of successful CSC initiatives. Organizations should establish transparent processes and clear shared goals to foster trust and collaboration. Policymakers play a crucial role by developing frameworks that support long-term partnerships and protect the interests of smaller organizations in CSC networks, aligning with the principles outlined by Prieto-Sandoval et al. (2018). Additionally, government bodies and industry associations must create supportive institutional frameworks, including policies that incentivize circular practices, provide financial support for technological upgrades, and reduce regulatory barriers, particularly for SMEs in developing countries (Ouro-Salim and Guarnieri, 2023)

This study enhances the theoretical framework of Circular Supply Chain Management (CSCM) by examining how factors from Organizational Information Processing Theory (OIPT) and Stakeholder Theory such as trust, information sharing, adaptability, risk perception, collaborative learning, and communication channels interact to drive CSC in SMEs. The configurations identified suggest that CSC success in resource-constrained contexts may hinge on adaptable, trust-based relationships rather than rigid adherence to traditional information sharing. These insights not only challenge existing theoretical assumptions but also provide a robust framework for guiding future research and policy on CSC practices among SMEs in developing economies.

This study has some limitations, which present opportunities for future research. First, the study's reliance on qualitative methods and subjective assessments from SME stakeholders presents limitations. Future research could incorporate quantitative metrics like financial performance or environmental impact to strengthen the findings. Additionally, exploring other organizational factors such as innovation capability, leadership styles, and digital readiness could provide further insights. Future studies could also expand the research to include diverse geographic and cultural contexts to assess the generalizability of these findings and deepen the understanding of circular economy practices in SMEs globally.