Towards circular economy through innovation: the role of entrepreneurial orientation and human resource management

EO´s origin can be traced back to the seminal paper by Miller (1983), where he described entrepreneurial firms —EO firms— as risk-taking, proactive, and innovative companies. Risk taking is the willingness to embrace uncertainty and support bold projects and ideas; proactiveness can be defined as the willingness to take an opportunity-seeking perspective that anticipates future trends and demands; and innovativeness reflects a firm’s tendency to engage in and support new ideas, novelty, experimentation, and creative processes that may result in new products, services or technological progress. As stated by Wales et al. (2023) “entrepreneurial orientation (EO) creates new value through a commitment to continues novelty within an organization’s product-market offerings” (Wales et al., 2023, p. 1763). Thus, innovation is at the core of the EO concept (Lumpkin & Dess, 1996) and research has consistently shown that EO enables firms to cope with challenges arising from disruptive innovation (Kraus et al., 2023; Rigtering et al., 2024). Following this line of reasoning, we view EO as an indispensable tool for coping with disruptive changes in nature, such as the transition from a linear economy to a CE (Kuhlmann et al., 2022). Similarly, Suchek et al. (2022) argue that entrepreneurial attitudes and behaviors enable firms to capture new business opportunities arising from the CE and Engzell and Kambanou (2024), highlight the importance of entrepreneurial spirit in the adoption of circular business models. In the same line, Mohapatra et al. (2024) find that entrepreneurial motivation leads to several actions related to business processes that are aimed at circular value creation.

Despite its relevance, the role of an EO in the CE context has been little explored in the literature, with the exception of Khan et al. (2022), who examine how small firms can minimize CE challenges in resource-, process-, and people-related aspects by properly configuring their EO. A related stream of the literature applies the EO concept in the context of sustainability. For example, Andersén (2022) finds a positive relationship between an EO and green product innovation among small Swedish manufacturing firms. Although green and CE concepts are both environmentally friendly concepts, CE adoption is potentially disruptive, requiring fundamental changes in firms’ operations, market offerings and manufacturing processes (Kuhlmann et al., 2022), which may not be the case for green adoption. Some scholars attempt to incorporate the element of green or sustainability into the EO construct and propose constructs such as a “green EO” (Ameer & Khan, 2022; Khan et al., 2023) and a “sustainable EO” (Criado-Gomis et al., 2018). Since innovation is at the heart of EO, we believe that the constructs of green EO and sustainable EO already encompass the relationship we want to examine, i.e., how an EO relates to circular innovation. For this reason, we decide to follow Miller’s (1983) definition of EO and exclude studies on green EO and sustainable EO from our literature review.

The CE can be described as a restorative system in which waste is eliminated and resource use is minimized as materials are continuously looped back into the system (MacArthur, 2013). The successful implementation of this closed-loop system requires fundamental changes in the firm product design and manufacturing processes as well as in management practices (Ferasso et al., 2020), changes referred to in this study as “circular innovation”. As Kirchherr et al. (2017) note, there is currently no consensus regarding the definition of “circular innovation”. Some scholars propose using the term “eco-innovation” to refer to changes resulting from the implementation of CE principles. The European Commission (2018) defines “eco-innovations” as “… all forms of innovation– technological and non-technological– that creates business opportunities and benefits the environment by preventing or reducing their impact, or by optimizing the use of resources” (p. 1). Thus, although eco-innovation is related to the concept of circular innovation, the two concepts are not identical. Some eco-innovations share common characteristics with circular innovation, such as the support of energy efficiency and the exploitation of renewable resources; however, others do not necessarily implement CE principles despite their favorable effect on the environment (Scarpellini et al., 2020).

To address these potential differences, some researchers have recently begun to use the term “circular eco-innovation” to refer to eco-innovations that aim to circulate resources in loops of reuse, recycling, and renewal (Scarpellini et al., 2020). Since one objective of our study is to investigate the relationship between circular innovation and environmental performance, we prefer to use the term circular innovation instead of eco-innovation or circular eco-innovation. This is because the definition of circular eco-innovation includes the objective of benefiting the environment. Consequently, it becomes a tautology to examine the impact of circular eco-innovation on environmental performance. However, this is not the case for circular innovation. As Blum et al. (2020) observe, there is no guarantee that a circular solution is more sustainable than its linear equivalent. This uncertainty creates an opportunity for us to add value to the CE literature by investigating how circular innovation affects firms’ environmental performance.

As in other types of innovation, firms face high technological uncertainty when developing new circular products (Linder & Williander, 2017). Moreover, using CE principles to design new products often involves incorporating the latest technologies, which are less mature and less stable. However, an EO implies that the firm is constantly conducting bold experiments to develop new technologies that facilitate product and new value creation (Wales et al., 2023). Additionally, EO firms tend to engage in higher levels of information scanning (Barringer & Bluedorn, 1999), sharing, and exploitation (Matsuno et al., 2002) to leverage their innovative processes than non-EO firms. Thus, EO firms are more likely to accumulate greater innovation experience over time and develop relevant competences to cope with technical uncertainty (Covin et al., 2020).

The introduction of new circular products is also associated with high market uncertainty (Linder & Williander, 2017). As the concept of CE is still relatively new, circular product innovation may fail to attract potential customers. There is also the possibility of cannibalization, resulting in tension between the new and existing products (Kuhlmann et al., 2022). Faced with these challenges, EO firms embrace the risks associated with circular innovation and proactively involve their current and potential customers in new product development activities (Zucchella et al., 2022). Co-creating new products with customers improves customer loyalty, prevents cannibalization and leads to an increasing market acceptance of the new circular products, ultimately improving innovation rates (Wang et al., 2020).

EO firms are also more likely to be ready to act when new market opportunities arise due to their proactive and highly forward-looking nature (Dess & Lumpkin, 2005). Although it can be expensive to develop circular products (Demirel & Danisman, 2019), circular product design is self-sustaining and linked to material efficiency (Genovese et al., 2017), which ultimately improves firms’ productivity. In addition to enabling cost savings, firms can also benefit from the additional value created by the extended life of circular products (Murray et al., 2017). All of these benefits make circular product innovation attractive to firms and increase their willingness to sacrifice short-term gains in exchange for medium- and long-term benefits, which is particularly attractive to EO firms (Wales et al., 2023). Thus, EO firms often see the transformation of their business activities according to circular principles as an attractive business opportunity (Ghisellini et al., 2016). Based on the above arguments, we propose the following:

Despite its potential to reduce costs (Darmandieu et al., 2022) due to greater efficiency in resource use (Ghisellini et al., 2016), implementing CE principles in the manufacturing process can be challenging (de Arroyabe et al., 2021). Employees may resist the new manufacturing processes because they change existing work practices and are likely to disrupt employees’ work routines and/or slow their progress. However, EO firms are better positioned to deal with this challenge because these firms have organizational structures and management practices tailored to support adaptation and change (Suchek et al., 2022). Thus, employees in EO firms are likely to view task disruption as temporary, reducing their resistance to change. Another potential challenge associated with circular manufacturing processes is the complexity of setting up these processes. Because circular manufacturing often requires relatively new technologies, not all firms have sufficient knowledge to adopt them. Due to their willingness to proactively seek out and share relevant information (Barringer & Bluedorn, 1999; Matsuno et al., 2002), EO firms are more likely to have leading edge knowledge. If not, they are more likely to search for partners to access relevant technologies (Veleva & Bodkin, 2018). Implementing CE principles in the production process is also costly and technically uncertain, which may make firms hesitant. EO firms are more likely to take risks if they foresee significant benefits in the future. Therefore, we suggest the following:

The literature agrees that employees are valuable resources in achieving organizational goals (Jabbour et al., 2019). Thus, because employees interact with and behave according to management practices (Santos-Vijande et al., 2012), implementing changes in HRM practices is essential to capture the opportunities created by the CE. In other words, moving toward CE adoption requires implementing appropriate HRM practices aimed at developing employee commitment and competence with CE (Erdiaw-Kwasie et al., 2023; Veleva et al., 2017).

EO firms are likely to interpret CE as an opportunity and, thus, to act upon it (Suchek et al., 2022). Because of their ability to reconfigure resources to create new value (Wales et al., 2023), EO firms are likely to reconfigure human resources so that CE principles are incorporated into their HRM practices—CHRM. Different studies have confirmed that EO firms are more experienced in developing employees´ competences by implementing knowledge creation and sharing systems (Jiang et al., 2019). EO firms also tend to be proactive in providing employees with cutting-edge knowledge through training, which empowers and motivates them in the desired direction (de Angelis, 2021). Nevertheless, incorporating change into HRM practices is challenging. Not only do organizations have to deal with implementation costs, but they also face the risk of employee turnover (Islam et al., 2020). Since EO firms are more willing to take risks for long-term development, we argue that they are more likely to assume the risk of CHRM. Thus, we propose the following:

According to the literature, a workforce with CEcapabilities would enable firms to identify and pursue CE opportunities (Khan et al., 2021) and implement circular manufacturing (Scarpellini et al., 2020). Marrucci et al. (2022) highlight the relevant role of employees in assimilating the CE knowledge, which is needed to design circular products (Erdiaw-Kwasie et al., 2023) and to reduce costs and improve the implementation of circular manufacturing processes (Darmandieu et al., 2022). On the one hand, a workforce with CE knowledge can better understand customer needs in the CE domain and design new circular products accordingly. On the other hand, employees with better CE knowledge can help firms to select circular practices that fit best with the production processes, and ensure proper implementation (Marrucci et al., 2021).

Although adequate recruitment and training are two ways to improve employees’ CE knowledge and skills and constitute key practices for developing circular innovation (Pronti et al., 2024) and implementing circular manufacturing (Beducci et al., 2024), an adequate reward system is also necessary to incentivize employees to adopt CE (Jabbour et al., 2019; Marrucci et al., 2024). Studying the CE adoption in established firms, Kuhlmann et al. (2022) find that circular product and production innovations are more likely to occur when firms empower employees to participate in organizational efforts toward circular practice implementation. Similarly, Singh and Singh (2018) also find that empowering employees increases job satisfaction in CE contexts, which in turn enhances CE initiatives. As Dragan et al. (2024) and Veleva et al. (2017) indicate, enhancing employee engagement through proper communication and empowerment is critical to ensure advancement toward circular practice implementation. In the same line, Soni et al. (2023) highlight the relevance of the employees´ empowerment in CE adoption by emphasizing the role of distributed leadership. Given that the implementation of CHRM involves recruiting talent with CE values and providing CE training to enhance employees’ CE skills and knowledge, designing appropriate performance appraisals and rewards to encourage CE initiatives, and providing employees with opportunities to autonomously participate in decision making on CE-related issues, we propose the following:

While we argue for positive associations between EO and CHRM and between CHRM and technological circular innovation, we also point out that CHRM mediates the relationship between EO and technological circular innovation. In this sense, Liu et al. (2023) recently demonstrate the critical role that HRM systems play in translating a CEO’s entrepreneurial orientation into employee innovative behavior.

Examining the technological features of environmentally friendly innovations, Petruzzelli et al. (2011) show that environmental technological innovations tend to be more complex and novel than other types of innovations. Anticipating the need to cope with increasing complexity and novelty, we expect that EO firms are more likely to respond by reconfiguring their resources, due to their tendency to be proactive, to facilitate circular innovation. As Rivera-Torres et al. (2015) show, EO firms often make changes in their organizational design and management practices while implementing pro-environmental practices in their product design and production processes. Sawe et al. (2021) also recommend that entrepreneurial firms focus on people-driven factors to achieve successful CE adoption. Therefore, we propose the following:

Compared to technological innovation in a traditional linear model, technological circular innovation emphasizes not only efficiency but also resource value retention (Kirchherr et al., 2017). Thus, going beyond slowing down resource consumption, circular product and circular manufacturing innovation allow resources to be used more than once. For example, by designing new products with as many recyclable materials as possible, firms allow these resources to re-enter a new production cycle at the end of their use. Similarly, waste and residues can have a “second life” if they are integrated into a new production process. Because of the ability of technological circular innovation to create multiple lives for resources, it is deemed an environmentally valuable resource (Singh et al., 2020).

As suggested by the resource-based view, owning a valuable resource enhances firm performance (Barney, 1991). Chan (2005) defines environmental performance as “meeting and exceeding societal expectation with respect to concerns about the natural environment” (p. 632). Based on this definition, we argue that circular products and manufacturing processes will meet and exceed societal expectations by complying with environmental protection regulations, pursuing efficiency and creating multiple lives for resources (Kirchherr et al., 2017), thereby reducing negative environmental impacts. This is consistent with Zhang et al.’s (2022) finding that investments in circular innovation positively impact firms’ environmental performance. Similarly, Valero-Gil and Scarpellini (2024) prove that firms that successfully register patents related to waste reduction achieve improved environmental performance. Furthermore, a recent meta-study covering more than 13,000 manufacturing firms also confirms that preserving the inherent value of products and materials for reuse leads to improved environmental performance (Pan et al., 2024). Therefore, we propose the following:

Technological circular innovation also creates market value by delivering value to customers. Circular manufacturing is characterized by resource efficiency, which leads to cost savings and potentially lower prices for customers (Bag et al., 2022). Incorporating CE practices such as waste management and recycling into the manufacturing process can also enhance a company’s brand reputation (Mazzucchelli et al., 2022). Circular products are more durable, which is valued by consumers because they can be repaired and upgraded (de Arroyabe et al., 2021). Technological circular innovation can also create value for customers by addressing public concerns. Kim and Hall (2020) find that customers perceive value when dining in restaurants that employ food sustainability and waste reduction practices. This is because customers perceive value through self-gratification from engaging in benevolent activities that also benefit the society. In this line of reasoning, Tsai et al. (2023) find that emotional perceptions increase individuals’ willingness to use the service provided by CE-based platforms. Thus, we argue that technological circular innovation creates value for customers and, thus, can be considered a valuable market resource.

Based on the resource-based view (Barney, 1991), we expect that owning circular innovation will improve firms’ market performance because firms can use their circular products to develop differentiation strategies and build a strong brand to attract customers who prioritize environmental issues (Mazzucchelli et al., 2022; Reinhardt, 1998). Previous studies have shown that this type of customer is willing to pay a price premium for products designed or manufactured according to environmentally friendly principles (Reinhardt, 1998). Circular product innovation and manufacturing also signal to customers that firms are committed to environmental issues, which is likely to enhance the loyalty of customers who share similar values (Kim & Hall, 2020). In fact, Bataineh et al. (2024) confirm that firms engaging in activities leading to reduction of energy or material use are likely to achieve a competitive advantage over their peers due to improved public image and reputation. Thus, we hypothesize the following:

Technological circular innovation can act as a socially valuable resource to improve a company’s social performance in several ways. First, environmental and social issues are often linked together, so that firms´ efforts to address environmental issues often have social consequences (Dey et al., 2020). For example, by integrating waste into the production cycle, firms not only incorporate CE principles but also improve local residents’ well-being by eliminating potentially harmful residuals. Using a case study approach, Cullen and De Angelis (2021) illustrate how the initiative to use local apples, which would otherwise have been wasted, as a key ingredient in the beverage manufacturing process benefits the local community by better conserving nature capital and building local social capital. By adopting circular principles in product design and manufacturing processes, firms are also likely to create new jobs (Moreno-Mondéjar et al., 2021). In addition, technological circular innovations often require employee involvement (Dragan et al., 2024), which is likely to increase job satisfaction (Iranmanesh et al., 2017).

Furthermore, the introduction of circular products and the implementation of circular manufacturing processes demonstrate the firm interest in taking a proactive approach to environmental issues, which can be interpreted as a proactive response to stakeholder concerns. As suggested by Shaukat et al. (2016), a proactive approach to stakeholder concerns is likely driven by corporate governance characteristics, which tend to be persistent over time. Thus, we expect that these firms would also respond proactively to societal concerns. Indeed, Dias-Sardinha and Reijnders (2005) found that firms that demonstrate the ability to address environmental issues tend to have better social performance, likely because they proactively seek solutions to meet stakeholders’ social expectations. As Zhang et al. (2022) argue, patenting environmentally friendly technologies often helps SMEs improve their corporate image in the eyes of employees and other stakeholders. Therefore, we suggest that:

The empirical analysis is based on a cross-sectional database of Spanish firms involved in CE practices and operating in different economic sectors. To this end, we identified a population of 1,357 companies using information published by private associations dedicated to the promotion of circular practices, ministerial reports, news and various sectoral reports developed by consulting firms. Primary information was collected through a survey conducted by a market research company. Qualified respondents were defined as CEOs or senior managers to ensure that they had sufficient knowledge about the strategic direction and CE practices implemented in their companies. After eliminating incomplete responses, the final sample consists of 218 organizations (16.06% response rate).

The description of the sample of companies can be done (a) by sector: plastics and rubber (15.84%), metallurgy and metal products (10.41%), automotive (9.95%), textiles and clothing (9.50%), construction (8.14%), packaging (8.14%), furniture, wood and cork (7.69%), other non-metallic products (7.69%), chemicals (6.79%), mechanical machinery and equipment (6.33%), food and beverages (4.98%), and electronics and TIC (4.52%); (b) by company age: up to 10 years (11.76%), 10–25 years (25.34%), 26–40 years (38.46%), more than 40 years (24.43%); (c) by turnover: up to 300,000 € (7.3%), 300,001 € − 600,000 € (8.7%), 600,001 € − 1,500,000 € (14.7%), 1. 500,000–3,000,000 (13.8%), more than 3,000,000 (55.5%); and (d) by company size: up to 10 employees (17%), 11–49 employees (39%), 50–249 employees (27.5%), more than 250 employees (16.5%).

Following Armstrong and Overton (1977), we conducted a nonresponse bias test. After analysis, no significant differences were found between early and late respondents, so the results do not reveal a nonresponse bias problem. In order to avoid common method bias (CMB) in the data, we applied ex-ante remedial procedures in the questionnaire design and tested for the presence of CMB ex-post using the marker variable technique suggested by Lindell and Whitney (2001). In this case, we used an industry-related item (“customers easily switch suppliers”) as a marker variable. In all cases, the correlations of the model constructs with the marker variables are not significant, ranging from 0.014 (sig = 0.84) to 0.044 (sig = 0.56). Finally, we recalculated the construct correlations in the research model, excluding the lowest positive correlation between the constructs and the marker variable. The results show that all correlations are not significant after adjustment. Therefore, it can be concluded that common method bias (CMB) is not a problem in this study.

We used 7-point multi-item scales to measure the constructs in the conceptual model in Table 2. EO is measured as a second-order reflective construct that includes the three first-order dimensions established by Covin and Slevin (1989): innovativeness (Brettel et al., 2015), risk-taking (Chen et al., 2012), and proactiveness (Chen et al., 2012). The CHRM scale is adapted from Del Giudice et al. (2021) and Jabbour et al. (2019), including for the first time a reference to employee selection, training, compensation, consciousness and active involvement in CE implementation. The scales for measuring circular product innovation and circular manufacturing are adapted from the studies of de Arroyabe et al. (2021), Bag et al. (2022), and Khan et al. (2021). The scales for measuring environmental, market, and social performance were inspired by Jabbour et al. (2020). Finally, technological turbulence was measured using a scale based on Chen et al. (2012). We included firm turnover and firm age as control variables.

In order to analyze the psychometric properties of the reflective scales we use covariance-based structural equation modelling (CB-SEM) with STATA software. Table 2 summarizes the measurement scales used and the psychometric properties of the scales. Due to the limitations imposed by the size of the sample in relation to the observable variables in the research model, we first tested a CFA model in which the underlying dimensions of EO are allowed to correlate. Once this first model showed a good fit and the adequacy of the scales’ properties, we tested a second model in which EO, represented by the mean values of its three underlying dimensions, and the rest of latent variables in the research model are correlated. Former exploratory factor analysis of the scales used in the research model revealed the presence of two distinct factors in the circular product innovation scale, whose configuration is coherent with the two main motivations for circular product design: (1) to narrow resource flows, or (2) to slow down or close resource loops. Circular product innovation involves creating new products or redesigning existing products to be more efficient throughout their life cycle (EMF, 2015). The aim is to develop goods and services that: (1) use fewer resources that are renewable and/or recycled whenever possible, (2) have high quality components that allow their lifecycle to be extended, and (3) are easier to maintain, repair, upgrade and recycle, whether for the original purpose or for other purposes. These initiatives achieve both the so-called narrowing of resource flows (or resource efficiency) and the slowing down or closing of resource loops (through reduced resource use) (Jensen, 2018). Accordingly, we considered two distinct factors in the second measurement model where EO, and the rest of the constructs in the research model correlate (see Table 2). We then used CFA to test the convergence of the two dimensions of circular product innovation into a second-order factor, in order to use circular product innovation as a single construct in the Seemingly Unrelated Regressions (SUR) model utilized to test the hypothesis. The goodness of fit indexes of this model will support this approach.

The analysis of the psychometric properties of the scales (reliability, convergent and discriminant validity) shows that most of the item loadings are above the threshold of 0.7, and the associated t-statistic is statistically significant (Anderson & Gerbing, 1988), which suggests internal consistency. Furthermore, Cronbach’s alpha and the composite reliability index (CR) values exceed the recommended threshold value of 0.7, indicating construct reliability. The average variance extracted (AVE) values range between 0.546 and 0.820, which are above the acceptance levels established by Hair et al. (2011) to assess convergent validity.

To determine the discriminant validity, we considered the Fornell and Larcker (1981) criterion and Heterotrait-Monotrait ratio of correlations (HTMT) criterion -Table 3. The results confirm that the square root of the AVE of each construct is greater than the cross-correlations between each construct and other constructs in the model and is not less than 0.50. Additionally, the HTMT values are less than 0.90 for conceptually similar constructs. We also confirmed that multicollinearity is not a problem in our study because the variance inflation factor (VIF) scores range between 1.21 and 4.50, well below the commonly threshold of 5.

In this study, we used a SUR method to test the proposed hypotheses. This model consists of multiple regression equations, so PLS-SEM could be used for its analysis. However, SUR allows simultaneous evaluation of equations and is more effective than SEM in estimating relationships between variables (Chen et al., 2023; Gao et al., 2022). In addition, SUR, also known as multivariate regression or Zellner’s method (1962), estimates system parameters taking into account heteroskedasticity and contemporaneous correlation of errors across equations (Bočkus et al., 2023).

The use of the SUR allows for more efficient estimation by combining information from different equations and mitigating potential endogeneity problems (Autry & Golicic, 2010; Cambra-Fierro et al., 2020). For example, previous studies emphasize that multi-linear systems of equations produce more efficient estimates when the error terms of different regressions are correlated, as in the case of SUR (Autry & Golicic, 2010; Gao et al., 2020). Similarly, “since some variables are dependent and independent variables in different regressions, this technique allows us to alleviate endogeneity problems that can potentially present in the data” (Autry & Golicic, 2010, p. 95). Thus, although an ordinary least squares (OLS) method could be used to test the hypotheses, the SUR method is particularly useful when the model has more than one dependent variable, as the regressions can be run simultaneously (Cambra-Fierro et al., 2021).

Furthermore, our research model includes several dependent variables that represent the “two sides of the same coin” such as circular product innovation and circular manufacturing (technical innovation) and CE performance (environmental, market, and social). In this case, SUR is a suitable method to estimate the parameters of all our proposed model equations simultaneously, as the parameters of each equation take into account the information provided by other equations (Wang et al., 2024). The result is an increase in the efficiency of parameter estimation as additional information is used to describe the system. Accordingly, SUR models have been used in previous studies for similar purposes (Cambra-Fierro et al., 2021; Gao et al., 2020, 2022).

In this case, we developed a six-equation SUR to empirically test the conceptual framework and related hypotheses proposed in Fig. 1. In this step, we used the composite variable scores obtained in the CFA estimation for the variables measured with multi-item scales (Chen et al., 2023). The first linear regression analyzes circular product innovation. Second linear regression examines circular manufacturing. Third linear regression examines CHRM. Fourth linear regression explains environmental performance. Fifth linear regression analyzes market performance. Finally, the sixth linear regression explains social performance. In all these cases, we use firm age, number of employees, and sales as control variables. In addition, we included the moderating role of technological turbulence on performance (market, social and environmental). The linear equations in the SUR model are presented below:

Where β is the unknown coefficient and ε is an error term of the model that is independently and equally distributed.

Motivated by the need to accelerate CE adoption in established firms and answering to the call of further research on the drivers and barriers in the CE adoption (Engzell & Kambanou, 2024), this study confirms the relevance of EO as a driver of technological and non-technological circular innovation. Our findings contribute to the literature in several ways. First, we advance knowledge about EO by highlighting its potential in managing disruptive change such as CE adoption, (Kuhlmann et al., 2022; Mohapatra et al., 2024), extending the list of CE drivers. Thus, our study provides evidence that EO is a valuable strategic orientation to overcome the potential barriers associated with CE adoption (Sonar et al., 2023). To this end, we view the adoption of CE as a holistic innovation effort across technological and non-technological domains, highlighting the breadth and depth of change required to put CE principles into practice (Reike et al., 2023). From this perspective, our study confirms that EO contributes to the firm’s technological circular innovation, that is, to the development of circular product innovations and the implementation of circular manufacturing processes. Moreover, in addition to determining the effect of CE adoption on technological innovation, we go beyond and examine the effect of CE adoption on non-technological innovation, in line with the suggestion of Ahmad et al. (2023). Specifically, we propose, measure, and validate the construct of CHRM and prove that CHRM is a key mechanism through which EO leads to circular product innovation and circular manufacturing, demonstrating (1) the relevance of EO in transforming HRM to incorporate the values of CE, and (2) the key role of CHRM in the adoption of circular practices in the technological domain. In fact, our results show that the impact of EO on CE adoption is stronger in the case of non-technological innovation and show that EO promotes circular innovation not only directly, but mainly indirectly through CHRM, which is a novel contribution to the EO literature. Furthermore, our results also show that EO in firms directly leads to performance improvements in the environmental, market, and social domains by promoting circular innovation, which allows us to respond to recent calls in the literature for a better understanding of the impact of EO on performance (Wales et al., 2023), especially in the context of environmental concerns (Ameer & Khan, 2022) and CE (Mohapatra et al., 2024). Additionally, our work is innovative research into a key knowledge cluster recently identified by Muldoon et al. (2024), referred to entrepreneurship for sustainability and CE.

In contrast to most previous studies focusing on certain performance dimensions (Jabbour et al., 2020; Khan et al., 2021), we simultaneously evaluate the impact of circular innovation on environmental, market, and social performance. This allows us to respond to the call to examine how CE practices affect performance from a holistic perspective (Yin et al., 2023), as well as to fill the research gap identified by recent review articles (Pan et al., 2024; Magnano et al., 2024) regarding the lack of studies on the relationship between CE adoption and social performance. To the best of our knowledge, only Le et al. (2022) and Obeidat et al. (2023) have tested and found a positive relationship between CE adoption (not green nor eco) and firm performance across the three domains. However, both studies differ from ours in two important ways. Our study focuses on circular innovation; however, the aforementioned studies examined CE adoption in the context of business models and value creation (Le et al., 2022) or exclusively in the manufacturing process (Obeidat et al., 2023). In addition, both studies treat firm performance as a single construct with items measuring the environmental, economic, and social domains. Instead, we consider economic, market and social variables as three distinct but interrelated variables and analyze their relationships with CE adoption respectively. Given the development of CE is imperative, firms need to have a clear understanding of the performance implications of CE in order to justify their adoption (Bag et al., 2021; Yin et al., 2023). By treating firm performance in environmental, market and social domains as three distinct but interrelated variables, our empirical results reveal that adopting CE not only improves environmental performance but also allows firms to cultivate a better market image and to satisfy demands from different stakeholders. That is, our findings help alleviate concerns about potential conflicts between firm performance in the three domains (Iranmanesh et al., 2017; Kuhlmann et al., 2022; Schroeder et al., 2019; Walker et al., 2022). This provides a full set of incentives for firms to transform towards CE. Finally, despite the prevailing technological turbulence in many markets, which may hinder the benefits of circular innovation, our results overall reinforce the positive impact of circular innovation on firms’ environmental, market and social performance in turbulent markets, which reinforces the relevance of understanding how to favor firms’ CE adoption.

In addition to contributing to the literature, our findings have several managerial implications. Our results highlight the importance and need for firms to cultivate a strong EO to support the adoption of CE. Adopting CE is not without risk. It requires significant organizational change to capitalize on new market opportunities. In order to unleash the potential of firms to take advantage of emerging CE opportunities, managers are urged to develop their firm’s EO as a key tool to facilitate the transformation to CE. To achieve this goal, managers must first develop an adequate understanding of CE in order to anticipate the new opportunities associated with the non-linear economy. Attending CE-related events and workshops and collaborating with knowledgeable partners are promising avenues for knowledge acquisition. In addition, managers need to encourage risk-taking and proactively assess all the requirements for developing circular products and manufacturing processes, which may include actively seeking the necessary technologies and collaborations with partners (Aarikka-Stenroos et al., 2022) to achieve adequate implementation of CE. Managers also need to pay special attention to HRM from a circular perspective, as CHRM is a critical variable for designing new circular products and implementing circular manufacturing. Managers are therefore advised to include CE values in recruitment processes and to provide specific training for employees. Creating a compensation and reward system that recognizes employees’ CE initiatives and encourages interaction among interdisciplinary work teams will also encourage the development of business activities based on CE principles. It is also useful to note the heterogeneous effects of technological circular innovation across the three performance domains, which provides managers with options to prioritize CE adoption according to their needs. For firms seeking to improve their environmental performance, implementing CE principles in the production process would be the more effective path. Finally, a workforce with innovative CE knowledge and skills can also enable firms to cope with the increasing demands from a variety of stakeholders. In this respect, our findings have implications for the management of business schools. To the extent that their students may be future employees and managers, strengthening their awareness and knowledge of CE principles and related market opportunities is crucial. Therefore, we recommend that business schools include CE issues in the curriculum. This would help future workers and managers to develop CE competencies, leading to an accelerated transformation towards CE.

Despite its contribution and relevance, our research has limitations that provide opportunities for future studies. Our empirical evidence is obtained by analyzing Spanish firms, which may limit the generalizability of the findings. Also, using the SUR method with a larger and more diverse sample in terms of the type of countries analyzed would allow better generalization of the results. Thus, scholars could replicate the study in different national contexts or analyze firms in different countries from a comparative perspective to test the boundary conditions of our research model. In addition to national differences, it would also be interesting to analyze how our proposed model varies across industries. This would require researchers to collect data from a larger sample size, making it possible to explore the differences and similarities across sectors.

We are also aware that EO is unlikely to be the only antecedent of firms’ transformation to CE. Recent studies suggest the relevant role of firms’ ambidexterity in the pursuit of frugal innovation, which is also closely related to circular practices (López-Sánchez & Santos-Vijande, 2022). In addition to CHRM, future studies should also investigate other non-technological innovations to promote technological circular innovation. Another line of research is to clarify the boundary conditions of the relationships by searching for additional potential moderators and trying to clarify the role of technological turbulence in different contexts. Our study is also limited in terms of the measurement of CE principles. We treat CE as a homogeneous concept without distinguishing its specificity. Although they share the same idea of circularity, the implementation of the 3R, 6R, and 9R principles (de Arroyabe et al., 2021) may require different knowledge and skills. In the future, scholars could use a more fine-grained measure to explore heterogeneity within CE concepts. Finally, our data are cross-sectional, which limits the ability of our analysis to clarify the causality between variables. The single-country nature of the dataset also calls for caution when attempting to generalize the results.