Different but the Same? Comparing Drivers and Barriers for Circular Economy Innovation Systems in Wood- and Plastic-Based Industries

Circular economy is an emerging concept that places an emphasis on strategies (e.g., reduce, reuse, recycle) to decouple resource use from economic growth, minimize waste and emissions, and maintain the highest utility along a product life cycle. The transition to a circular economy requires innovative solutions along entire value chains. This literature review was carried out to investigate the respective innovation systems that emerge along the wood-based and plastic-based value chains. To investigate different barriers to and drivers for the transition to a circular economy, the system functions of the technological innovation system framework were used. The results reveal that the two sectors hold different strategic positions and that barriers are dominant in the innovation system for plastics, while drivers are more prevalent in the innovation system for wood. This study is one of the first to direct a focus toward different industrial origins and their underlying logic, contributing to a better overall understanding of the circular economy.


Introduction
Global grand challenges such as climate change, resource depletion, or deforestation force humanity to act on and to create a more sustainable future [1]. In this context, the circular economy (CE) concept has received an increasing amount of attention, not only from researchers [2][3][4], but also from political authorities and institutions (European Commission (EC) [5,6], Ellen MacArthur Foundation (EMF) [7]), whereas private businesses are considered to play central roles in the development of the CE [8,9]. The CE concept has been developed to decouple resource use and economic output, emphasizing reusing, recycling, and reducing strategies, minimizing waste and emissions, and maintaining products at their highest levels of utility [10]. The European Commission's action plans for CE address several priority areas, including plastics, food waste, critical raw materials, construction, and biomass/biobased products [5,6]. As described in these plans, the CE covers a broad spectrum of all kinds of fossil-based (i.e., non-renewable) as well as bio-based (i.e., renewable) innovations that are being put into practice to increase circularity [7]. Therefore, several scholars have highlighted the importance of forming alliances among fossilbased and bio-based industries in order to enable the development of CE innovations and to encourage them to share knowledge, raw materials, technology, and information [11].
Next to non-renewables, bio-based innovations (new materials, chemicals, or processes) are considered to be integral parts of a CE, because renewable resources are not infinite either; therefore, they should be used as efficiently as possible (e.g., through cascading utilization (CU)) [5,12]. However, resource types are not only differentiated by the nature of the component materials, but also by how the resources can be handled in a circular way, as illustrated in the EMF's butterfly diagram [7]. This seems to be especially true in the wood-based and plastic-based industries since these two materials will be strongly affected by the transition to a CE. In this study, the wood-based industry is defined as industry pursuing activities that include woodworking with different kind of woods, such as producing furniture, pulp and paper manufacturing and converting industries, according to the EC 1 (2022). Not relevant for this study was food and agricultural production. The plastics-based industry is defined as industry which includes actors such as plastics producers, plastics converters, plastics recyclers of common fossil-based polymer types but not bio-based plastics, according to PlasticEurope 2 (2021). For reasons of comparability, we have limited this analysis to the two established and mature industries, the wood-based and plastic-based industry, and their value chains, respectively. Given the increasing use of these materials in a wide range of products and energy sources, and the importance of preventing pollution [13][14][15], there is interest in increasing the reusability, recyclability, biodegradability, as well as the potential cascading use of these materials to return them to the system. When considering these two industries, scholars are encouraged to refer to the EMF's butterfly diagram, which depicts a strong association between the wood-based industry and the bio-based side, which supports returning materials to the biosphere, and between the plastic-based industry and the technical side, which supports resource recycling in closed loops [16].
Moreover, both materials are associated with sustainability problems. According to the FAO 3 (2018), the global roundwood production amounted to 3912 million m 3 in 2018. A new data series collected to monitor the progress of the transition to the "circular bioeconomy" shows that the amount of post-consumer recovered wood consumption has exceeded 27 million tons (Mt). Because the wood resources are limited, products must be maintained at the highest level of utility [10], but this goal conflicts with a growing demand for wood-based products [17] and the steady increase in the utilization of wood for energetic purposes [18,19]. Although the use of wood products from sustainably managed forests has increased significantly, contributing to improved recycling and material substitution, the sustainability discourse raises the question of whether it would be more appropriate to retain wood in forests for carbon storage, since deforestation is responsible for about half of the greenhouse gas emissions generated by the AFOLU (agriculture, forestry, and other land uses) sector [20]. A large share of the available wood is currently used for energy generation; this represents the largest share of biomass for renewable energy generation [21]. In addition, this increased use of wood may have adverse effects on the environment, resulting in deforestation, land-use change, or biodiversity loss [22,23]. The incorrect disposal and management of plastic materials, meanwhile, causes widespread pollution that affects various components of the ecosystem and increases the GHG emissions from plastic production [24,25]. According to PlasticsEurope 4 (2021), the global plastic production amounted to 367 Mt, 29.5 Mt (2021) of which were collected as post-consumer waste. Only 34.6% of the collected plastic waste was recycled, 42% was used as energy recovery, and 23.4% ended up in the landfill. Furthermore, plastic production accounts for around 4-5% of the annual petroleum production.
Previous efforts of businesses to develop solutions aligned with the defined CE strategies have shown varying degrees of success. In 2016, 73% of the companies surveyed (N = 10,618) undertook some CE activity, according to the Eurobarometer study 441 on CE implementation among European SMEs [26]. The latest Circularity Gap Report 5 (2022), however, stated that only about 8.6% circularity has been reached in terms of global material flows. Apparently, CE implementation is still in the early stages [27,28]. Thus, it is not a surprise that the body of CE-related literature that addresses barriers to and drivers for a CE is steadily growing (e.g., [29][30][31]). However, in previous empirical research on CE diffusion and adoption, researchers mainly focused on specific branches rather than crosssectoral developments [9,32,33]. At the same time, little attention has been paid to sectoral differences in CE-related activities [34], even though industrial origins are an important predictor of such activities [35,36]. In other words, a need to analyze the peculiarities and differences among different industrial contexts still exists [9], considering both the organizational and system levels.
Previous research on barriers to and drivers for a CE has tended to neglect the fact that business activities are embedded in wider sociotechnical systems. Such systems are clearly described by the so-called technological innovation system (TIS) framework, a conceptual approach that has received an increasing amount of attention from innovation and transition scholars in recent years [37][38][39]. This framework is used to analyze innovation barriers and drivers from a system perspective [40][41][42] and highlights the fact that innovations emerge from a system of actors, networks, and institutions that jointly interact and contribute to the development and diffusion of new products or processes [43]. Furthermore, the framework presents a set of key processes (so-called system functions [38,39]) that are typically associated with an innovation system. Although the TIS framework is typically used to analyze innovation processes in the context of specific technologies [44,45], it can also be applied to broader knowledge fields associated with a set of related technologies [39].
In this study, we used the TIS framework as a conceptual lens to analyze and compare the barriers to and drivers for CE-related innovation activities in wood-based and plasticbased industries. In terms of actors, technologies, and practices, both industries are associated with quite diverse CE-related innovation activities. However, we view the innovation activities within each industry as the outcome of a wider innovation system that is connected by a shared sector logic [46]. Accordingly, we conducted a literature review on empirical case studies from the body of CE-related literature, where we focused on the wood-based and plastic-based CE innovation systems (WCEIS and PCEIS, respectively) and addressed the following research question: How do the wood-based and plastic-based CE innovation systems differ in terms of their logic, drivers, and barriers?
The aim of this comparison is twofold. On the one hand, we aim to identify potential differences between the two innovation systems in the field of CE as regards the understanding and mentality of actors with different industrial origins. This can be helpful for an adaptation process of CE business strategies and further to develop alliances under consideration of such differences. Second, this approach will contribute to the discourse of CE as a concept by emphasizing different industrial origins and building knowledge for scholars or policymakers.

The TIS framework
In their seminal paper, Carlson and Stankiewicz [38, p.111] defined a technological innovation system as "a [dynamic] network of agents interacting in a specific economic/industrial area under a particular institutional infrastructure or set of infrastructures and involved in the generation, diffusion, and utilization of technology." Although several alternative definitions have been proposed since then [39,40], the central idea has remained the same: The development and diffusion of technology is driven by actor networks that are embedded in specific institutional structures. Technology, in turn, is understood quite broadly "in terms of knowledge, product or both" [38, p.408].
Institutional structures are regarded as imposing established rules that both constrain and enable innovation activities within an innovation system [43]. While three different types of institutions can be distinguished (i.e., regulative, normative, and cultural-cognitive institutions [47]), our study places a particular emphasis on cultural-cognitive institutions, which refer to the dominant logic shared by actors within an innovation system. This emphasis was chosen for two main reasons: First, both regulative and normative elements are expected to be quite diverse in the context of CE-related innovation systems. While regulative institutions (e.g., regulations, laws) are strongly country-specific, normative institutions (e.g., technical standards and norms) can differ greatly depending on the specific innovation. Second, green political economy concepts such as CE are inherently ambivalent and allow for different interpretations [48][49][50][51]. CE-related innovation activities are thus expected to depend strongly on the industry-specific interpretations of the CE concept.
In the innovation literature, the important roles of industry-specific logic in shaping innovation activities have been widely acknowledged [52,53]. Nelson and Winter [54] observed that the innovation activities performed by companies in the same industry tend to be shaped by common search heuristics, which the authors subsumed under the term "technological regime". As an example, they referred to the DC-3 aircraft regime, which defined the innovation in aircraft design for multiple decades. Malerba [46] applied this regime concept in the context of sectoral innovation systems, emphasizing how they facilitated or inhibited innovation activities. This author further stressed the fact that regimes may change over time and can either create conditions that are conducive to the entry of new innovators or rather rigid structures that prevent new innovators from entering the market [46]. Adhering to the regime concept, we expect that the underlying logic of CErelated innovation activities differs between the wood-based and plastic-based innovation systems.
While the early TIS literature placed a primary focus on the structural aspects of an innovation system (in terms of actors, knowledge networks, and institutions), scholars have taken dynamic perspectives in more recent work, emphasizing the processes that occur within innovation systems. Along these lines, the concept of so-called system functions was introduced [38,39]. System functions represent the key processes that have to take place for an innovation system to perform effectively. Analyzing these functions thus goes beyond merely describing the system structure and allows scholars to more clearly understand the deficiencies and weaknesses [45]. System functions, however, are closely interwoven with structural elements, and an innovation system is formed as a result of this interaction [55]. Whereas different lists of system functions have been presented in the literature, our analysis is based on the list provided by Bergek and colleagues [39,43], which is presented in Table 1.

Material and Methods
According to Snyder (2019), conducting a literature review proceeds in four phases: 1) designing the review, 2) screening the body of literature, 3) applying data analysis procedures, and 4) writing the review. These four phases are described below. Table 1 Adopted TIS functions for the present paper according to Bergek [43] Function Explanation of the system functions for the present paper (F1) Knowledge development and diffusion "Processes that result in a broadening and deepening of the knowledge base of a TIS, sharing of knowledge between actors within the system and new combinations of knowledge as a result of this." (F2) Entrepreneurial Experimentation "Uncertainty reduction through trial-and-error experimentation with new technologies, applications and strategies." (F3) Influence on the direction of search "Mechanisms that influence in what direction firms and other actors look for new opportunities and to what problems and solutions they apply their resources." (F4) Resource mobilization "Acquisition of different types of resources that are needed for innovation to occur, most notably financial resources (capital), human resources (competence and manpower) and complementary assets (e.g., infrastructure)." (F5) Market formation "Creation of a space or an arena in which goods and services can be exchanged (…) includes subprocesses such as articulation of demand and preferences, product positioning, standard-setting and development of rules of exchange." (F6) Legitimation "Process of achieving regulative, normative and cognitive legitimacy (…) in the eyes of relevant stakeholders." (F7) Development of positive externalities "Creation of system-level utilities (or resources), which are available also to system actors that did not contribute to building them up."

Design and screening of the literature
The scientific database Scopus was used to identify relevant scientific literature. This phase (2) of the literature review was conducted in January 2021, and retrieval was limited to articles in English.
The screening was carried out to identify relevant empirical case studies published between 2015 and the end of 2020, in which information from or about companies (at least one central company) was examined. This enabled us to gain insights into the current CE discourse as well as the change in the CE discourse over time between the time the first CE action plan was published in 2015 [5] and the newest CE action plan in 2020 [6] in the WCEIS and PCEIS. To identify relevant literature and to address the research question, criteria for the keyword search ( Fig. 1) were developed.
By using Boolean operators (e.g., TITLE-ABS-KEY, AND, OR) and adding wildcard characters (*), which allowed the use of different word expressions to configure the query, the search string (left side of Fig. 1) was created that was used for the present review. The screening process step resulted in the identification of 597 potentially relevant articles in Scopus. Abstract screening and the application of eligibility criteria involved the following steps ( Fig. 1): 1) The term CE and/or its synonyms (e.g., circular bio*economy) had to be relevant in the context of the work. In addition, strategies related to "cascading use" are similar to CE in that they both address resource efficiency and the promotion of resource circulation; however, CU is often conflated with downcycling rather than with recycling [56]. Mair and Stern [57] argued that both concepts share similar end-of-life practices and, therefore, can be applied equally in both bio-based and non-biobased economies, although CU has been used more frequently in the context of renewable resources, while CE refers to all kind of resources. Campbell-Johnston et al. [58] proposed using CU as an overarching, more fundamental concept that goes beyond biological nutrient cycling; therefore, both concepts were considered in this review. 2) To address the WCEIS or PCEIS, several synonyms were used to restrict the screening to both innovation systems and centrally related industries (Fig. 1). Some industries in which both innovation systems are used were not included in the search string (e.g., con-

Fig. 1
Literature screening process used for this study struction), since the literature on cases where plastic or wood is mentioned as a central keyword in those industries was already identified using the selected search terms.
3) The screening was limited to the identification of empirical case studies that included at least one company in their analysis. For this purpose, keywords referring to business activities were used (Fig. 1).
After the abstract screening was performed, a set of 137 documents remained for which a clear link to the WCEIS and PCEIS could be identified. After performing full text screening (45) and screening the references used, a final set to 51 documents remained for which both inclusion criteria (right side of Fig. 1), the industrial origin and the company focus, could be met.

Analysis procedure
The analysis was conducted using MAXQDA 6 software, which supported the qualitative and quantitative coding procedure of the prevailing literature. Because we wanted to go beyond specific regions and to include CE innovation activities carried out along the whole value chain, we decided to use the TIS framework as a conceptual lens. While TISs are typically delineated by specific technologies [48], we applied a broader system definition and regarded all CE-related innovations associated with an industrial origin as part of one innovation system. Although these innovations can be quite different in nature, we argue that they are still part of a common knowledge field that emerges around the CE concept.
Based on the understanding of cultural-cognitive institutions, we explored how different actor networks interpret CE and looked for overlaps or contradictions between or within the two innovation systems. In other words, we examined the "thought collectives" [59] or shared logic that were associated with the analyzed innovation activities. To describe these collectives/this logic, we mapped the occurrence of R-strategies [60,61] applying the 10R framework described by Potting et al. [60] (Fig. 2). This framework systematically describes the different CE strategies that exist and is based on the waste hierarchy of the waste framework directive [62], which was issued to minimize the environmental impacts by increasing the degree of circularity. However, we do not consider these strategies merely as options to approach a CE. According to Scott [47], behavior is not only shaped by rules and norms, "but also by common definitions of the situation and common shared strategies of action" [48, p.47]. Therefore, the distribution of R-strategies in the reviewed documents for the two innovation systems provides information about a predominant attitude towards a CE. Strategies that promote smart product use (e.g., product-as-a-service, product sharing) are considered to be more beneficial than the strategy of a product life extension, as the need for natural resources is decreased. With a product life extension, raw materials are still used, but are maintained longer in the economic system, and the lowest circularity level is related to way the material is recycled or energy is recovered from it [60].
As the cultural-cognitive institutions that were associated with the analyzed innovation activities refer to structural categories, the functional categories encompass several processes that help to identify differences and similarities between different innovation systems [43].
However, like Bauer et al. [44], we use the TIS system functions as a structuring aid for our coding strategy rather than as an explanatory framework. Due to its thematic overlap with other functions [55], we excluded the function "development of positive external economies". To compare both innovation systems, an inductive procedure was then used to identify the barriers to and drivers for a CE from the documents reviewed, which were then assigned to the corresponding system functions.
As the collection of barrier and driver studies for CE is steadily growing, the selection of the codes considered the key umbrella categories of barriers to and drivers for a CE that are predominantly mentioned in the CE-related literature, including cultural, organizational, social, regulatory, political, financial, legislative, legal, economic, market, technological, technical, barriers or drivers (e.g., [30,[63][64][65]).

Sample characteristics
The number of publications in the field of CE as well as for the present study has increased since 2015. In Fig. 3A, the publications per year are divided according to whether the documents relate to the WCEIS (n = 27) or the PCEIS (n = 24). Figure 3B illustrates the "CErelated" terms used in the reviewed documents. CE was mentioned the most frequently (n = 40), except in the years 2015 and 2016 when the term CU was mentioned more frequently. However, since 2017, the number of empirical studies using CE as a central term has increased exponentially. As of 2019, we observe an extension of CE towards circular bioeconomy (CBE), especially in the literature on the WCEIS (n = 8).
For both innovation systems, the most frequently used definition of a CE was taken from the EMF 7 [66], and the second-most frequently used definition, from the EC [5]. CE was defined less frequently in the publications referring to the WCEIS than in publications referring to the PCEIS. Occasionally, authors referred to the CE definitions of Kirchherr et al. [67] and Geissdoerfer et al. [7].
The literature on the WCEIS addresses an ambiguity of the CE concept (in relation to sustainability) [33,68], and issues related to CE implementation in the supply chain [69][70][71], the implementation of a circular business model [9,35], sectoral path dependencies [72], and the substitution of fossil raw materials [19]. The methodological approach is mainly characterized by qualitative studies (n = 22) and more rarely by quantitative approaches (n = 2) and company-related life cycle assessment (LCA) cases (n = 2). Studies on WCEIS were performed in Finland (n = 7), Brazil, Germany, and Italy (each n = 3).
Most of the companies with the WCEIS covered in the reviewed literature operate in the areas of product manufacturing (esp. furniture), wood construction, fiber-based packaging, and bioenergy. In terms of knowledge networks, the largest number of documents that cite WCEIS emerged from the Netherlands, then the UK and Finland (Appendix Fig A1.1). The Netherlands is a leading country in terms of knowledge production related to CE and has the highest circularity score (24.5%) in the Circularity Gap Report (2022).
The literature on PCEIS contains primarily information about the selection of appropriate methods (LCA, eco-design strategies, impact assessment) indicators to measure the environmental performance [73][74][75] of a process (e.g., upcycling process [76]) and uncertainties in assessing environmental impacts [77,78] in a CE context. The methodological approach is characterized by more company related LCA cases (n = 11) than for the WCEIS, and an even distribution of qualitative (n = 7) and quantitative approaches (n = 6) was observed. Studies on PCEIS emerge from Italy (n = 8), Brazil and Spain (each n = 3), and the UK and Belgium (each n = 2). Most of the covered companies in the documents are product manufacturer or refer to packaging and place a much higher focus on waste management (esp. related to recycling). In terms of knowledge networks, the most frequently cited country domains are Belgium, followed by Sweden, Brazil, and Spain (Appendix Fig. A1.2).
Both samples (wood and plastic) address comprehensibility issues of the CE concept and the lack of knowledge regarding how CE can be approached in the company, the supply chain, or in the broader industry [8,49,79,80]. Thus, approaches for an industrial symbiosis [81,82] or (reverse) supply chains [81,83] are central topical areas of investigation with regard to both innovation systems. However, the most common objective in all reviewed publications is the investigation of barriers to and drivers of CU [19,84,85] or CE in the WCEIS [86][87][88] and in the PCEIS [89][90][91].
A more specific role of actors and their partnerships and collaborations in networks will be shown below in relation to the innovation system functions. Figure 4 presents the prevalence of the different R-strategies in the two innovation systems. If we examine the 10R framework from right to left, the first division is "useful applications of the material" that contains the strategies "Recover" and "Recycle". The body of literature that addresses PCEIS has the most entries in this division, especially in the field of recycling. "Recover" plays a role in both innovation systems, and an even distribution between them is observed ("Wood": 6.9%; "Plastics": 7.0%). Energy recovery is a common practice in both innovation systems. In the reviewed documents, energy recovery from wood and plastics resources is mentioned as an end-of-life strategy [68,92,93], as an end-of-life assumption for environmental life cycle assessments [73,77], as well as a potential outcome to substitute fossil resources with renewable resources in energy generation [76]. "Recycle" is the most frequently used strategy in both innovation systems (WCEIS: 27.0%; PCEIS: 58.7%), as it is mentioned in all documents as a key R-strategy for a CE. With regard to the WCEIS, the focus is placed on improving efficiency and capacity [31]. With regard to the PCEIS, recycling is discussed particularly as a technical issue (e.g., quality improvement [92], recyclability [77], closed-loop recycling [94]) or as a more holistic strategy to achieve sustainable development [95].

Circular economy logics
Examining the middle of the 10R framework, we find the second division "extending the life of products and parts", which includes the strategies "Reuse", "Repair", "Refurbish", "Remanufacture", and "Repurpose". While these strategies are less frequently represented in the PCEIS, the WCEIS has the most entries in this section, especially for remanufacturing and reuse strategies. In the PCEIS, parts/components created with 3D printing [94] are often reused or the "Reuse" strategy is already considered in the product design phase [96,97] to avoid single-use plastics. However, the strategies described in this section are much less common in PCEIS, which is partly due to a lack of maintenance of functionality [79]. Highlighting maintenance, longevity, and lower quality requirements provide an opportunity for the reuse of the wood material [84]. In this context, some of the reviewed documents address circular business models developed to slow resource loops [32] by emphasizing longevity or the return of materials to a regional refurbishment/remanufacturing center [98] through a reverse supply chain [69].
On the left-hand side of the 10R framework, "smarter product use and manufacture" has the fewest entries found in the sample. "Refuse" and "Rethink" are strategies that are rarely mentioned in connection with either innovation system. These are mostly associated with degrowth scenarios [49], innovative solutions such as the transformation from product to a service [8], or new eco-design approaches [99].
Regarding the aspect of "Reduce", this strategy was only counted if it referred to resources and not to a reduction in emissions, energy, or environmental impact. Unlike the other two strategies, "Reduce" is well represented in both innovation systems: for PCEIS, it is the second most frequently used strategy (26.1%) in the entire hierarchy. This is due the fact that the use of recycled materials [77,92] is counted as an "Reduce" strategy. Furthermore, this strategy has often been mentioned in the context of energy or resource efficiency through the reduction in material inputs [9], in cooperative formations and side stream utilization [100], or the general reduction in packaging material [101] and waste [48,50]).

A functional perspective on barriers and drivers
To further explore how CE innovation systems differ in both innovation systems in terms of processes and their functionality and to see which factors identified in the studies under review support or hinder the respective industries in the transition to a CE, we used the system function approach described by Bergek [43]. Table 2 provides an overview of all identified barriers to and drivers for a CE, which were described in detail in the reviewed studies.
To increase readability, we present the distribution and relevance of barriers (A) to and drivers (B) for a CE in a combined illustration (Fig. 5) before presenting each of the six system functions in detail. Referring to Fig. 5 and the system functions, we then discuss the relevance of the categories for the respective innovation system.

Knowledge development and diffusion (Function 1)
The first function "knowledge development and diffusion" describes how knowledge is generated, diffused, and shared in the TIS [43].
The most frequently mentioned barrier in this function for the WCEIS is a lack of expertise [84,90], which includes the unavailability of technologies [102] and a lack of access to knowledge [19,83] which is a consequence of not sharing knowledge [85,88]. In contrast, the most frequently mentioned barriers to PCEIS refer to a lack of the application of methods (esp. LCA) to evaluate recycling targets [68], to model waste supply chains [81], or, in general, to a lack of data [78]. Another aspect mentioned was the difficulty of comparing assessments due to different system boundaries [75,77]. Another hindering factor that is mentioned with regard to both innovation systems is the lack of knowledge and awareness of customers [80,87] about recyclability [101] or appropriate collection and disposal [92].
Providing supportive training programs [83], expertise on methods and best practices [68,71,100], awareness raising [8,32], and knowledge sharing in the industry [70,101] is mentioned several times as a driver by both innovation systems. Knowledge exchange or collaboration between academia and industry [9,84], however, is almost exclusively mentioned with respect to the WCEIS. Digitization in the form of databases and platforms [80,83] is cited as a driver for CE in both innovation systems, a holistic life-cycle approach, and the combination of new tools and indicators in CE (e.g., eco-design and LCA) [77,99] are mentioned more frequently regarding PCEIS.

Entrepreneurial experimentation (Function 2)
In the second function, "entrepreneurial experimentation", processes of uncertainty reduction by "experimenting with new technologies, applications or strategies" [41, p. 8] are described. Therefore, this function includes phenomena that reduce or increase uncertainty in the experimental processes that support the transition to the CE.
Regarding the WCEIS, uncertainties in the conception/design phase have been identified. Extending the product lifespan conflicts with the idea that products become obsolete Table 2 Overview of barriers to and drivers for the CE within the system functions too quickly [100], an aspect that is problematic for remanufacturing process [69] as well. Concerning (packaging) design, an imbalance was observed between the actors in the innovation systems, as design is determined by the brand owners [50] or change agents [19], while the industry has to create this design. Regarding PCEIS, this function includes most of the inhibitory factors (Fig. 5), which is evident in the recognition of several conditions that increase uncertainty, such as postconsumer waste [76,100]. Such waste is more difficult to collect [96] and limits further options to process the plastic used due to its heterogeneous composition and impurities, leading to a downcycling process [79,80,92]. Other barriers contributing to this maladministration include a lack of recyclability [101,103] and multi-layered packaging that makes disassembly difficult [90] and prevents the preparation of the product for reusability [104].
For the WCEIS, however, function 2 made up the highest share of reviewed studies; most the reviewed papers are dedicated to business model innovation in the wood (esp. furniture) industries as well as to industrial symbiosis. Business models such as productservice-systems [98,105], sharing of goods or transport [102], servitization [9], and business models related to overall resource or energy efficiency [84,91], such as a product life extension [69] or sufficiency [32], are mentioned not only to save resources but also costs Barriers to a Circular Economy PlasƟc Wood <5% 10% 15% 20% 25% 30% 35% 40% 50% 60% 45% 55% 65%

Share (%) of studies within a literature sample (wood: n = 27; plasƟc: n = 24)
Note: The share of studies is ploƩed in relaƟon to the radius of the circles.  [8,49]. Furthermore, industrial symbiosis [32,100] is identified as a driver for companies, including the promotion of local exchange [70] and the utilization of side streams [33,82]. PCEIS show less business model innovation activity in the reviewed papers except for Gall et al. [95], where a socially oriented business model of waste collectors is described in detail. Furthermore, industrial symbiosis in terms of joint logistic solutions in the recycling chain [92] is addressed with respect to PCEIS as well.

Drivers for a Circular Economy
Both innovation systems display examples of (circular) design of a product in early life stages [74,79] that reduces the dependence on non-renewable resources [71]. Furthermore, technological advancements such as optimization processes in waste processing through automation [106], better separation technologies [19], or usage of bioenergy [85]) play a role.

Influence on the direction of search (Function 3)
The function "influence on the direction of search" includes mechanisms that both influence actors in the knowledge field and promote or hinder entry into the knowledge field [43].
Barriers preventing the transition to a CE can be found in both innovation systems; however, they were more common in the documents reviewed for the WCEIS. These barriers include a lack of political support in the form of missing incentives for companies [48,98] too little political power [19,68], and an unawareness of the potential benefits of CE/ CU [9,48], resulting in an unwillingness to change [49,106]) as well as a lack of communication with stakeholders along the value chain [19,92] and the lack of a shared vision [32,72]. The remaining barriers to a CE in this function occur in equal proportions in both innovation systems, namely, a lack of uniform legislation, characterized by the absence of a coherent international strategy [104,107], as well as too many regional differences and priorities [49,50]. Furthermore, a lack of engagement of the industry was observed due to a lack of green culture [84,90], while a lack of dialogue in the value chain [88], the challenge of operating profitably in a CE [35], a lack of understanding regarding how to achieve a payback [102], and a lack of customer engagement [69,89,101] as well as higher upfront costs [48] for, for example, recycled materials [92] affect both innovation systems as well.
A regulatory push in the form of climate change policies to reduce environmental impacts [86,107], as well as green image [71,108] and corporate vision for sustainability [9] are more prevalent driving factors for the WCEIS. Driving factors to intensify circularity for both innovation systems include supporting measures such as public procurement [72,73], inter-organizational initiatives [85] such as the EU Waste Directive [62], the CE Action Plans [5,6] or the UK Plastics Pact [101], reference systems [9], certifications [19,94]), environmental product declarations [99], or the promotion of international standards [80,107] that act as guidance for a CE. Another factor that affects both innovation systems is the corporate culture, namely, the motivation of top management [35] to reduce environmental impacts [8,90]. Factors that enhance a corporate culture for stakeholders in the innovation system are, for example, flat hierarchies [88] to create an innovative environment, transparent and comprehensible information (e.g., material flows) [79] on the positive impacts of a CE which foster trust [85], and a positive reputation [70,101].

Resource mobilization (Function 4)
The function "resource mobilization" includes the resources needed to enable innovation in a system. Bergek [41] divides this function into the categories "human resources", "financial resources", and "complementary assets". Subsequently, the barriers to and drivers for these are listed with minor adaptations.
A lack of management resources [83,85] is slightly more prominent in the reviewed WCEIS literature, namely, the high bureaucratic [68] and administrative burden [79,84] that must be accepted to fulfill environmental requirements. This burden often arises due to the small company size [90], lack of managerial skills to address sustainability-related issues [84]; these, in turn, result in a lack of marketing strategies [87] or the market research [9] needed to identify and successfully implement CE benefits in the company.
Barriers related to financial resources are mentioned least often in both innovation systems. However, this is not related to market price barriers, which are mentioned in function 5; high upfront costs are also mentioned in function 3 as a barrier that prevents actors from moving into TIS. In particular, barriers related to this function are due to the lack of funding opportunities [90,106] or financing options [88] to cover innovation costs [108] and transport costs [80,95] or to shift costs away from recyclers and enable reverse logistics [81].
Complementary assets were identified as the main barriers for both innovation systems in this function. In the PCEIS the resource availability is the most relevant barrier (i.e., regarding recycled materials [101,103], followed by low return rates of post-consumer waste [76,81], and associated contamination of waste [79,100]. Technical difficulties, such as the quality of recycled materials [76,79,92], disassembly [99], and aesthetics [101] related to resource provision, were identified as barriers as well. In addition, barriers related to the supply chain were identified that are relevant to both innovation systems, such as a fragmented supply chain [50,80], coverage of the entire value chain [48], miscoordination of a reverse supply chain (remanufacturing) [35,69], and a lack of supporting infrastructure [87] to support the reuse or recycling process [49].
Fewer drivers than barriers were identified overall for this function, and, in some respects, the drivers reflect the barriers. Financial resources, as with barriers, were scarcely mentioned; those that were found were financial incentives for companies, such as economic incentives as well as funding opportunities [95,103] to help companies remain competitive [68].
Regarding the WCEIS, complementary resources are a major driver, notably addressing the supply chain [33,35], fostering collaboration and dialogue between sectors [19,80], and shortening the supply chain (focus on regionality) [68,92]. Another aspect is reverse logistics [69], which enables the return of material to producers such as public and private institutions and retailers [81].

Market formation (Function 5)
Market formation covers various factors (e.g., price, demand, norms) that influence the exchange of goods and services [40]). Within this function, most barriers were identified for the WCEIS in this study, namely, in an absence of a market [19,107] as well as a lack of demand [33,84] for refurbished products (e.g., remanufactured furniture, [69], recycling wood [19] or upcycled products [83]), reflecting the lack of a common strategy in the wood-based industry [70,80]. Furthermore, competition for different applications was identified, such as alternative material use versus use for energy production [49,82,93] or cascading/recycling versus use for energy [19,50]. However, to make the recycling of wood viable and scalable, improved separation and processing of mixed wood waste is needed [84]. A lack of standards for side streams [108] and overall weak standards (greenwashing) [19] are more frequently mentioned with respect to the WCEIS, and standards for recycled or bio-based materials [92,107] and the export of waste [78,94] were identified with respect to both innovation systems.
Low prices for virgin materials [81,92] and a changing price/demand situation for recycled/remanufactured materials [69,79] are barriers that are found more frequently often in the PCEIS.
Barriers that persist for both innovation systems are consumer-related barriers, namely, a lack of interest, acceptance, or trust [50,107] and an unwillingness to pay higher prices for more sustainable products/services [19,98]. Furthermore, a lack of interaction was reported between the supply chain and the consumer [83,88] as well as regional differences [89,101], which lead to a competitive disadvantage.
On the contrary, the most prominent driver mentioned in the WCEIS-related literature is a perceived alternative consumer demand [9,68] from fossil-based to bio-based products [19] and also the willingness to spend more money on sustainable products [87,98].
Regarding PCEIS, barriers are flipped; therefore, a creation of a market for recycled materials [48,76] and consequently a price reduction of products made from these materials [83,89] can occur. Restrictions that include taxes to promote transformation for CE [50,74] and regulatory pressures such as a restriction on waste exports [79], extended producer responsibility (EPR) [89,103], or incineration bans on recyclable plastics [80] were also identified for PCEIS.
One driver identified for both innovation systems is the promotion of CE as a competitive advantage [9,102] which can be used to differentiate between products/services (e.g., through certifications) [50,80] from competitors or endorsements to remain competitive through, for example, tax cuts [73,90].

Legitimation (Function 6)
This function describes the adaptation process (e.g., norms, regulative or behavioral patterns) to the investigated phenomena, which may be reflected in an institutional change [43].
With regard to this function, the fewest codes were found in both innovation systems in the examined studies. According to Bergek [43], this could be because the function performs poorly as long as the benefits of using the central technology are not clear. Applying this to the present study, the most common barrier to both innovation systems is a lack of a common understanding of CE [33,79,83] and an ambiguity about the benefits of CE [48,72], CBE [9], and CU [19].
In the WCEIS, the lack of a social function [50,84] is more frequently identified in terms of creating health and employment [8,105]. In the PCEIS, a lack of green culture or society is more frequently identified due to a lack of lifestyle adoption [99], an increase in sorting/collection [81,96], and a skeptical attitude towards the use of waste as a resource [92]. Identifying the further development of a CE by clarifying the concept itself [9,33], promoting positive regional effects [49], and creating jobs [35,102] were often found as drivers for the WCEIS. Drivers for both innovation systems include increased environmental awareness and a shift of consumers toward more sustainable behaviors [48,106], accompanied by more education [83,103] to help the people to become more sustainable citizens.

Discussion
To support the transition to a CE, we examined the CE-related literature to find out how the emerging CE innovation systems (wood and plastics) differ in terms of their internal logic and to identify barriers to and drivers for a CE. The number of studies that address barriers to and drivers of SMEs [29,109], business models [63,110], or innovations for a CE [11] on a national or even an international level [30] is steadily growing. However, this study is one of the first ones to follow the suggestions to investigate different industrial origins [8,36] regarding the different understanding and implementation of CErelated innovation activities. Using a heuristic approach based on the TIS framework [38,39], we highlight the unique and shared aspects of the two innovation systems studied and clarify the different forms of logic from diverse industrial origins used by providing a nuanced allocation of barriers to and drivers for a CE. This not only strengthens the development of a coherent discourse of a CE from a conceptual point of view [68] but also provides starting points for the highly desirable establishment of cross-sectoral developments based on an in-depth understanding of the forms of logic applied in a single industry [9,32,33]. Fig. 4, we show the institutional contexts (forms of inner logic) of the TISs in both innovation systems differ, but also identify similarities in terms of the most common strategies (esp. recycle, reduce). This finding confirms the statements of Mair and Stern [57] that both bio-based and nonbio-based industries utilize similar end-of-life strategies. Furthermore, for both innovation systems, higher R-strategies were scarce, because profitability through CE is hard to achieve [89,104], and little experimentation is occurring with natural or bio-based materials [9,92]. The least common R-strategies (esp. refuse, rethink) that are associated to, for example, degrowth scenarios [49], innovative solutions such as the transformation from a product to services [8], or new eco-design approaches [99] seem to be unpopular in both innovation systems. Overall, driving factors are mentioned more frequently with respect to the WCEIS than to the PCEIS (Fig. 5), suggesting that more CErelated activities occur in the WCEIS. The R-strategies may again provide the rationale for this (Fig. 4). Even though recycling is the most important strategy in both innovation systems, the strategies differ, especially in the division in the middle (especially reuse, repair and remanufacture). Prieto-Sandoval et al. [111] developed a knowledge map of CE and outlined the concept, principles, and determinants of a CE. Still, an examination of the innovation systems addressed in this study, with its forms of inner logic and functions, indicates which strategies might support or hinder this institutional change. The function "legitimation", which describes a process of adaptation (e.g., regulatory or behavioral patterns), however, is assigned the fewest codes, which could indicate the absence of institutional change [43] (e.g., in this case, a shift of the R-strategies (Fig. 4) to the left-hand side). Figure 5 shows that different priorities are defined for the WCEIS and PCEIS, illustrating the barriers to and drivers for a CE embedded in the six system functions. One of the main drivers for the WCEIS is "business model innovation", which is located in the entrepreneurial experimentation function, which implies a willingness to experiment with business models as product-service systems [98,105], product sharing [102], or product life extension examples [32,69] that coincide with the R-strategies shown in the division in the middle (Fig. 5). According to Pieroni et al. [112], the variability among the product characteristics that occur in a sector influences the development of circular business model patterns. The advancement of CE business model patterns [110], and especially in manufacturing sectors (e.g., furniture [112] textiles [113]), is clearly apparent. A higher prevalence of business model innovation in the WCEIS could be due to the need to overcome barriers, such as a multitude of competing applications [19,50] or the absence of a market [19,107], by finding innovative solutions to respond to consumer needs [9,68]. The characteristics of the WCEIS (e.g., longevity, renewable energy) may also enable a greater diversity of services to be offered than those of the PCEIS, which is mainly driven by the trend to close the loop [94] (especially through recycling) rather than by developing more beneficial R-strategies. One of the main barriers to the WCEIS is the absence of a market for CE-related products, but the increasing demand and further refinement of the CE concept is perceived as an important driver in that innovation system as well.

Barriers and drivers in the plastic-based circular economy innovation system The main
barriers to the PCEIS relate to (technical) conditions, such as difficulties associated with the collection process, achieving suitable quality, or the processability (e.g., recyclability, reusability) of plastics (e.g., [80,92,103]), which hinder experimentation in that innovation system. Our findings are thus in line with those of Siltaloppi and Jähi [25], who found that barriers to establishing a sustainable plastic value chain were related to three main conundrums: the limited production of sustainable materials, the lack of users and demand for sustainable plastics, and the lack of an economic logic for recycling development. On the other hand, the main drivers for the PCEIS help to address these challenges through technological advancement, which, in turn, may explain the strong emphasis on recycling and the feasibility of using recycled materials (reduce) shown in Fig. 4. The creation of demand and markets for recycled materials is a finding that also agrees with those of Siltaloppi and Jähi [25]. In contrast, these technical conditions are less of a concern in the WCEIS, where competition for the right application is more relevant [49,85]; still, these barriers are strongly associated with uncertainties in the conception/design phase with regard to both innovation systems [69,100]. Kirchherr et al. [30] found that the main barriers in the EU are cultural, including the lack in consumer awareness and hesitant company culture, issues that are amplified by the high upfront costs and low virgin material prices. The outcome of the present study supports this statement due to the focus placed on the industrial origins of the materials; thus, high costs and low virgin material prices can be assigned to the PCEIS, and an absence of a market can be assigned to the WCEIS. A hesitant corporate culture applies to both industries, although not to a high extent, as does a lack of consumer awareness [8,48]. However, raising awareness and building knowledge through employee training programs [84,103] and consumer awareness campaigns [83] are seen as important drivers, particularly in the WCEIS. Some progress has been made in the previous years, and a green corporate culture is being mentioned more frequently as a driving factor. Finally, an interesting, shared factor is the strong emphasis on similar topical areas for the WCEIS (esp. business model innovation, sharing knowledge, and expertise), for the PCEIS (esp. digitalization, (eco-)design, (LCA-)tools/ indicators) and the ways both are encouraged-especially by forming strategic alliances-to collaborate in the supply chain. In forest-based industries, the use of residues by and the exchange of materials between industries are firmly established practices [33,114]. CU can increase the efficiency of wood utilization; however, the barriers to CU are consistent with those identified in the results of the present study, although few documents cited a relationship between CU and CE [57]. Actors in both the WCEIS and PCEIS face similar challenges, namely, the need to ensure the quality and availability of recovered products and renewable energy targets, the need to gain the commitment from policymakers, and the lack of a market for secondary materials [84,114]. Overall, the function "market formation" is a broad function for both barriers and drivers, as it has been assigned the most codes in this study.

Insights gained by considering industrial differences
Innovations are crucial in the transition to a CE [11,111]. Suchek et al. [11] pursued a comparable objective: to compare or measure the progress of the transition to a CE. Thus, they presented a slightly different approach by identifying seven innovation clusters (e.g., strategic alliances, business model innovation, biological cycle, technology, or waste management) in a CE. However, the industrial origin within the clusters was not considered, which means that the documents reviewed in this study are found in several clusters. Lastly, a growing number of studies that are being carried out to identify future pathways for a (circular) bioeconomy [9,50,78], as well as initiatives and policies that encourage that transition [68,107], are especially prevalent in the WCEIS. A review of the overall body of CE literature reveals an increase in the number of publications on CBE, although it is still unclear whether CBE represents an overlap between CE and BE, whether the BE is a part of the CE, or whether CBE represents more than BE and CE alone [66,[115][116][117]. Scholars continue to criticize the fact that both concepts are resource-centered and leave social aspects mostly unconsidered. Likewise, degrowth concepts still seem to be neglected [117].

Limitations and Conclusion
This literature review study is subject to certain limitations which might be overcome in future studies: First, the temporal limitations (years 2015-2020) set certain boundaries in this study. The literature on CE is currently growing exponentially. Our screening in Scopus using the search term circular *econom* revealed 156 articles published in 2015, but a total of 4901 articles published in 2021. Our literature review covered the period through 2020 therefore, we focused on the time period between the first [5] and second [6] CE action plan. The body of literature from 2021 and onwards was not included in the analysis. However, we have compared selected literature from 2021 and 2022 with the results of the present analysis to illustrate the point that the TISs for both innovation systems exhibit similar topical scopes after 2020.
Second, we based our analysis on secondary data presented in the empirical studies that formed our sample. Therefore, the results may be subject to certain biases, such as an overrepresentation of topics that are relevant for research and publication, whereas topics that are less relevant might be underrepresented. For example, the industry perspective chosen in this study could be responsible for the fact that the issue of microplastics is hardly addressed in the publications used. Furthermore, a certain selection bias is inevitable in a review to a certain extent due to the limitation of selected keywords. To counteract this, an iterative procedure was used after the full paper screening to include other papers considered important for the review (Fig. 1). Collecting primary data gained from those two industries in the future might allow a more detailed validation to be made; however, with the boundaries chosen, it was feasible to achieve conclusive results.
Third, since we compared knowledge fields and not a single technology in the TISs, the formation of a whole system structure for each TIS is beyond the scope of this study, as the systems encompass a multitude of actors and networks that cannot be covered by a literature review alone. Therefore, we used the system functions as a structural aid [44] for the analytical comparison of the innovation systems.
The national, sectoral, and cultural environments influence decisions made about CE activities [36]. To reach a specific point, you may need to take different routes, depending on where you start. This is also the case when different industrial origins are examined to study the progress in the transition to a CE. In this case, the different starting points can be attributed to the different raw material characteristics (renewability, processability, repairability) as well as the distinct industry structures (company sizes, business model diversity, power distribution, cooperation). Those differences may be understood as grounds upon which the distinct sectoral cultures are formed; thereafter, these manifest in innovation system functions [38,39]. As shown in Fig. 5, the respective starting points from which the two innovations systems begin to transition to a CE are quite different. The barriers dominate in the analysis of the PCEIS, but drivers dominate in the analysis of the WCEIS. Figure 4, however, illustrates considerable similarities between the two innovation systems. The rather low amount of consideration given the "refuse" and "rethink" approaches can be attributed to the radical nature of these approaches [118], raising questions about the respective industry as such. In contrast, "reduce" seems to be an acceptable strategy in both cases, since this implies an incremental process of sector reconfiguration. In the backend of the life cycle of the innovation systems, a similar tendency to use recycling and recovery strategies can be seen. The smaller relative share of recycling strategies (i.e., for wood-based products) may result from the higher share of reuse and remanufacturing strategies. Reuse and remanufacturing are more relevant for wood-based products, which is not surprising, considering such important material characteristics as easy processability. In conjunction with the growing literature on CBE, an appropriate extension of R-strategies that may become more important in the future is a "replace" strategy, 8 particularly in the context of substituting fossil-based materials with bio-based materials. Those observations may reflect the different perception of a CE in each of the innovation systems. Whereas CE is primarily seen as a mandatory solution to improve the environmental performance in the case of plastics, it is also considered as an opportunity to improve the economic performance in the case of wood. Therefore, the study results suggest that the technological regime [46] of the WCEIS seems to be more accessible to innovators than the PCEIS. While the TIS approach employed in this study offers a perspective on current drivers and barriers in the respective innovation system, it could also be used as an instrument to compare future changes in the field of CE. Finally, this work strengthens the link between a bio-based (WCEIS) and a fossil-based (PCEIS) innovation system, which may be a step towards a CE, as knowledge and awareness about the different understandings and Funding Open access funding provided by University of Graz. Open access funding was provided by University of Graz. The authors gratefully acknowledge funding of the project Start Circles (Project Nr: 199) co-financed by the Cooperation Programme Interreg V-A Slovenia Austria from the European Regional Development Fund.

Conflict of Interest Statement
The authors have no competing interests to declare that are relevant to the content of this article.

Competing interests
The authors declare no competing interests. Performed with the R package"biblioshiny" [119]