The technocentric consensus: a discourse network analysis of the European circular economy debate

ABSTRACT

How do different framings of the circular economy structure relations between key European policy actors? The circular economy is emerging as a dominant environmental governance framework, but multiple interpretations of what it entails and how it is to be implemented abound. This article investigates the way policy actors in the European Union frame the circular economy, and how certain framings facilitate coalition-building. A Discourse Network Analysis, which combines content and network analysis, allows us to draw connections between frames and actors. Building on an existing typology of circular economy discourses, we combine inductive and deductive coding of selected policy documents from policy actors (EU institutions, business associations, think tanks, trade unions, NGOs). This content analysis is visualized as a discourse network, which disaggregates discourses into their subcomponents and connects them to policy actors. The strongest discourse coalition emerges around technocentric framings of circular economy. In particular, we identify production, innovation and waste management as central sub-components which strengthen the congruence of this coalition. More transformative frames advanced by environmental organizations struggle to break the technocentric consensus. We argue that the current framing of a technocentric circular economy is unlikely to change unsustainable patterns of production and consumption.

KEYWORDS:

• Framing
• discourse coalitions
• circular economy
• social network analysis
• content analysis

Introduction

European societies are facing a multiplicity of challenges, such as global warming, loss of biodiversity and deepening social inequalities, which has intensified the search for alternative economic models. To address these challenges, the concept of the ‘circular economy’ (CE) has gained traction among all sectors of society, and more recently, it has been adopted as one of the cornerstones of the Von der Leyen Commission’s European Green Deal project (COM/2020/98).¹ Broadly, the concept entails the closing of input and output cycles in the economy, thereby making production and consumption systems ‘regenerative and restorative’ (Ellen MacArthur Foundation, 2015, p. 23). In doing so, it aims to solve problems of environmental degradation, social inequality and resource scarcity, while delivering economic growth to its economies and businesses (COM/2014/398). Hopes are pinned on CE as a ‘magic bullet’ that can target multiple challenges simultaneously (Calisto Friant et al., 2021, p. 377). However, many questions remain unanswered regarding its assumptions, implementation, and acceptance by different actors (Hobson, 2021).

A concern common to scholars studying definitions of the CE is its inherent ambiguity (Geissdoerfer et al., 2017; Korhonen et al., 2018). An overly broad application of the concept runs the risk of green-washing (Calisto Friant et al., 2020, p. 2), while fragmentation and contestation of the concept due to multiple interpretations can lead to deadlock and collapse (Kirchherr et al., 2023, p. 2). Acknowledging the multiplicity of definitions and narratives surrounding the CE, scholars have called for more research into the connections between actors and narratives – who influences the development of CE agendas, which narratives prevail, and with what effect on relations between actors (Palm et al., 2022, p. 381; Hobson, 2021, p. 174)? Calisto Friant et al. (2020, p. 14) argue that we need more analysis of CE discourses from public and private actors, and we need to test existing frameworks to better understand the relative strengths and weaknesses of different CE discourses. Such work has been carried out on the European Commission through a discourse analysis of official CE policy documents (Calisto Friant et al., 2021). But there are no studies of the CE discourses circulating in the broader, Brussels-based European Union policy environment, in which the Commission is located and to which it is responsive and dependent upon for policy expertise and relevance (Georgakakis & Rowell, 2013).

This article takes a step towards addressing the lack of detailed empirical studies on how policy actors relate to different kinds of CE discourses, as well as how discourses facilitate coalition-building between various groups of stakeholders to advance disparate notions of CE. Drawing on a theoretical tradition that allows us to analyze connections between discourses, policy coalitions, and frames (Hajer, 1995; Leifeld, 2017), the research question guiding our investigation is as follows: How do different framings of the circular economy structure relations between key European policy actors? Our analysis is informed by an established typology of CE discourses (Calisto Friant et al., 2020), but in this study, we apply a greater level of granularity that lets us reveal the specific strengths and weaknesses of the constituent parts of CE discourses. By broadening our analysis to consider not only the European Commission, but also the environment of business associations, think tanks, trade unions, and non-governmental organizations (NGOs) that make themselves heard in CE policy debates, we aim to discover how different discourses map on to the landscape of stakeholders, and how the discourses establish ideological coherence and facilitate the formation of coalitions. Thus, our study addresses both how actors frame the CE and which discourse networks (Leifeld, 2017) these acts bring about. Our analysis finds that business interests in the EU have succeeded in establishing a technocentric storyline that is shared by a powerful discourse coalition including key think tanks and the European Commission, and they are therefore effective in influencing policy. Surprisingly, we also find that coalitions aiming for a more radical CE policy approach fail to achieve similar levels of congruence on what a more radical CE model entails. Storylines and discourse networks thus matter for the policy direction and implementation of CE in the EU.

The article is structured as follows: The literature review will focus on recent scholarship that has advanced a discursive analysis of the circular economy, bringing a range of different framings into play and asking questions about their effect. We then specify how theories of discourse networks and discourse coalitions can help us take the next steps towards understanding the way CE framings intervene in relations between policy actors. In the methods section, Discourse Network Analysis is described together with the coding procedure and criteria for the sampling of policy papers and operationalization of the theoretical framework. The analysis presents an affiliation network that visualizes connections between CE framings and policy actors, on which we draw broader inferences on the underlying structure of CE policy debates in Europe. Before concluding with some short remarks and directions for future research, we discuss the implications of these results for the legitimacy of the CE as a sustainability vision and environmental governance framework.

Closed-loop policies and circular economy discourses in EU policy debates

The principle of circularity is not new in EU environmental policy making. Closed-loop inspired policy measures have been adopted in waves since the 1970′s, where resource waste was first identified as an environmental issue which required integrated regional waste management systems and recycling initiatives, starting with the Environment Action Program (EAP) in 1973 (Fitch-Roy et al., 2020, p. 988). The concept of the CE won acceptance at the European Resource Efficiency Platform in 2012, which pushed CE arguments onto the primary economic agenda (Leipold, 2021, pp. 1053–1054). Early CE policy packages were seen as too ambitious and unrealistic by industry stakeholders, which caused a reevaluation of the CE agenda with the advent of the Juncker Commission (Leipold, 2021, pp. 1049–1050).² The Ellen MacArthur Foundation has played an instrumental role in this reevaluation by shaping CE policies to be more in line with the updated, business-friendly Commission priorities (Fitch-Roy et al., 2020; Hobson, 2021; Palm et al., 2022). As a consequence, businesses, originally one of the primary targets of environmental regulation, are now the primary change agents, making environmental protection a ‘by-product’ of sound business practices and dismissing the need for new regulation (Leipold, 2021, p. 1051).

One example of this is the European Strategy for Plastics in the Circular Economy, which constitutes the EU’s approach to governance of plastics as a policy issue and the first policy framework to regulate a material-specific life cycle according to CE principles (Palm et al., 2022, p. 366). Although the aim of the strategy overall was to curb plastic littering, targets were only set for recycling and not production reduction (Leipold, 2021, p. 1059; Mah, 2021, p. 13; Palm et al., 2022). This is in line with the actions taken more broadly through the Commission’s 2015–2020 CE policy package, which has mostly aimed at improving the recycling of different waste sources, and has shied away from the more difficult end of the waste hierarchy: refusion and reduction of waste (Calisto Friant et al., 2021, p. 345).

Taking in these developments broadly, several authors have criticized the EU’s CE policy agenda for perpetuating earlier ecological modernization narratives, in which businesses as the central actors are incentivized into producing sustainable alternatives for society through lighter touch regulatory measures that tend to prioritize economic considerations over environmental or social ones (Fitch-Roy et al., 2020; Hobson, 2021; Hobson & Lynch, 2016; Leipold, 2021). To succeed in transitioning towards a more sustainable future, these scholars agree that CE per definition should prioritize ecological and social aspects higher, which they perceive to be largely absent from implemented policies and white papers. This suggests there is a noticeable gap between much social scientific treatment of CE and how practitioners tend to approach the term. Knowledge about who draws on CE conceptions, in what ways and which concepts facilitate coalition-building remains scarce, which is echoed by other environmental policy scholars (Eckert, 2021).

To help us make sense of how different conceptions of the CE are connected to policy actors, we use a typology of CE discourses as suggested by Calisto Friant et al. (2020) as a starting point, but provide a greater level of granularity in our subsequent empirical analysis. By using the typology, we can more easily compare different conceptions and narratives which practitioners draw upon in their framings of the CE (Calisto Friant et al., 2020, p. 2). Additionally, since this typology builds upon and has been utilized by other scholars, it helps us compare the overall discourse emerging from these framings with previous scholars’ findings (Calisto Friant et al., 2021; Palm et al., 2022).

The aim of the typology is to better differentiate discourse in a field where the actual definition and conceptions of CE are unclear and inconsistent. The typology classifies circularity conceptions according to their position on ‘fundamental social, technological, political and ecological issues’ (Calisto Friant et al., 2020). The typology is based on a critical literature review of CE scholarship and previous classifications of environmental discourses similar to the CE. The authors identify what they see as five obstacles facing different strands of the CE literature they reviewed: (1) the relation between entropy on one hand, and growth, capitalism and decoupling on the other, (2) trade-offs between needs for retaining material value, energy demands and biodiversity, (3) systematic evaluations of the impacts of the CE, in particular with respects to rebound effects, (4) the implications transitioning to a CE has for questions of governance, social justice and cultural change, and (5) lack of inclusion of alternative circularity visions, which complements mainstream definitions (Calisto Friant et al., 2020, pp. 4–6). Based on the extent to which different CE definitions and conceptions take these challenges into account, they can be positioned inside the realm of four different CE discourses organized along two different axes.

The first axis builds on the distinction between circular society and circular economy discourses, where in the former not only resources are kept in loop but also wealth, technology, and power (Calisto Friant et al., 2020, pp. 10–11). Holistic (society) discourses thereby ‘integrate the social, ecological and political considerations of circularity’, while segmented (economy) discourses focus only on economic and technical aspects of CE. The second typological axis is based on attitude towards the key issue of green growth (Calisto Friant et al., 2020, p. 10). This could be ‘the most crucial element to transition discourses’ as the fundamental question is whether capitalism, supplemented by the CE, can decouple economic growth from environmental degradation. One can be either optimistic or skeptical of whether technology and innovation can overcome ecological challenges. This yields four different CE discourses: reformist (holistic/optimistic), technocentric (segmented/optimistic), fortress (segmented/skeptical), and transformational (holistic/skeptical).

Theory: discourse, frames, and storylines

In order to investigate how the European CE debate is structured, we engage in Discourse Network Analysis (DNA). DNA allows us to understand policy issues in the intersection between networked forms of policy coalitions and the discursive rationalities that govern what can be said and done (Leifeld & Haunss, 2012, p. 383). DNA is not a theory, but a method that can be informed by different theories – in our study, we are guided by Hajer’s theoretical contributions to ‘Argumentative Discourse Analysis’ (Hajer, 1995, 2002). Discourse is thus defined as ‘[..] an ensemble of ideas, concepts and categories through which meaning is given to social and physical phenomena, and which is produced and reproduced through an identifiable set of practices’ (Hajer & Versteeg, 2005, p. 175). The agents of interest seeking to shape policy debates are the ‘discourse coalitions’ which are actors sharing certain social constructions of particular phenomena (Hajer, 1995; Leifeld & Haunss, 2012, p. 384). This shaping is done through the act of ‘framing’, through which a solution and ways to reach it are constructed from a given definition or conception of a political issue. Discourse coalitions are defined by an alignment between members in terms of how they frame a given political situation. Additionally, discursive alignment tends to be correlated with more explicit political coordination (Leifeld, 2013). From this perspective, a coalition is dominant if the institutional practices and policy outcomes reflect the coalition’s framing of the political issue. DNA is a more recent approach to the study of policy discourse that leverages the strengths of both qualitative and quantitative modes of inquiry, and it is increasingly being applied to environmental issues (Leipold, 2019, p. 449, 457). This approach is in line with studies that emphasize the strategic practices of ‘discursive agents’ in policy debates (Simoens & Leipold, 2021, p. 822) and with previous studies on the use of narratives in CE policymaking (Calisto Friant et al., 2020; Palm et al., 2022).

According to Hajer (1995, pp. 54–57), actors seeking to influence environmental policy do so by constructing storylines that situate environmental problems in discursive and political structures. A storyline is defined as a ‘generative sort of narrative that allows actors to draw upon various discursive categories to give meaning to specific physical or social phenomena’ (Hajer, 1995, p. 56). Storylines are the expressions of discursive systems, which are deeper and more comprehensive orderings of ideas, concepts, and categories. When policy actors make sense of the political reality, they are (consciously and unconsciously) drawing from discursive systems and manifesting them in storylines. To go from the typology of CE discourses identified by Calisto Friant et al. (2020), which we described in the previous section, we require a greater level of granularity to allow policy actors to express aspects of the CE discourses through more specific storylines. We draw on the notion of ‘frame objects’ (Hasselbalch, 2019) to theorize this level of granularity. Frame objects are the concrete pieces of communication that ‘select some aspects of a perceived reality and make them more salient in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation and/or treatment recommendation for the item described’ (Entman, 1993, p. 52). Frame objects are the building blocks of storylines, which are the expression of discourses. For example, the technocentric CE discourse might be expressed in storylines about how the CE is an opportunity for corporate expansion and profit, which we can observe in discrete frame objects about technocentric innovation (need for new corporate technologies) or technocentric growth (CE investments will create positive returns).

In the contestation and communication of the various storylines of CE discourses by different policy actors, the structure of their relations becomes apparent. Actors express themselves to encourage others to show their support or opposition to policy ideas (Leifeld, 2013, p. 172), and the stories they tell are often directly learned from other actors or meant to convince them (Leifeld, 2017, p. 2). On this basis, DNA assumes that discursive alignments also reveal social or political alignments. To the policy practitioners, think tanks, and interest groups that we are observing, the CE is often defined in accordance with its implementation and the broader set of goals it should serve (Calisto Friant et al., 2020, p. 2). By engaging in DNA, we aim to capture these varying conceptions of the CE by observing the frame objects that are communicated by actors in policy documents, and assessing the discourse network that thus emerges from alignments of actors communicating similar frames. We expand on how to turn frames into networks in the methods section and the beginning of the analysis. The approach we outline here allows us to address the research question on how framings structure relations among European actors discussing the CE.

Methods: discourse network analysis

Policy and position papers from twenty-seven organizations were selected in order to study framings and discourse networks in the European CE debates. A few criteria were used for the selection. First, organizations were picked based on relevance, judged by how frequently they appear in the reviewed academic articles and in the Commission’s CE policy documents. Second, frequent media activity suggests that actors proactively seek to influence the policy debate by moving public discourse. A search on the pan-European media network specialized in EU affairs, Euractiv.com, was therefore done to identify actors seeking broader public support for its views through print media communications. Finally, actors engaging the Commission through written responses to its 2020 CE Action Plan and on the Circular Economy Stakeholder Forum made up the final source of sampling.

We made keyword searches on each organization’s website in order to find their flagship publications on CE. These publications then had to present a broader idea of what the CE is and have an EU-wide scope, leading to further omissions. In total, our data selection left us with forty-three publications of varying length, which were selected for further analysis (see Appendix A).³ These publications were published between 2011 and 2023, starting with the Commission’s 2011 Roadmap to Resource Efficiency communication and ending at the time of writing. This time frame is both coherent with previous historical accounts on the emergence of CE policies (Fitch-Roy et al., 2020; Leipold, 2021) and captures the most recent reactions to the Commissions 2020 Circular Economy Action Plan (CEAP), which is the latest iteration of CE policy initiatives. The publications were produced by ten NGOs, fourteen interest groups, and three public sector bodies (including the European Commission). Table 1 provides an overview of the sampled organizations.

Table 1. Sample of organizations and number of policy documents.

Organization	Type	Size (N of employees)^4	N of sampled papers
European Commission	Public	Large	6
Ellen MacArthur Foundation	NGO	Large	3
Club of Rome	NGO	Medium	1
Institute for European Environmental Policy	NGO	Small	4
European Trade Union Institute	Interest Group	Large	1
Business Europe	Interest Group	Large	1
Circle Economy	NGO	Medium	3
European Environmental Bureau	NGO	Medium	2
European Social Partners	Interest Group	Large	1
Euratex	Interest Group	Large	1
EuroCommerce	Interest Group	Large	1
Zero Waste Europe	NGO	Small	2
European Committee of Manufacturers of Electrical Machines and Power Electronics	Interest Group	Large	1
European Waste Management Association	Interest Group	Large	1
European Recycling Industries’ Confederation	Interest Group	Large	2
Association of Cities and Regions for Sustainable Resource Management	Public	Large	1
RREUSE	NGO	Small	1
Plastics Europe	Interest Group	Large	1
Friends of the Earth Europe	NGO	Small	1
Forest Stewardship Council	NGO	Large	1
American Chamber of Commerce EU	Interest Group	Large	1
ECOS	NGO	Small	1
European Organization for Packaging and the Environment	Interest Group	Large	1
European Economic and Social Committee	Public	Large	1
European Steel Association	Interest Group	Large	2
European Construction Industry Federation	Interest Group	Large	1
Chemical Recycling Europe	Interest Group	Large	1

We operationalize ‘frame objects’ in the analysis as the expression of policy preferences or standpoints on the circular economy. We recovered these frames through an initial inductive round of coding, which was aimed at discovering the components of the circular economy discourses that were being discussed by actors. This initial inductive coding revealed the following seven frame objects: growth, innovation, production, jobs & training, society, consumption, and waste management. We then combined these inductive codes with the four-way typology of CE discourses described by Calisto Friant et al. (2020, p. 10): reformist, technocentric, transformational, and fortress. This left us with a matrix of 28 possible CE frame objects, each of which are expressed as storylines that activate different aspects of the broader CE discourses (see Appendix B for examples and detailed descriptions of frames).

The forty-three policy reports were then analyzed interpretively within the Discourse Network Analyzer software (Leifeld et al., 2019) according to our final coding framework. The DNA software combines content analysis and network analysis by creating graph objects that connect framings to organizations. By doing a content analysis of each report, we gradually built a picture of which frame objects were used by which organization and how frequently. Frame objects were identified as statements, either sentences or entire paragraphs, that we interpreted as consistent with one of our 28 possible codes. Multiple statements of the same frame increase the frequency count for that frame. The data is then exported from the DNA software in a two-mode network that shows all connections between all actors and all frames in the ‘affiliation network’ (Leifeld & Haunss, 2012, p. 390). This allows us to deploy network analytic methods to identify important actors, frames, connections, and so on. We can also quantify the degree of congruence between actors and the projected frames to discover which actors or frames are most closely aligned in the discourse network.

Analysis: storylines and discourse coalitions in the European CE debate

Figure 1 presents the overall discourse network for the CE policy debate by visualizing the affiliation between actors and concepts in a networked form. The network was created using network visualization packages in the R statistical programming environment and laid out according to a ‘stress majorization’ algorithm, which reduces edge overlap in the network, placing similar clusters of nodes close to each other (Gansner et al., 2005). Edges are drawn between the actors and the frames they communicate. Hence, actors that communicate similar sets of frames are placed closer to each other, and different frame users are placed more distantly. Actors are depicted as black circles and frame objects as gray ones. Edges with greater weight (meaning they are more frequently communicated) are depicted in darker shades and greater width. The size of the nodes is mapped to their degree centrality, which is the total amount of other nodes each node is connected to. We normalized edge weights according to average statement activity, which is an important step to avoid over-representing the actors that were more heavily sampled (Leifeld et al., 2019, p. 9).

Figure 1. Discourse network of the European circular economy debate.

To get a sense of the level of congruence in the network, we introduce statistics in Tables 2 and 3 on inter-node edge weights in the one-mode projections of the affiliation network. These statistics are derived from the two-mode network in Figure 1. Where the two-mode network depicts connections between actors and concepts, the statistics show connections between actors (on the basis of common use of concepts) or concepts (on the basis of common use by actors). The greater the edge weight, the more actors or concepts are aligned, and the more prominently they figure in the data.

Table 2. Normalized edge weights in the actor congruence network (top 20).

From	To	Weight
Ellen MacArthur Foundation	European Commission	36.39
Circle Economy	Institute for European Environmental Policy	13.68
Circle Economy	Ellen MacArthur Foundation	13.15
Circle Economy	European Commission	13.03
Ellen MacArthur Foundation	IEEP and Ellen Macarthur	9.93
European Commission	IEEP and Ellen Macarthur	9.88
Ellen MacArthur Foundation	Institute for European Environmental Policy	9.61
Club of Rome	Ellen MacArthur Foundation	9.60
Club of Rome	European Commission	8.52
European Commission	Institute for European Environmental Policy	8.24
European Social Partners	Institute for European Environmental Policy	8.09
Club of Rome	Institute for European Environmental Policy	7.87
Club of Rome	European Social Partners	7.80
Business Europe	European Commission	7.61
Circle Economy	Club of Rome	7.34
Circle Economy	IEEP and Ellen Macarthur	6.79
Ellen MacArthur Foundation	European Social Partners	6.64
Club of Rome	IEEP and Ellen Macarthur	6.32
Circle Economy	European Social Partners	6.18
European Commission	European Social Partners	6.05

Table 3. Normalized edge weights in the concept congruence network (top 20).

From	To	Weight
Innovation Technocentric	Waste Management Technocentric	41.28
Innovation Technocentric	Production Technocentric	39.83
Production Technocentric	Waste Management Technocentric	36.79
Growth Technocentric	Innovation Technocentric	29.96
Growth Technocentric	Waste Management Technocentric	27.44
Innovation Technocentric	Society Technocentric	25.82
Growth Technocentric	Production Technocentric	24.11
Society Technocentric	Waste Management Technocentric	23.27
Production Technocentric	Society Technocentric	23.19
Innovation Technocentric	Society Reformist	22.45
Production Technocentric	Society Reformist	20.55
Society Reformist	Waste Management Technocentric	19.61
Consumption Technocentric	Waste Management Technocentric	17.89
Consumption Technocentric	Innovation Technocentric	17.89
Consumption Technocentric	Production Technocentric	17.65
Growth Technocentric	Society Technocentric	17.34
Production Reformist	Society Reformist	16.97
Production Reformist	Production Technocentric	14.77
Growth Technocentric	Society Reformist	14.02
Innovation Reformist	Innovation Technocentric	13.94

Three storylines can be observed in the network (Figure 1): one consisting of technocentric frame objects which is communicated primarily by business lobby groups but also receiving attention from the Commission and the Ellen MacArthur Foundation in particular (right side of the diagram), a storyline based on reformist frame objects communicated by primarily NGOs and public institutions (left side of the diagram) and, finally, a less coherent transformational storyline which is sporadically voiced primarily by NGOs (bottom of the figure). Starting with the technocentric storyline, this is built on technocentric frame objects, in particular focusing on the role of production, innovation, growth, society and waste management in the CE (see Table 3). The coalition around these frame objects is structured by a vision of the CE that hinges on optimizing resource-use in industrial production and manufacturing and claiming waste as new input, in order to cut costs. Regulation (coded as the frame object ‘society’), in this vision, is both considered ‘red tape’ to the CE across the EU bloc, but also as an important prerequisite in order to make secondary materials competitive with virgin materials (e.g. EuroCommerce, 2020). Business Europe in particular has championed a market-driven approach to CE, highlighting the role of suppliers’ expertise in setting the rules for this new economic model:

If the Commission chooses to extend its scope, it will be crucial to clarify the exact definitions, options and thresholds it is considering to define the circularity criteria, and how this would all feed into the broader discussion on applying a life cycle approach to products and on making them more sustainable overall. As businesses need to secure the safety and performance of the products the Commission is targeting, business stakeholders should be involved in defining these criteria effectively (Business Europe, 2020, p. 5).

The technocentric discourse coalition arguably holds a strong grasp on the CE policy imagination due to strong coherence in their expression of storylines and limited intermingling with competing CE storylines. First, the business associations concentrate three quarters of their statements on four technocentric frame objects: production, society, waste management, and consumption. Limiting the storytelling to less rather than more components can be regarded as an advantage, insofar as it renders the CE actionable by boiling it down to a small number of interventions like optimizing production processes, creating a market for second-hand materials, institutionalizing circular product offerings, and clawing back the ensuing waste for new material input. Secondly, the business associations rarely draw upon reformist or transformational frame objects, keeping overlap with competing storylines to a minimum (Table 3). Discourse network scholars have argued that ideational congruence is closely associated with political coordination among policy actors (Leifeld, 2013). Looking at Table 3, we see that nine out of the ten most heavily weighted edge ties are between technocentric CE frame objects, suggesting that actors pushing these are the best aligned in this policy space. Given the closing of ranks around a bounded set of technocentric frame objects, we expect the business associations to be successful in pushing a CE storyline that is internally coherent.

The set of actors in the middle of the discourse network, which includes NGOs, public sector bodies, and interest groups, embraces both reformist and technocentric CE storylines. The Commission and Ellen MacArthur Foundation lean most heavily on technocentric frame objects, focusing in particular on the innovative potential of more circularity and how this will improve competitiveness in the Single Market (COM/2020/98). These actors are, however, brought closer to the reformist camp by insistence on more regulatory intervention in support of the CE, not just voluntary initiatives. The Commission’s proximity to the business coalition is hence not surprising according to the previously mentioned historical account (Leipold, 2021), but we see it takes a stronger regulatory tack. Drawing more heavily on reformist storylines we find the European Trade Union Institute (ETUI) and European Social Partners (ESP), who are both advancing a CE storyline concerned with adequate labor conditions and reskilling of the workforce, which is required in order to better handle secondary raw materials:

[..] it is important that the circularity of companies goes hand-in-hand with strategies to preserve employability, re-skilling, suitable health and safety safeguards and good working conditions. Collective bargaining, including by way of collective negotiations and agreements, are a useful tool in this respect, in line with national industrial relations systems (European Social Partners, 2021, p. 2).

In the perspective of the actors most committed to the reformist storyline, the CE must be driven by stricter regulation on production standards, market-making, and marketing of circular products. In addition to the ETUI and ESP, the core reformist actors are ECOS, the Association of Cities and Regions for Sustainable Resource Management (ACR+), RREUSE and the European Economic and Social Committee (EESC). The CE continues to be framed as an engine for domestic economic growth, but this objective is subservient to the overarching goal of preserving the environment and thus limiting the extraction of primary resources and the resulting pollution. Additionally, it seeks a holistic approach to the CE that makes space for other actors than businesses, such as municipalities and civil society which are often closer to the consumption- and waste management practices on the ground (ACR+, 2020; EESC, 2019).

The reformist storyline is built on three primary frame objects: production, society, and consumption. First, businesses and industry should not only aim to optimize production processes, but aim to have a net positive ecological footprint, for which they should also be held accountable (ECOS, 2019). Secondly, more stringent regulation is needed in order to reflect the true cost of primary resource extraction. This necessitates improved environmental regulation, surveillance of the secondary materials market, and fiscal policy which shifts taxes from labor to products (ECOS, 2019; EESC, 2019). Finally, reducing consumption is considered necessary in order to bring down absolute extraction levels (ECOS, 2019). To enable such sustainable consumption, repair options as well as product sharing models must be more readily available for consumers who are trapped in a linear consumption regime (RREUSE, 2022).

While a broader set of reformist frame objects enjoy support from the actors in the center, these are what mobilize the ‘core’ reformist coalition. This can be read off of Table 3, in which Production Reformist and Society Reformist features as the only reformist pair in the top twenty co-mentions, suggesting that ideational congruence in the reformist camp is built on a narrow set of frame objects. For instance, Jobs & Training is only integral to the CE narratives of the ESP and ETUI, receiving otherwise little attention by others in the reformist camp. Similarly, growth only receives modest attention in this core reformist grouping, suggesting that this frame object is more important to actors who wish to achieve consensus between the reformist and technocentric camp. In summary, the reformist discourse coalition achieves congruence through common reference to a small set of frame objects and remains divided on the remaining frame objects, arguably limiting the coalition’s capacity for political coordination.

In contrast to the technocentric and reformist coalition, no strong, coherent discourse coalition has emerged out of the broader transformational discourse due to, first, ideational incongruence between the actors, reflecting significant ambiguity in how the CE is framed as a form of socio-political reorganization, and, second, strong reliance on individual technocentric and reformist frame objects. The transformational policy actors (comprised mainly of environmental NGOs) broadly agree on the need to maintain consumption within planetary boundaries, implying that reduction in absolute material extraction is necessary for developed countries. They critically point out that an overt focus on enhancing productivity and substituting less sustainable materials in the end does not address growing material consumption within the EU:

To achieve a more sustainable economy, it is insufficient to only increase recycling and focus on (partial) improvements in the degree of circularity, but it is essential to also achieve absolute reductions in resource extraction and consumption, that is, to downsize the socioeconomic metabolism. (Zero Waste Europe, 2020)

However, the actors differ on whether increasing consumption constitutes a ‘cultural issue’ on the level of the individual (Circle Economy, 2021) or a structural issue which places responsibility with businesses and regulation in the EU’s internal market (EEB, 2020; Friends of the Earth Europe, 2016; IEEP, 2020). Additionally, what sets these storylines apart from more reformist ones is the emphasis not just on economic but social reorganization, in which more circularity is equated to a more socially just economic system. In these storylines, current economic models should not be retrofitted into circularity as this would only perpetuate current inequalities within and outside the EU (IEEP, 2022). Multi-stakeholder governance structures, quality-of-life indicators replacing traditional growth indicators (such as GDP), and a production system relying less on material extraction constitutes policy initiatives the EU should work towards to achieve a CE.

The congruence between these transformational frame objects is weak compared to the technocentric and reformist storylines, as indicated by their absence in the top 20 frame object co-mentions (Table 3). This is interesting as certain actors who draw on transformational frame objects are highly interconnected (Table 2), suggesting that these actors rely on frame objects from the other discourse coalitions to achieve this connectivity. This suggests that actors making reference to transformational frame objects are more disorganized in comparison to the other camps. The transformational storyline struggles to exist independently from the other storylines, making it less obvious what a policy agenda built on a transformational CE storyline looks like.

In summary, while the analysis corroborates previous scholars who find the CE policy debate dominated by a pro-business consensus (Calisto Friant et al., 2021; Fitch-Roy et al., 2020; Leipold, 2021; Mah, 2021), we go further by identifying which frame objects this consensus is built upon, which are absent, and how a more radical alternative is struggling to emerge. The technocentric discourse coalition has seemingly managed to close ranks around a few central frame objects while also keeping its reach into the other storylines limited, creating the conditions for a strong independent storyline. The technocentric storyline rests on policy commitments towards production, innovation, waste management and growth. Frame co-mentions of growth with waste management and innovation (Table 3) make it clear that the rationale of the technocentric consensus is to turn waste into a driver of business growth.

The core reformist coalition also focuses their attention on a limited number of frame objects: consumption, production, and society. This coalition aims to open up the CE to a broader set of objectives, such as curbing production and consumption and more actively involving civil society and labor in this economic reorganization. However, there is only limited acceptance of a broader reformist CE storyline from the Commission, especially when viewed in contrast to the strength of the technocentric consensus. Equally important are themes that remain largely absent inside these coalitions, such as jobs and training. Transitioning to a CE will increase the need for labor for maintenance and repair services and require upskilling or retraining many workers. Keeping this in mind, the marginal amount of attention the topic receives is surprising, alluding to the strength of the technocentric discourse coalition. Finally, we find no evidence of a coherent, transformational CE storyline. The actors communicating these frames are too scattered and do not seem to agree on a common telling of what a more radical, socio-political circular society entails or how it is reached.

Discussion

Early reviews of CE literatures found theoretical development to be fragmented due to the inherent ambiguity of the concept and its ability to appeal to many different actors (Calisto Friant et al., 2020; Geissdoerfer et al., 2017; Ghisellini et al., 2016; Kirchherr et al., 2017; Merli et al., 2018). In a more recent review, Kirchherr et al. (2023, p. 2) warned that these multiple interpretations of CE would lead to further fragmentation and contestation ultimately resulting in deadlock and collapse. In contrast, our analysis suggests that the fragmentation of CE discourses is still far from leading to a situation of deadlock – rather, fragmentation provides rich opportunities for political maneuvering and coalition-building in the current ecology of EU-level CE policy actors. However, some coalitions have advanced much further in consolidating their particular vision of CE into specific storylines, building stronger and larger discourse coalitions.

We found three distinct CE storylines in our analysis, each consisting of multiple frame objects, and each a manifestation of a broader CE discourse (Calisto Friant et al., 2020). The coalition built on technocentric framings has so far been most successful in this regard, succeeding in dominating the policy discourse that we mapped in our analysis. The storyline attracted many powerful actors, but especially secured ideological alignment between the European Commission, the Ellen MacArthur Foundation, and the vast majority of business interests. The transformational coalition is the most radically opposed to this technocentric consensus, but in comparison to the technocentric coalition, the storyline is not nearly as coherent, making the coalition more scattered and less able to speak with one voice. There are fewer policy actors communicating transformational frames, and when they do so, they tend to emphasize different frame objects – this was not the case with the technocentric coalition, where the storyline clearly emphasizes the same frame objects: growth, innovation, waste management, and production.

In-between these two coalitions we find the reformist coalition, which is better organized than the transformational group, but less organized than the technocentric one. The reformist coalition emphasizes the frame objects of society, consumption, and production, and jobs and training to a lesser extent. As such, reformist conceptions seem to bridge the transformative and technocentric coalitions, which in line with Palm et al. (2022, p. 379) suggests that reformist policy narratives in the CE can provide an opening to challenge prevailing technocentric assumptions. According to these authors, the incremental changes from policy implementation will most certainly fall short of the ambitious expectations that have been invested in the CE. This could provide the window of opportunity for more reformist narratives to become implemented in legislation. Perhaps surprisingly, in a European policy environment that increasingly worries about resource security and critical raw materials, storylines evoking a ‘fortress’ style CE discourse were not evident during the sampled period. However, we expect this to become a novel feature of the CE landscape in coming years.

The analysis contributes to the literature on CE in environmental policy in two ways: first, by disaggregating broader CE discourses into inductively generated sub-components, namely the frame objects and storylines, and second, by empirically investigating the presence and connections of CE discourse coalitions in the EU-level policy debate. Regarding the first contribution, the analysis shows that when CE discourses are manifested into specific frame objects and storylines, certain topics come to play more dominant roles in stabilizing the discourse coalitions, such as production for the technocentric camp and society for the reformist camp, while some topics are marginalized in both of the main camps, such as jobs and training.

Regarding the second contribution, the DNA approach complements the analysis of the European Commission’s CE communications carried out by Calisto Friant et al. (2021), building upon their previously developed discourse typology (Calisto Friant et al., 2020). According to these authors, communications by the Commission fall inside the reformist circular society discourse realm, while actual implementation has been in line with more technocentric CE discourses (Calisto Friant et al., 2021, p. 346). However, we find their discourse to be largely technocentric in comparison to other actors while also making it explicit that its position inside the technocentric coalition rests on framing the CE as issues of production, innovation, and waste management, while abstaining from discussions of jobs and consumption. The catchphrase ‘restorative by design’ (EMF, 2015; COM/2014/398) that is frequently mentioned in conjunction with cross-sectoral collaboration and industrial symbiosis is emblematic for the framing of circularity as a mission for ‘greener’ market offerings and less wasteful supply chains. We also go beyond merely analyzing communications from the Commission to providing a full picture of the Brussels-based ecology of think tanks, NGOs, interest groups, and other policy actors that engage actively in shaping the CE debate.

Business scholars have already identified the CE as an avenue for market-expansion, insofar as the CE is perceived as a means to create more value from already-paid-for materials (Esposito et al., 2017, p. 11). According to Mah (2021, pp. 122–123) the themes of production, innovation, and governance are particularly important to corporations, because these are focal points for further containment and capitalization on the CE. From this perspective, corporations engage proactively with environmental governance in order to mitigate reputational risk, attain power over sustainability governance, and justify a regulatory setting that accommodates profit-making and increased production (Dauvergne, 2018, p. 40). Leipold’s (2021, p. 1057) account on business pushback to the initial CE policy proposal under the Juncker Commission bears witness to such strategies. Simoens and Leipold (2021) have shown how this tends to produce incrementalism (as opposed to transformation) in organizations implementing CE policy. Focusing on themes of production and innovation also helps avoid the unpleasant question of reducing primary production through secondary production (recycling goods), which from a business perspective amounts to ‘market cannibalism’ (Hobson, 2021, p. 167).

Faced with the strength of the technocentric consensus and the discourse coalition that promotes it, the marginal positions of transformational framings are especially worrying for those who place higher hopes for stronger sustainability emerging out of the CE. Our analysis complements recent findings from Corvellec et al. (2021) and Hobson (2021), who bring together critiques of the CE, highlighting that current conceptions largely fail to recognize inherent limits to circularity, such as rebound effects. Our findings corroborate this, showing that transformational frames which do recognize the limits are on the periphery of the discourse network, which may lead to alternative models of the CE being overlooked. Kirchherr et al. (2023) argues there has been a growing agreement that a fundamental shift is necessary for the CE, but the lack of a transformational consensus in the definition of a CE may hinder its ability to affect policy making.

We were surprised to see how weakly aligned NGO contestation and alternative CE framings were. In other policy issues, environmental NGOs in Brussels have succeeded in forming strong and cohesive discourse coalitions that have been able to speak with one voice when asking for policy change or attention to issues (for an example in chemicals regulation, see Joachim & Locher, 2008). However, when it comes to the transformational CE, a cohesive storyline has not emerged. Friends of the Earth Europe and the European Environmental Bureau have taken more radical stances than the Institute for European Environmental Policy (IEEP) or Zero Waste Europe (who both lean more towards compromising with reformist and technocentric storylines – IEEP in several cases coauthoring reports with the Ellen MacArthur Foundation, see Valeche-Altinel et al., 2021). As a result of this, a coherent and strongly connected discourse coalition has not formed. We should not expect the direction of European CE policy to become more transformational as long as a coherent alternative discourse coalition is absent.

In the realm of broader EU environmental politics, scholars argue the CE policy agenda represents reframing and perpetuation of previous environmental policy rather than actual policy change (Fitch-Roy et al., 2020, p. 995). The CE agenda perpetuates discourses of ecological modernization, that is, the general conviction that economic growth can be reconciled with ecological sustainability through innovations and efficiency increases in production and consumption (Leipold, 2021). Eckert (2021, p. 191) also finds that current CE conceptualizations from both policymakers and industry players adhere relatively close to the status quo and therefore are limited in their transformative potential. They focus on the linear end of the waste hierarchy such as recycling and energy recovery, relying on technocentric solutions in their bio-economy vision. They corroborate Mah’s (2021) findings that industry players, particularly in the plastics sector, are determined on ‘future-proofing’ existing business practices, even as this risks tripling the amount of plastic waste circulating on land and in the oceans by 2040. In addition to supporting these findings above, our analysis can also suggest one of the reasons as to why alternative framings of the CE have thus far not been successful in reorienting policy away from the technocentric consensus: the alternative is simply not strongly enough realized even in the realm of discourse. Ultimately, the CE risks losing its sustainability credentials as an environmental governance framework given the discrepancies between the current light-touch approach to regulation and the urgent need to curb primary material production and consumption.

Conclusion

This article has investigated how circular economy (CE) debates structure relations among key actors in the EU by studying acts of framing in policy documents and publications. It has done so by drawing on Discourse Network Analysis, a methodology connecting actors by their common adherence to framings of the CE, which allows network relations to emerge from discursive alignments between actors. We disaggregated discourses into their sub-components across four ideal-typical conceptualizations of the CE (Calisto Friant et al., 2020). Two overarching findings emerged from the analysis. First, we find that a few central frame objects (technocentric framings of innovation, production, waste management, and reformist society framings) structure the two central coalitions, of which the technocentric coalition exhibits the strongest congruence in storylines. Here, the CE generally emerges as a market-driven approach to optimize currently inefficient production and waste-management systems through innovations and new business models, leaving unaddressed the challenges of decoupling, socio-cultural barriers to circularity, and societal inequality. Secondly, we find no evidence of a coherent transformational CE storyline, as actors drawing on frame objects within this broader discourse do so in an uncoordinated and scattered manner. A strong and coherent storyline of what transformational CE policy entails, backed by a cohesive discourse coalition, is an important prerequisite to actual political mobilization and influence.

While this corroborates those scholars indicating that current conceptualizations perpetuate ecological modernization discourses (Calisto Friant et al., 2021; Hobson & Lynch, 2016; Kirchherr et al., 2017), we do see reason for some slight optimism regarding the concept’s capacity to leverage more reformist framings for broader coalition-building. We see especially reformist society conceptions as an avenue for coalition-building as this frame bridges several actors, connecting critics and the core of the discourse network. Finally, we reiterate previous critical scholarship when we argue that the current win-win narratives accompanying technocentric CE frames will face substantial challenges. As real-world implementation unfolds and the CE moves increasingly from the realm of discourse into policy, this will undoubtedly reveal further tensions and frustrate rhetorical agreements on concepts and assumptions between actors (Palm et al., 2022, p. 378). Much work still remains to be done to challenge the technocentric consensus and allow a truly transformative CE to emerge.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Notes

1 In this article, we refer to the communications from the EU Commission using their official document numbers.

2 The Juncker Commission was the European Commission in office from 2014 to 2019 under President Jean-Claude Juncker.

3 One of the publications was co-authored between the IEEP and the EMF. We treat this as a separate actor in the analysis, rather than as two separate publications by each actor.

4 Organization size is based on the Commission's categorization at their public initiatives platform.