Localised waste reduction networks, global destruction networks and the circular economy

Abstract

Creating a circular economy (CE) is considered central to solving problems like climate change and resource depletion. In this context, the concept of global destruction networks was developed to better theorise using waste in new production. However, CE advocates also seek to avoid waste production by extending products’ lives. These efforts occur within what we term waste reduction networks. Examining how these two sets of networks intersect allows a fuller understanding of the CE’s emergent geographies. We analyse 17 European and US clothing firms focussed on value creation through waste minimisation to illustrate our argument.

Introduction

Since McDonough and Braungart (2002) published Cradle to Cradle: Remaking the Way We Make Things, much thought has been given to how to (re)design manufacturing to create circular economies (CEs). A central goal of efforts to build these CEs is to decouple economic growth from natural resource depletion and environmental degradation through activities that reduce, reuse and recycle materials in production, distribution, and consumption processes (Hvass and Pederson, 2019; Murray et al., 2017). Key to this is shifting away from production processes configured around linear flows of materials and energy towards circular or “closed loop” systems configured around questions of resource use and waste residuals (Brydges, 2021). This requires a fundamental change in how products are designed, produced and consumed so as to place sustainability, closed-loop thinking and post-consumer practices at the centre of business models (Hvass and Pederson, 2019). For manufacturers, a CE approach can include engineering products for greater longevity and for better circularity through designing them to be more easily taken apart by recyclers and/or making them easier to repair. It can also involve transforming products into services through rental relationships, with the seller/producer taking back the product at the end of its useful life for one consumer and renting it to another consumer or disassembling and recycling the product (Stahel, 1982).¹

Although there may be agreement on many fronts concerning the need to create CEs, what constitutes one is much contested (Kirchherr et al., 2017). Indeed, the history of the development of the CE concept reveals a diversity of understandings of the constituent parts, processes and outcomes of circularity in production (Murray et al., 2017). For Linder and Williander (2017, 183), though, the defining characteristics of a circular business model are that “the conceptual logic for value creation is based on utilising economic value retained in products after use in the production of new offerings”. These “new offerings” take multiple forms. For any firm wishing to foster circularity, however, product design and the development of efficient material salvage and recycling industries are unquestionably critical considerations, as production must be configured around facilitating resource recovery from discarded commodities to avoid waste. This means that value creation occurs differently from how it occurs in linear production. Thus, in a linear system value creation for, say, apparel manufacturers includes “procurement of raw materials and intermediaries (textile, accessories, machinery), and the management of production” (Nachum and Uramoto, 2021, 39). Alternatively, a CE approach requires including in the production process intermediaries specialising in product repair, resale, recovery and material recycling/upcycling. These intermediaries play a central rather than peripheral role, including maintaining and releasing value from materials that a linear approach would define as waste. Given that these actors are embedded in diverse social networks and labour processes (labour-intensive versus capital-intensive forms of taking discarded materials apart, for instance), there is consequently no single pathway towards circularity (Bryson and Ronayne, 2014; Frenken and Boschma, 2007). Equally, the overall aims of circularity are often (problematically) assumed to be universal when, in fact, different actors may seek quite different things from its implementation. There are many different CEs, with each firm (or even product line) developing a tailor-made solution that shifts production/consumption towards some type of circularity. Finally, whereas some see the CE as key to solving contemporary problems like climate change and resource depletion, scholars like Corvellec et al. suggest that it is not a panacea and that the “pathway towards circularity [will] be a circular economy that is modest” (2022, 429).

Despite differences of opinion concerning what constitutes a CE and how to achieve it, without question breaking extant links between mass production, mass consumption and waste generation to create a CE requires understanding the social and spatial lives of commodities in terms of both their production and consumption. The former is reliant upon the latter—production is often driven by marketing campaigns and design-centred product obsolescence (Bryson and Rusten, 2011) wherein consumers are persuaded to replace functional objects in response to short-term changes in taste or fashion. It also requires the creation of alternative production and consumption pathways that focus on reducing waste and consumption and encouraging reuse, repair and remanufacturing. However, historically research on firms’ efforts to implement CEs has been somewhat spatially circumscribed, largely focussing upon three somewhat localist scales of analysis (Murray et al., 2017): (i) single enterprises involving firm-level studies of cleaner production (Yuan and Shi, 2009); (ii) supply chains within inter-firm clusters (Baldassarre et al., 2019) and (iii) firms’ CE practices within individual cities (Bolger and Doyon, 2019; Zhang et al., 2010). As such, CE studies have generally explored either particular locations within which waste is managed or the waste itself (Gregson et al., 2015). More limited attention has been given to value chain-specific circularity and/or points of differentiation between sectors or discussion of where the challenges or geographic limits of the circular approach lie.² As a way to further develop this latter approach, we focus here explicitly upon how places are linked across space through being incorporated into various what we term waste reduction networks (WRNs).

For us, WRNs, which may be configured locally, regionally, nationally and/or internationally, involve firms reducing the generation of waste and thus the need to recycle by, for instance, extending product life cycles. In this regard, we see WRNs as complementing what Herod et al. (2014) have termed global destruction networks (GDNs). However, there is an important distinction to be made between GDNs and WRNs. Specifically, whilst GDNs are “networks of places where products are disassembled and their constituent parts are extracted for processing and re-use” (Herod et al., 2014, 427), WRNs are networks of places where products’ initial manufacturers and/or their intermediaries seek to extend such products’ lives to delay their transfer to GDNs. Whilst they can complement one another, each type of network makes a different contribution to efforts to develop circularity. This is an important point to recognise because each firm’s location—both organisational and geographical—within the economy can shape how it responds to exogenous and endogenous circularity drivers, such that whilst a firm may opt to become engaged in a GDN at some moments it may drop out of this and pursue involvement in a WRN at another (and vice versa).

Empirically, our study focuses upon the fashion industry, which is paying considerable attention to questions of circularity—in recent years circular and “slow” fashion have emerged to counteract the industry’s negative aspects, particularly those associated with “fast” fashion. These developments on the part of some manufacturers reflect ambitions to move away from the “take-make-use-dispose” model of production and towards a more regenerative system based upon closed loops (Henninger et al., 2021). However, given the ephemeral nature of clothing tastes, there are concerns about the degree to which circularity can be achieved (Brydges, 2021; Henninger et al., 2021). Acknowledging this, we explore the implementation specifically of circularity practices by US and European producers of “eco-friendly” activewear. This is a sub-section of the industry that is growing dramatically,³ given mounting consumer interest in “healthy lifestyles”, and that often associates itself with a broader environmental ethic—its products are frequently sold with the intent of encouraging consumers to have a closer relationship with nature, a goal often reflected in product advertising.

Although we focus here on matters of circularity, it is important to recognise that concerns with circularity are part of the much broader debate on corporate social responsibility, responsible business and sustainability. This is important because, in an ideal world, any reconfiguring of pathways towards circularity would not compromise other aspects of social responsibility, such as bringing pressure to bear upon clothing workers’ wage levels. In practice, though, there are sometimes tensions between these different goals which can lead to political frictions—some environmental activists may see ensuring circularity as more pressing than guaranteeing labour rights whereas trade unionists may view the latter as more important. These frictions can shape how both WRNs and GDNs are structured and developed. Equally, social responsibility must be differentiated from some of the more operational and design challenges related to enhancing circularity. For example, concerns around working conditions in garment factories have received ongoing attention from companies, but this does not speak to circularity. Likewise, broader interests in, say, reducing energy usage can be motivated by an environmental impact or response to energy volatility rather than circularity (Mulhall and Bryson, 2013). In our understanding of circularity in eco-friendly garment production, then, we distinguish between general corporate social responsibility initiatives and specific circularity initiatives.

The paper has four parts. First, we lay out a conceptual framework for characterising and locating WRNs in the CE, together with how these activities might mesh with those of GDNs. Second, we detail our methodology. Third, we present some of our empirical analysis of firms’ efforts to develop and geographically configure and reconfigure WRNs. We end with some general conclusions.

Circular economies, global destruction networks and waste reduction networks

Production and consumption are inherently geographical processes, linking myriad places together. Both generate waste. In recent years much conceptual and empirical work exploring production geographies has done so through investigating the structure of Global Value Chains (GVCs) or Global Production Networks (GPNs). Influenced by the “follow the things” framework articulated by Appadurai (1986), GVC/GPN analysis has aimed to track commodities through production processes to understand how different parts of the planet are being connected in new and different ways. This work, though, has “predominantly adopted a linear view of the life of commodities” (Herod et al., 2014, 421) and has not really thought much about what happens to commodities after they have been discarded by their initial user, other than to imagine them as having been transformed into “waste”. In recent years, though, the study of such waste has become a burgeoning research area.

Waste studies and discard studies

The emerging fields of waste studies and discard studies (Doherty and Brown, 2019; Lau, 2023; Liboiron and Lepawsky, 2022) add an important dimension to the follow the thing research agenda.⁴ Within these fields much analysis explores the dissolution of things (Gregson, 2023). In this context, in an effort to foster a more complete understanding of commodities’ lives and to better understand the geography of what happens to them post-consumption Herod et al. (2014) developed the concept of GDNs. These are networks within which previously used goods are dismantled and transformed into raw materials for inclusion in new commodities as part of an emergent CE. Research on GDNs has focussed on understanding secondary resource recovery, including illicit recycling, involving local formal and informal practices (Gregson and Crang, 2017, 2018; for more on geographic research into illicit activities, see Chiodelli et al., 2017; Hudson, 2014, 2019). Across the planet, several agglomerations, or waste clusters, like Agbogbloshie in Ghana and Bekasi in Indonesia, have become important nodes in various GDNs, a development that reveals how “everyday economic relations, transactions and interdependencies at the scale of the local play a prominent role in establishing and consolidating global networks” (Wang et al., 2022, 549).

Increasing numbers of studies have focussed on understanding what happens to things once they have been discarded by consumers and have entered global recycling networks and/or GDNs (Gregson, 2023). There is much complexity here, though, as recycling may involve something as simple as composting biodegradable materials or as multifaceted as decommissioning a ship. However, we want to suggest that studies of recycling and component recovery need to be complemented by research that explores initiatives that extend the life of things to delay the moment at which they are finally discarded as waste. In so doing, it is important to appreciate that waste is generated at all stages of the production process. In the electronics industry, for instance, e-waste is often assumed to be a post-consumer problem. Yet, as Lepawsky (2018) notes, the bulk of the industry’s waste is generated before an electronic item is ever sold to a consumer. In addition to examining the material recovery processes that take place in GDNs, a fuller understanding of waste must therefore include examining efforts to reduce its generation in the production process as well as investigating the importance of post-consumer repairing and refurbishing of commodities. Developing CEs will thus require establishing new business models or innovation within existing models and operational processes, with an emphasis on the control of products and materials throughout their entire life cycle (Wells, 2013). One such innovation that several clothing firms have increasingly turned to in this regard is what we are here calling WRN.

Waste reduction networks

So, what is a WRN? We conceive of a WRN as a network of actors who are involved in extending commodities’ lives to delay their entering the waste stream. This may involve actors in the production process (e.g., engineers and/or designers who configure products to last longer) and/or actors working in the post-consumer marketplace (e.g., people who repair goods already sold to consumers or who resell and/or repurpose used goods). These actors and networks often work in conjunction with GDNs in CE creation. However, there is a critical difference between GDNs and WRNs that relates to where in a commodity’s life cycle efforts to capture value and circularise its flow occur. Hence, GDNs are networks in which commodities like old computers that have been discarded are dismantled, with the materials secured from them used as inputs into new production processes. In this process, they carry forward the value contained within these materials or components. Sometimes these recovered parts may be reused pretty much as is (as with motherboards that become the basis for building bespoke, local-branded computers in the back alleys of places like Bangalore) and sometimes they may be more fundamentally transformed (as when copper from wiring is melted down and transformed into cooking pots by local manufacturers) (Herod et al., 2014). Hence, diverse types of inputs into new manufacturing processes can be treated quite differently in GDNs, depending upon whether they have high recycling potential (e.g., cardboard), lack systemic reuse (e.g., polymers), are by-products that can displace virgin material intake (e.g., plastic bottles turned into polyester for clothing manufacture), or are innovative products that are fully restorative by design and intention (e.g., bio-materials) (Ellen MacArthur Foundation, 2014; Murray et al., 2017).

Whereas GDNs focus upon processing post-consumer products that have become waste, WRNs, on the other hand, form around production and post-consumer initiatives to minimise waste generation in the fabrication process, as well as around processes configured to extend product life. Pointedly, rather than being geared towards releasing materials from waste for re-use via a GDN, WRNs use different types of feedback loops, or “self-replenishing circuits”, —including reuse, repair, remanufacture and recycling (King et al., 2006), with the number of “Rs” varying according to the sector examined and/or the entry point for circularity—to ensure products’ continued utilisation. Consequently, businesses that develop WRNs can transition from only generating profits by selling artefacts to also generating them by engaging in various post-consumer practices (reusing, repairing, adapting and upgrading commodities). This shift, however, may require quite radical, even revolutionary, alterations in commodity design, production processes and consumer behaviour (Bakker et al., 2014 and; Bocken et al., 2016). If a GDN thus represents a W–NP–W (waste–new product–waste) circuit (Herod et al., 2014), WRNs might be thought of as complementary NP–M–R (new product–maintenance–reuse) circuits.

Following the above, it is evident that combining WRN and GDN activities can potentially provide economically viable ways to create a more robust and far-reaching CE. However, in considering the potential to develop WRNs and how they might intersect with GDNs, it is important to recognise the role not just of the original manufacturer but also of various intermediaries. These include companies that innovate to create new eco-friendly materials designed with circulatory practices in mind and companies involved in reselling, repairing, and renting products rather than selling newly manufactured goods.

To summarise, WRNs configured by firms or intermediaries can extend the life of goods and materials in one of two ways. First, resource loops can be restructured to slow them down. This can involve four possible strategies on the part of manufacturers and suppliers—designing for longevity, designing for leasing or servicing rather than for sale, designing for reuse in manufacturing and designing for material recovery (RSA, 2020). These strategies avoid designs that are configured around fast or temporary fashions They can also involve designing out the creation of waste within production processes.

Second, in addition to designing their products for longevity firms may also engage in consumer-orientated initiatives to encourage maintenance, continued use and reuse. There are, though, at least two tensions here. First, those who operate within GDNs often have an incentive not to see products’ life cycles extended because any extension may reduce the amount of waste to which they can have access. In turn, this reduces their potential to capture value through dismantling this waste and extracting materials which they then sell onwards to manufacturers. This is quite different from the work of WRNs, which is geared specifically towards extending commodities’ lives. Second, by intent WRNs are designed to decelerate product obsolescence whilst accelerating alternative moments of value capture based on repair, reuse and resale. This runs counter to the model of post-war capitalism in which planned obsolescence has long been a central aspect of the accumulation process by encouraging consumers to constantly buy new products. It also represents an important difference concerning value. Hence, a GDN generates value through the labour applied in the dismantling process. This is combined with the residual value incorporated in the original commodities being dismantled that is then sent on to be incorporated in new commodities in which any useful materials recovered are used. In a fashion industry WRN, by way of contrast, on-going processes of value creation come through the labour involved in repairs, alterations, upcycling and in reselling and/or renting items, which might be conducted by the original manufacturer or may be undertaken by others.

The discussion above has important implications for understanding the unfolding economic geography of any form of CE. Clearly, infrastructure is required to support repairing/reselling/renting activities and this infrastructure may be configured internationally, nationally or sub-nationally. Its geography will reflect both sectoral and firm-level determinants (below we focus on firm-level determinants of WRN geographies in the activewear industry). Equally, GDNs for particular commodities are likely to have very different geographies than the WRNs in which those same commodities are embedded. For instance, in the case of activewear repair facilities, the repair elements of a commodity’s WRN might be located close to initial consumers whereas the entry point to the GDN to which old clothing is sent for dismantling and turning into, say, rags for paper manufacture or into Circulose®, a viscose fibre made using cotton waste that can be used for manufacturing new clothing, may be on the other side of the planet. It is important also to recognise that much as different GDNs have different geographical characteristics—e-waste GDNs’ spatiality is quite different from those involving paper or ships, for instance—so, too, are different sectors’ WRNs organised differently, with all the geographical consequences this brings with it. Hence, some may focus more upon remanufacturing whereas others focus more upon repair and/or enhancing reuse of commodities ‘as is’. With remanufacturing the emphasis is on recovering a commodity to return it to its original manufactured form in terms of quality, specification and performance to prevent it from ever becoming waste. This may involve sending it back to a factory at some distance from its owner. Enhancing reuse, on the other hand, is really more about delaying the point at which an article eventually becomes waste and may simply involve facilitating reselling the article in a second-hand store. In the case of activewear companies, however, the scope for remanufacturing is limited, largely because the industry lacks the technology required to produce garments from old materials and many of the materials lack durability (Brydges, 2021). Nevertheless, activewear firms can readily offer repair services (sometimes in-house and sometimes via third parties) to consumers to reduce or delay the need for landfilling or recycling.

Production-orientated and consumer-orientated WRNs

Putting all of this together, we suggest that the concept of WRNs allows us to better theorise the different strategies being developed by firms in their commodity circularity journeys. We want to distinguish, though, between two types of WRN, both of which reflect firm-level product differentiation strategies. First, there are “production-orientated WRNs”, where firms demonstrate to consumers the additional value of their products based upon them being more sustainable (if not always circular), with the focus being upon waste minimisation in production processes. This is about product and production design focussed on removing/reducing waste. With this type of WRN, new and recycled materials may be obtained from significant distances via GPNs and GDNs but some firms may define circularity around the configuration of much more local networks to access materials. Second, there are “consumer-orientated WRNs” in which firms create additional value by configuring consumer and post-consumer WRNs. There are, therefore, different types of value creation involved in WRNs based on how feedback loops are generated between various stages of the value chain. These include: designing for repair, reuse and recycling; producing using more sustainable methods and reusing waste materials; selling products and retaining value through maintaining and supporting products whilst with the consumer; and facilitating return post-consumer.

Importantly, these two types of WRN require infrastructure that has different geographies. For instance, production-orientated WRNs may be more globally orientated, allowing for production in, say, China for European and US consumers. Consumer-orientated WRNs that focus upon repairing, reselling and upcycling products, however, require infrastructure that can be readily accessed by consumers. This suggests that they are more likely to be configured locally, regionally or nationally rather than internationally, as a consumer in Europe is unlikely to send a pair of jeans to China to have them repaired to extend their life. Every firm, and even every product line, will have to develop its own approaches to configuring these two types of WRN. This may include focussing on obtaining material inputs locally through highly localised GDNs and/or configuring highly localised commodity repair, resale and recovery infrastructure.

Having laid out a conceptual framework for exploring firms’ waste reduction strategies, our empirical research now turns to examine the sustainability journeys of eco-friendly activewear companies to explore how various firms have developed post-consumer WRN strategies.

Methodology

This paper is based upon extensive secondary data collection of self-declared eco-friendly US and European activewear companies. We began with a systematic review of the activewear market in both territories, using and comparing rankings of firms based on their claimed eco-friendly credentials. This enabled us to identify 86 small and medium-sized enterprises involved in the production of eco-friendly activewear. In-depth research was then conducted on each, with the criterion for selection for our study being that firms must have at least one identifiable environmentally friendly practice or policy related to circularity (covering materials choice, production standards and/or waste reduction). The removal of several firms suspected of “greenwashing” on the basis that evidence could not be identified to support the claims made by these firms narrowed our database to 80 (40 in the USA, 40 in Europe). Information collected included that on materials (which used, where sourced), production (location and type of factories, working conditions, certification, auditing, logistics and distribution) and post-consumer activities (circular initiatives, “giving back”, trade-in-services, etc). The firms identified provided extensive public details via their websites and other media (such as sustainability reports and statements) regarding their applied and evolving definition of circularity, including their product-design philosophy, selection of textiles and other materials, waste minimisation strategies and production geography (this latter included naming factories and details of any existing or planned post-consumer recycling initiatives). All 80 had multiple certifications and all were evaluated by rating agencies, with these processes including independent assessments of production processes, waste strategies and labour practices.

All companies included in our analysis talked about trying to minimise the environmental impacts of their products through aiming to be carbon neutral, through reducing water waste, through implementing “green” logistics (using recycled and/or organic fabrics, employing highly localised sourcing and fibres that had been produced using closed-loop systems, and making use of recycled and recyclable packaging), and through designing products for circularity to enhance longevity and educating consumers regarding garment care to enhance longevity. They also sought to show concern about working conditions in their factories. However, no firm was completely circular, with the primary challenge they faced being based around post-consumer recycling. Nevertheless, of the 80, 17 (Table 1) went beyond simply using recycled packaging materials (a common practice across the firms in our larger dataset) to include post-consumer and end-of-life considerations as a central element in their business model and were using this as a unique selling point. They are, in other words, concerned with the whole product life cycle and are going beyond the production stage in the value chain to consider what happens after a garment has entered the post-consumer stage. In this sense, they had created what we are calling consumer-orientated WRNs and each had processes in place that cut across the product life cycle and provided evidence that they were engaged in designing for longevity, reducing and removing waste and post-consumer initiatives based upon repair and reuse. This made them the core subset of our analysis. A spreadsheet was constructed that included detailed information on each of the 80 firms, though these seventeen had additional columns of data detailing their post-consumer initiatives.

We decided that interviewing was not required, as sufficient information regarding these firms and their activities was available from the secondary sources consulted. This was primarily because these firms position themselves as being eco-sustainable and part of this positioning includes much greater data transparency than is the case with many other types of firms—providing consumers with a way to evaluate their eco-sustainability is an important element in these firms’ marketing strategies. Our method, then, is based on identifying firms and then engaging in a complex and time-consuming data identification and assessment process that includes triangulation to ensure that any claim made by these firms is supported by practice. Post-consumer initiatives were assessed by exploring how these could be experienced by potential/actual consumers—assessing what was actually possible (via visiting, for example, company websites reselling used items) rather than assuming company claims unquestioningly. The decision not to interview is grounded in a research design based on testing claims made by firms, with the extent and nature of this testing process being highly unusual. Our approach is one based on speculative theorising to formulate propositions that we then test via rigorous assessment of information that is available to all consumers and researchers.

Our analysis was supported by information provided by ratings agencies and interviews with the founder(s) undertaken by journalists and/or presentations made by founder(s) on YouTube and other platforms. An important advantage of our methodology is that firm names do not have to be anonymised. We use actual company names in the text to identify firms and codes (US1/EU1, etc.), and details of these firms are provided in Table 1. Following the database’s formation, the 17 firms were re-examined in January 2022, June 2023 and January 2024 to investigate their statements regarding post-consumer recycling and circularity. We considered not only their claims but also the reality of the consumer experience. The process of data collection and analysis involved co-authors independently researching all seventeen firms, focussing upon their reduce, repair and reuse activities (with reuse broken down into second-hand, downcycling, upcycling and renting), as well as returning to the original 80 firms to re-review their activities in the interests of research rigour. Co-authors met to discuss their independent findings and conducted a second round of firm review to ensure we had not missed something.

Characterising activewear waste reduction networks

In what follows we show how the 17 firms in our subset have gone about constructing WRNs in their efforts to achieve post-consumer circularity. Importantly, each activewear company has a slightly different circularity strategy. These strategies reflect the firms’ leadership teams’ objectives and policies centred around competing based on their products’ eco-credentials. One conceptual and methodological challenge it is important to note when it comes to exploring WRNs, though, involves differentiating between different types of corporate sustainability strategy, specifically concerning the question of what differentiates a company that is configuring some form of circularity strategy from one that is merely “eco-friendly”. To illustrate our argument, below we detail some of the activities engaged in by these firms.

All firms in our final subset use “eco-friendly” materials, including biomaterials and/or recycled artificial fibres.⁵ The varied use of these materials, though, positions these firms differentially within various (and different) biological networks and GDNs. This means that there is an important geography to recognise when it comes to how they are developing their WRNs, as there was no common approach to manufacturing or post-consumer recycling. Rather, each firm was developing a solution that engages with its own specific organisational structure, production processes and circularity definition and each had different feedback loops designed to enhance product life or return its products after use. These different choices indicate how the ways in which firms are embedded in all sorts of local and extra-local relationships shape their agency and how such agency, in turn, is shaping the uneven spatial development of the emergent CE. Hence, whilst a minority acted to minimise locally their carbon footprints by using deadstock fabric or left-over fabric obtained from co-located firms that would normally be sent to a landfill, others relied on fibres and fabrics produced by value chains that were configured nationally and/or globally.

WRNs in practice

One US apparel company that has sought to develop both what we are calling a production-orientated WRN and a consumer-orientated WRN is éclipse (2023) (US27), which highlights its “thoughtful consumption and sustainable business practices”. This includes implementing three goals: “taking back our used garments, increasing our recycled content, designing for physical and emotional durability” (éclipse, 2023). Whilst a focus solely on sustainable business practices would not in itself reflect a circularity agenda, éclipse engages with sustainable practices across the value chain. First, it designs and manufactures durable clothing, which comprises providing resources to consumers on how to care for their clothes (including a slow fashion guide). This emphasis on durability, including repair, supports the configuration of the firm’s consumer-orientated WRN. A second activity is the firm’s use of recycled fabric to ensure that its products are manufactured ethically. This supports the configuration of this firm’s production-orientated WRNs. For éclipse, ethical manufacturing involves ensuring that the firm’s “pieces are sewn locally” within the USA to “reduce our carbon footprint, support local jobs and contribute to our economy” (éclipse, 2023). This includes enforcing a supplier code of conduct and external certification. The focus of a third activity is on logistics, given this activity’s environmental impacts. For éclipse this includes “planting a tree with each order” and “transitioning to compostable/recyclable packing and shipping material” (éclipse, 2023). The use of the word transitioning highlights that éclipse, like the other 80 firms, sees itself as being on an eco-journey towards what it defines as sustainability.

As éclipse has developed its sustainability strategy, then, it has sought to integrate its consumer-orientated WRN with its production-orientated WRN. Post-consumer concerns are expressed by éclipse as part of its fourth activity, which is based on the circularity principal that “[e]ven after éclipse clothing leaves our shop, it doesn’t stop being our responsibility”. Hence, éclipse includes a take-back programme to keep its products out of landfills. Although this company ships worldwide, there is a clear geography to this takeback programme; consumers are encouraged to upcycle, mend or swap garments in their home communities or they can post used éclipse garments to the company’s headquarters in Colorado and obtain 20% off their next purchase. This is important, because whilst ensuring that all fabric scraps from the production process are recycled back into the manufacturing process contributes to a firm’s sustainability and production-orientated WRN, having post-consumer repair services and take-back programmes in place is what gives it circularity through the configuration of consumer-orientated WRNs. Nevertheless, for this firm, the emphasis is placed on consumers posting garments back to a US address, a fact which highlights that this activity is orientated towards US consumers. The fifth activity involves having external certification and internal and external processes in place to confirm the circularity of a firm’s products. For eclipse, this includes being monitored by an external Advisory Board.

In the case of Swedish manufacturer Filippa K (2022, 7) (EU22), on the other hand, the firm notes that

Sustainability has been at the core of Filippa K’s DNA since the beginning, with a circularity strategy in place since 2014. The brand takes a holistic approach centred around mindful consumption: the responsible creation of high-quality fashion that promotes a lifestyle of owning fewer, better pieces and keeping them in use for as long as possible.

This company sells worldwide. It has long-established practices around product differentiation but is more recently concentrating on becoming more sustainable by organising its activities according to the three sustainability pillars of circularity, traceability and impact reduction, so that it might “design for the full life cycle of a garment with the intention of reducing, repairing, reusing and recycling” (Filippa K, 2022, 7). It seeks to implement this strategy through designing products for longevity. This includes ensuring that 62% of garments are made from mono-fibres to enhance recyclability. It also includes providing repair and garment-remaking services and facilitating a second-hand market by providing a mechanism for consumers to resell used garments. However, like éclipse Filippa K’s Preowned initiative has a very restricted geography, with the firm’s Preowned space being restricted to “our community in Sweden”. Filippa K has, though, configured its consumer-orientated WRN initiatives to retrieve used products from consumers for recycling. This highlights that a direct link can exist between WRNs and GDNs, as used products are sourced via WRNs and then transferred to GDNs for recycling. This connection is detailed in Figure 1, which is an adaptation of a value chain diagram developed by Filippa K (2022, 4) to illustrate its approach to sustainability. Informed by engaging with the conceptual literature, this Figure is also grounded in empirical analysis, as it highlights the distinction applied in practice by these firms between production, use and activities designed to extend product end-of-life.

Filippa K’s sustainability strategy includes five measurable goals that it hopes to implement by 2030 (Filippa K, 2022, 7):

• (i) 100% of garments received “through claims or collection, as well as materials from garment waste, will be remade, resold or recycled”;

• (ii) 100% of the company’s products to be fully traceable and certified;

• (iii) reduce overall emissions by 50% by 2030 and be carbon neutral by 2050;

• (iv) use fully-recyclable fibres in 100% of its products; and

• (v) ensure that there is no exploitative labour or corruption in the company’s supply chain and that all people are paid a fair wage.

To ensure that these policy commitments are embedded throughout its operations and value chain, the company employs two “protection agents to ensure that no action or behaviour goes against the policy commitments” (Filippa K, 2022, 8).

From our brief overviews of Filippa K and eclipse, it is obvious, then, that following and implementing a circularity strategy requires a holistic approach that begins with product design and the configuration of a sustainable approach to manufacturing. Applying circularity principles to activewear, however, comes with many tensions. For instance, in an interview, Eva Karlsson, CEO of Houdini (EU3), noted that:

… [we] clearly promote a circular user-phase —including high frequency and versatility of use, care and repair, and reuse … [Our clothing is] also circulating in our rental system. So, there is plenty of circularity in there, even though we haven’t yet solved the material issues. This is a typical Houdini dilemma, or healthy friction, and why we always look at the holistic impacts … We take extended producer’s responsibility. Products are returned to us either for recycling, reuse, or second-hand sales (or reuse as we call it). Since 2006, all worn-outs that have been returned to us which cannot yet be recycled are stored at our warehouse. As owners of natural resources in the shape of fibers, fabrics, or products, we have committed to being a custodian of those resources. The way we see it, ownership gives us the right to use but not the right to waste (Karlsson cited in Klingelhöfer, 2022).

This highlights the complexity of trying to practise circularity as a firm balances design and production with approaches to waste management.

Processes involved in consumer-orientated WRNs

Well-designed products support an approach based on firms configuring production-orientated WRN initiatives based on minimising waste in the production process before continuing to take responsibility for their goods by developing consumer-orientated WRNs. Pulling back from a focus upon these two firms to a broader analysis of the 17, we identify 5 activities that support consumer-orientated WRN operations: (i) providing additional repair guidance; (ii) making available repair services; (iii) operating take-back programmes for resale or upcycling; (iv) providing resale programmes and (v) offering rental schemes (Table 1 and Figure 1).

Each of these five activities is important, even if some are more tokenistic than others. Each also has quite different geographies, with some being more geographically restricted than others. Hence, on the one hand, additional repair guidance might be provided pretty much anywhere via the Internet, with someone downloading a PDF of instructions. On the other hand, in the case of the seven firms that have take-back programmes in place for product reuse or upcycling (Vaude (EU21), Filippa K (EU22), Globe Hope (EU24), Veja (EU5), Pact (US6), Mate the Label (US21), Girlfriend Collective (US16)), a consumer’s physical location will play a role in how the WRN develops—it will be easier for customers living close to these firms’ outlets or to a mailing service to take advantage of such programmes than it will be for those living further away. Equally, although some firms provide customers with a discount voucher to encourage them to return products, geography matters here too, as participating in such initiatives will be easier for some consumers than others. An example of this type of WRN is provided by the Girlfriend Collective (US16)—although this firm accepts returned items in exchange for a $15 store credit, with the items upcycled into new pieces, this initiative is only available in the USA even though this firm ships to consumers located in Canada, the UK and Australia, and sells in other markets through local stockists. Such constraints clearly limit the geographical reach of many post-consumer WRNs identified in this paper and highlight that, for these 17 firms, these are local, regional or national networks rather than global ones. In turn, this can limit their effectiveness, since persuading consumers to return items is key to being able to justify the infrastructural investment required to make them work.

Whereas éclipse and Filippa K have both focussed on changes in the manufacturing process and initiating take-back programmes, three other firms (REI (US1), Houdini (EU3) and Vaude (EU21)) emphasised providing additional care and repair services. For them, this is a key part of their circularity strategy, although it should be noted that it can also be used to generate additional revenue. Hence, Vaude places Care and Repair on the frontpage of its website with detailed advice, including videos, provided for all product categories whilst REI offers detailed expert repair advice as well as selling items to support the product repair process. Overall, seven companies provide repair services (Houdini (EU3), Vaude (EU21), Filippa K (EU22), Globe Hope (EU24), Veja (EU5), Picture Organic (EU8), Outerknown (US25)), either directly or through a repair partner network. There is also a geographical constraint here, though, given the relationship between the expenditure needed to support the provision of repair services versus demand. Specifically, repair services may only be available to customers in some markets or locations—large retailers with stores in many cities will be more able to provide repair services close to their customers than will smaller retailers with perhaps only one or two outlets. For instance, French manufacturer Veja (EU5) has cobblers available in only two stores in France, although in 2020 it did open its first repairing and recycling project store. However, the company sells products through 3000 retailers and in 50 countries. Meanwhile, one firm (Globe Hope (EU24)) offers an almost lifetime repair service for some products and this reflects products designed for longevity.

Five companies have developed resale programmes on their websites (Filippa K (EU22), Globe Hope (EU24), Mate the Label (US21), Outerknown (US25) and Wolven (US17)). These take different forms. Filippa K established Filippa K Preowned as a space on the firm’s website where customers can list their items for resale and either receive cash or a Filippa K credit. When customers want cash, the company recommends a listing price of usually around 50% of the garment’s original value, depending upon its condition. When sold a prepaid shipping label is provided to the seller, who is responsible for packing and posting the item to the buyer. Whilst this scheme supports recycling, it is only available in Sweden. Filippa K, however, usually offers a much higher value when store credit is sought rather than cash for reselling goods, something which obviously benefits to a greater extent consumers who live close to a store compared to those who do not. This is a strategy intended to create captive customers who are locked-in to this brand. In a similar fashion, US firm Outerknown (US25) established a resell platform, Outerworn, in 2021. Customers locate the item they want to sell in their order history, share a brief description with the company and set a price that is reviewed by Outerknown. Once it sells, a prepaid shipping label is provided by the firm and the seller receives store credits.

Whereas these five firms accept their products back for resale, 10 (59%) (two of which—Globe Hope (EU24) and Outerknown (US25) —are amongst the five) have take-back programmes in which products are returned by consumers and then recycled (these are Houdini (EU3), Vaude (EU21), Globe Hope (EU24), Picture Organic (EU8), éclipse (U27), Outdoor Voices (US9), Womsh (EU12), Vaude (EU21), Outerknown (US25) and Arielle (US20)). Whilst this type of take-back programme does not extend product life, it does ensure materials’ ongoingness. However, although companies typically try to avoid downcycling when recycling, as this is counter to circularity, they cannot always do so. The primary challenge they face here is that most do not have sufficient scale in terms of product returns to justify investment in recycling operations. Consequently, intermediaries like SuperCircle, established to manage the logistics of clothing recycling processes by investing in technology, logistics and warehousing, have formed to fill this gap. Los Angeles-based Mate the Label (US21), for instance, has a partnership with SuperCircle, where customers can return products in return for credits. As with Filippa K Preowned, though, this creates captive customers, as the credits must be spent with Mate. Nonetheless, the linking of SuperCircle’s website to Mate’s does allow the firm to know what an individual has purchased and the product’s material data. This is important information to have to avoid material downcycling. Indeed, what Mate’s partnership with SuperCircle highlights is that central to the configuration of some WRNs is the development of technology and reverse logistics infrastructures that support a company focussing on products designed for circularity. More particularly, companies like SuperCircle can reshape the geographical scale at which circularity may occur by creating regional logistics aggregation networks that are sufficiently large to support clothing recycling for many brands. The outcome of this is a more efficient sorting process that enables SuperCircle to curate bails of materials that have been sorted by type, which facilitates their ongoingness in ways that would not otherwise have been possible.

Finally, the early debate on circularity initiated by Stahel (1982) noted that products could become transformed into service relationships by companies renting rather than selling them. Although there are clothing intermediaries who specialise in renting clothing, significantly three of the seventeen firms we studied rent products directly: REI (US1), Houdini, (EU3) and Picture Organic (EU8). To Picture Organic (2021) renting “is a step to diversifying our business. It’s a step towards more recirculation programs. It’s a step towards achieving our mission”. Interestingly, this company’s introduction of a product-service system was supported by a discussion of the academic literature on rental clothing. There are three issues here. First, the company argues that producing five jackets and selling them to five consumers results in five satisfied people, but “quite a lot of emissions” (Picture Organic, 2021). The alternative is to produce one jacket and rent this to five individuals. The outcome is still five satisfied people but much-reduced emissions. Second, renting out goods not only supports circularity but also adds an additional source of profit to the firm. This includes profit from the eventual sale of a rented garment. Importantly, there is a geography to this as the environmental benefits of rental must include an assessment of rental store locations and customer travel to these locations. Third, rental is better suited to specific-use products that are infrequently required than it is to others—for example, skiing clothing and technical jackets. Whilst Picture Organic acknowledges that “rental is another drop in the bucket”, the firm also argues that it is another small-scale potential solution (Picture Organic, 2021).

To conclude, we should note that geography plays an important role in all of this. Distance can be a significant problem with creating consumer-orientated WRNs. Hence, the CEO of Houdini (EU3) has noted that “At the moment we are only doing rentals in the Nordic countries, because if you take freight into consideration, it isn’t really a sustainable alternative to ship back and forth across continents. It needs to be regional. So we’re building within regional clusters one step at a time” (Klingelhöfer, 2022). Geography and its consequences for establishing reverse logistics, then, is currently a major constraint on developing spatially extensive consumer-orientated WRNs. Whilst most companies hope to extend their initiatives to international customers, our analysis highlights that their WRNs, though important in circularity terms, are typically presently spatially fairly restricted to regional or, at best, national scales of operation, with the services they provide tending to be available at only a limited number of locations. Whether they will ever become global networks remains to be seen.

Discussion and conclusion

In this paper, we have sought to develop the concept of WRNs as a circuit that is juxtaposed with the GDNs circuit developed by Herod et al. (2014). We believe that approaching the CE through an appreciation of how firms engage with GDNs, or limit the exposure of their products to GDNs, is critical. Central to this process are the ways in which firms configure WRNs that are intended to increase the circularity of their products and to avoid the production of waste. For activewear, the analysis highlights that whereas production-orientated WRNs are quite spatially expansive (sometimes even global) when it comes to material sourcing and value chain configuration, consumer-orientated WRNs tend to be more local, regional or national networks, given some of the commercial and spatial constraints on having consumers return clothing. This means that whilst domestic consumers may benefit from post-consumer recycling initiatives, consumers located in other countries typically have much more limited (and often no) access to producer-provided recycling initiatives. This is an important spatial aspect of circularity as practised by these 17 firms and highlights how geography can act as a constraint on firms’ abilities to develop CE initiatives. However, innovation could perhaps lead to more globally orientated post-consumer WRNs. One solution is intermediaries like the e-Bay, Vinted and Depop platforms that operate as centralised clearing houses for the global resale of used clothing. Although these resale platforms are global, much of the consumer activity transacted on them is local and/or national. For clothing, firm-led post-consumer product initiatives will need to be supplemented by investments in waste management systems that ensure that most of the clothing is recycled rather than landfilled. This investment will likely require government intervention.

Overall, our analysis has revealed that there appears to be no singular path towards circularity, as each firm is developing its own set of solutions to embracing the CE. Some of these firm-level solutions are extremely local and are deeply shaped by the geographical contexts within which different firms find themselves, both contemporary contexts and prior ones that have created particular path dependencies for them. Although the blend of approaches that each firm has applied to configuring firm-level solutions to production-orientated and consumption-orientated WRNs means that inter-firm differences are evident, we identified no systematic significant differences between US and EU firms, a finding which likely reflects the fact that these firms are competing internationally and experience many of the same sector constraints.

In contemplating the creation of a CE it is vital to recognise that firm-level application of CE style strategies is an evolutionary and competitive process. Initially, an eco-friendly product segment emerged in the activewear sector that reflected either the interests of the founders or the application of a marketing approach that was grounded in product differentiation configured around production-orientated WRNs. In these early days eco-friendly practices largely revolved around fibre/fabric choices and implementing production processes intended to minimise waste. Nevertheless, the rapid development of this eco-friendly activewear market segment has led to firms increasingly having to deepen their CE practices to maintain product differentiation and enhance their market position. Consequently, not only have many firms been forced to implement more eco-friendly production methods to remain competitive, they have increasingly considered the need to develop consumer-orientated WRNs as part of their business activities. This is an on-going process for each firm, as each develops a local, regional or national solution to configuring its consumer-orientated WRNs. Furthermore, whilst the initial structure of a WRN may reflect just an activewear manufacturer’s decisions regarding product design, putting in place the WRN’s architecture may sometimes be done directly by that firm itself but may also be carried out through intermediaries. This is an important point, as it highlights both a circularity scale issue that any one firm faces but it may also lock consumers into specific firm-provided repair/reselling/recycling initiatives with consumers only being able to return clothing to its original manufacturer for a credit. This is different from how GDNs work, as (for example) e-waste processors typically do not restrict the waste that they accept to specific electronics companies but, rather, accept all brands of computers, TVs and other discarded electronics. A key question that arises from the involvement of intermediaries, though, concerns who captures any value created by consumer-orientated WRNs (the original manufacturer or some intermediary?) and what is the geography of this value creation process?

Another key question concerns the role of the state in all of this. Although individually any one firm’s CE initiatives might be considered modest, collectively there is a larger effect, as consumers are provided with options to enhance product circularity. This effect will only grow over time. An important driver of this growth will undoubtedly be the application by governments of policies intended to encourage or force firms to accept extended producer responsibility (EPR) as part of a policy tool intended to minimise waste by extending producers’ operational and/or financial responsibilities for the post-consumer stage of their products’ lives. These policies will have dramatic implications for shaping various WRNs’ geographies. This is important because in considering the geography of any WRN it is critical to recognise that different CE product solutions come with very different environmental outcomes. Hence, products containing natural fibres tend to have the most reduced environmental footprints because it is easier for them to be repaired, reused and upcycled within firms’ WRNs before they are downcycled. On the other hand, firms that use other types of fibre (like polyester made from plastic bottles) typically have quite different environmental impacts, especially because it is more difficult to repair and upcycle them. Ultimately, though, the best environmental outcome comes from producing long-lasting products that can be repaired and resold multiple times, as this avoids the challenge of dealing with product end-of-life management.

Clearly, there are major geographical challenges to enhancing product circularity within WRNs. Nevertheless, the efforts of some firms to create such networks are an important element, we believe, in creating a more circular economy in the clothing sector. At present, too much clothing ends up in landfills or is sent for resale in emerging economies (WRAP, 2017), a phenomenon that is only likely to worsen with the growth of the ‘fast fashion’ industry. However, whereas landfilling brings with it all sorts of ecological and other issues (an inefficient use of resources, which are literally thrown away, a burden on local tax bases as landfills need to be expanded because they fill up so quickly, and so forth), sending clothing to the Global South can prevent the development of local textile industries and raises important questions regarding who benefits from such clothing “recycling”. There is also an illicit side to WRNs that requires research, and this includes things that are stolen to order, with this process perhaps extending product life, and labour exploitation linked to product repair and recycling. The challenge faced by all CE producers, though, is in persuading consumers to return products for recycling or resale. Whilst an alternative approach is for capitalism to shift from a process based on selling products to one based on rental agreements, wherein a consumer would never own a physical good but would have responsibilities for ensuring that the product is returned to the producer for post-consumer recycling, it seems unlikely that this will ever come about. Given this, future research is required to explore the organisation and geography of production- and consumer-orientated WRNs, as these perhaps offer a more achievable way out of our present take-make-waste economic models.

Acknowledgements

We are grateful to the editor for handling the editorial review of this article and for his guidance on revisions. The constructive comments from the three reviewers challenged us to refine and develop our argument.

References

Footnotes