1. Introduction
The greatest challenge facing the world at present is the threat of the climate emergency disrupting the stability of the planet [
1]. Tackling this issue requires transformational change underpinned by sustainability principles and prioritised by global governance, business, and citizen practice [
1]. The circular economy (CE) adopts sustainability practices by maximising finite resources and increasing product lifetimes as well as decreasing waste sent to landfills. However, as there has been little research into the implementation of the CE, especially from a fashion context [
2], it follows that there has also been little acknowledgement of how the CE can be marketed to appeal to consumers. The research that has been carried out on the circular fashion economy (CFE) has identified numerous barriers, and it appears that sustainable fashion is still a niche market [
2,
3], dominated by the fast fashion business model [
4]. To date, sustainable fashion has sought to appeal to consumers’ moral and ethical values, underpinned by their concern for the climate emergency [
5]. In this article, we explore how the CFE can compete with fast fashion by adopting a similar marketing strategy that offers consumers viable alternative option to acquire fashion.
The fast fashion industry is inherently unsustainable, and efforts to address sustainability are considered inauthentic in tackling the core issues [
6]. Fast fashion marketing tactics encourage frequent impulsive consumption of inexpensive garments [
7,
8], of which production is reliant upon scarce resources and exploitative practices (of people and the planet) [
6]. Low pricing increases notions of disposability, and consumers have been socialised to acquire and dispose of fast fashion frequently [
9] in what have become established social norms of fashion practices [
10]. This once innovative and competitive fast fashion business model is now the dominant social paradigm (DSP) of fashion production and consumption, and both the industry and consumers are locked into this system [
5]. Fast fashion sustainability is a superficial response, passing responsibility to consumers to donate unwanted garments to charity rather than take responsibility to integrate sustainability into production and support disposal routes [
11,
12]. Yet, the volume of clothes donated to charity is significantly greater than consumers buying second-hand fashion [
3,
13]. As discussed below, the literature presents clearly defined barriers preventing participation in alternative fashion practices [
14,
15]. Advances in the CFE would make the most of scarce resources in circulation and reduce exploitative practices of people and the planet [
2,
3]; however, there is little understanding of how this might be actualised within fashion practices, especially when competing with the DSP of fast fashion marketing and retailing and the inclusion of social and aesthetic capital, which underpins involvement in fashion.
This paper proposes that, to advance the CFE, similar tactics that encourage the social and aesthetic capital of fast fashion consumption are necessary for competitiveness; there are lessons to be gleamed from the way in which the fast fashion industry operates that can be utilised to disrupt the DSP. The paper argues that, just as fast fashion disrupted the fashion industry, resulting in a race to the bottom [
4], innovation can break through this DSP to create new fashion practices that offer alternative distribution channels that will appeal to consumers’ sustainable values. Previous research has outlined an existential vacuum in which the fashion industry and fashion consumers stagnate, aware of the fashion industry’s contribution to the climate emergency but compelled to continue within the DSP for fear of losing market share and not conforming to social norms [
5,
14,
15]. Research has examined fashion engagement, demographics, values, life stage, and knowledge of the issues (see [
13] for an overview) but has found no clarity in what encourages sustainable fashion practices, as recognised in the well-established attitude–behaviour gap [
16]. Although change in societal, institutional, and business behaviours are necessary, we believe that learning and replicating how fast fashion marketing has captured consumers’ attention to entice rapid fast fashion consumption can innovate the CFE to disrupt the incumbency of the DSP and provide a resolution to this reported attitude–behaviour gap. Therefore, this paper will present an established CFE business, “Wardrobe, as a case study. Wardrobe is an online a peer-to-peer fashion-sharing platform operating in the USA that offers consumers access to luxury brands as well as celebrity wardrobes. We examine Wardrobe’s business model through a conceptual framework that was constructed from reviewing the fashion and marketing literature along with the theoretical framework of disruptive innovation developed by Christensen [
17]. Disruptive innovation has been successful in determining shifts in the marketplace that are supported by technological advancement and provides an opportunity to understand market innovations that disrupt current ways of business operations. Market expansion is often sought by moving into new geographical areas, yet there can be new forms of business operations that can support growth, and this has been evident in disruptive business models in the sharing economy, such as Uber and Airbnb. Goffin, Åhlström, Bianchi, and Richtnér [
18] assert that case study research can help to explain new concepts of business model innovation. The paper will first review extant literature to construct a framework of fast fashion marketing concepts that can underpin a CFE and then outline the theory of disruptive innovation and its application within the fashion sector. This will be followed by introducing Wardrobe and demonstrating how this CFE utilises similar tactics and terminology as those f fast fashion marketing to redirect consumers’ fashion practices. The paper will conclude with implications for theory, fashion marketing, and retailing.
3. Theoretical Framework: Disruptive Innovation
Christensen’s [
17] seminal theory of disruptive innovation (DI) has been much cited over the last few decades, as well as being challenged on theoretical and practical elements (see [
51]). Nevertheless, the theory does enable a closer examination on how to disrupt DSPs through the role of innovation and carve out new value propositions to support the CFE. As presented in
Figure 3 below, the theory of DI sits amongst other types of innovation theory that are utilised to solve problems; what makes DI more relevant for this paper is how it can advance technological and marketplace shifts [
52]. This is evident in how the fast fashion industry became possible through globalisation and technological advances, where low pricing was facilitated by cutting corners on good practices that protected the environment and workers. More recently, technology and the marketplace has shifted again through instant global communications accessible on smartphones [
3] and offers a platform to share the growing concern for the climate emergency.
Satell [
52] advises that when a problem is well defined (i.e., fast fashion produced through exploiting people and the planet and marketed to encourage frequent consumption and disposal), breakthrough innovation can help provide a solution. Breakthrough innovations introduce a new set of performance features to transform existing markets. Yet, sustainably produced fashion remains niche due to barriers of fashionability, price, and accessibility, as established above, and has not drawn consumers from the DSP despite their concern for the climate emergency. Therefore, new sustainable features have not resolved the problem. Conversely, when it is the domain that is well defined, disruptive innovation is more suitable [
52]. Although the problems of fast fashion regarding the climate crisis are well defined and increasingly filtering into social discourse, it is not enough to educate consumers on the impact and expect them to make sustainable fashion choices when it requires a sacrifice in price and style [
5,
14,
22]. Further, the new set of performance features of sustainable fashion do not address the social and emotive capital imbued in fashion engagement. Although some fast fashion retailers have introduced facets of sustainable fashion, such as one sustainable collection that is dominated by continued production of fast fashion, it is marketed on the same principles as fast fashion: cheap, fast, and of poor quality, and this superficial response is considered inauthentic [
5,
11,
12]. Therefore, as addressing the problem definition has been unsuccessful, it may be better to consider the domain—the marketplace—and disrupt the DSP by using similar social capital marketing tactics.
Christensen [
17] proposes two types of innovation: sustaining, either evolutionary or revolutionary, and disruptive. Sustaining innovations do not impact existing markets but offer improvements to products and services that evolve within markets or transform markets, often at the high-end and luxury sectors, without disrupting existing market players [
51]. This is illustrated in sustainable fashion, which occurs more authentically in higher-priced brands. Although it could be assumed that the fast fashion business model was disruptive, in driving down the price of fashion to be more affordable, convenient, and more responsive to trend changes, it does not align with Christensen’s [
17] framework, as will be discussed below. Rather, sustaining innovations are prevalent in the fast fashion industry, competing in terms of price in a homogeneous and crowded market. Similarly, sustainable fashion has not disrupted the DSP of the fast fashion business model; rather, sustainable fashion is perceived as more expensive, not fashionable, and less convenient to access and aligns with high-end encroachment theory, as outlined by Schmidt and Druehl [
51]. These established barriers support the cycle of unsustainability that the fashion industry and fashion consumers are locked into [
5,
9]. Therefore, to challenge the DSP, disruption is required, and the next section will examine Christensen’s [
17] framework to consider how to provide similar emotions and experiences to challenge fast fashion.
The theory of disruptive innovation creates a new market or enters at the bottom of an existing market, and Christensen et al. [
53] stipulate four elements that underpin disruption. The first is new consumer categories that can be attracted by new technology and business models. Christensen et al. [
53] provide the example of computers moving into consumer homes, whereas previously they were only owned by businesses, opening up a new consumer market. Mitzkus [
54] considers Ford to have disrupted the automobile industry by introducing a moving assembly line that reduced the price of production, enabling automobiles to become more affordable and enter the mass market. Although it can be assumed that fast fashion was disruptive by reducing the price and ensuring that consumption is more accessible, the model has cannibalised production to encourage frequent impulsive consumption that sustains the industry rather than creating new consumers [
4]. Historically, when fashion was not as affordable, consumers bought less, shared more, and repaired damaged garments [
15,
23]; fast fashion has simply accelerated the cradle-to-grave process. Disruption of new consumer categories will come from the CEF by designing a new business model of acquiring and practicing fashion that offers a viable sustainable and affordable route to acquire fashion through an innovative business model.
The second is new technologies that use technology to better meet consumer needs. Christensen et al. [
53] provide the example of Netflix moving from lending physical DVDs to consumers through postal systems to the instantaneous act of streaming films though a subscription service. This convenience and expansion of product availability offered a USP that provided a competitive advantage [
55]. Similarly, Mitzkus [
54] notes the Apple iPod as a technological development that, although smaller, enabled consumers to access a wider variety of portable music than the personal portable DVD and cassette players. Although the iPod was initially expensive [
51], the price points were reduced within a variety of models. New technologies also played a role in lowering the price of fast fashion and quickening the production process to respond to rapid trend changes. However, this represents sustained innovation that accelerated fast fashion consumption, exacerbating the exploitation of the environment and workers rather than using technology in a new way to enhance the consumer experience. Although fast fashion retailers harness technology to augment the consumer experience by minimising risk and effort for consumers and offering personalised marketing [
13], the aim here is to encourage more consumption, which does not support the sustainability agenda in addressing the issues that are of concern to Generation Z. Arguably, this capitalistic model benefits fast fashion business owners and shareholders more than consumers [
56], who profit from the insecurities of social capital, and this expansion of social inequalities and lack of diversity does not sit well with Generation Z [
19].
New business models are the third element of Christensen’s theory [
17], and examples include Alibaba, Uber, Airbnb, and Facebook as business models that have eliminated the need for physical assets; rather, they act as a conduit between producers, retailers, and consumers, and this is supported by digital technology [
3]. The Alibaba group (the Chinese equivalent of Amazon), the worlds’ largest retailer, does not own any goods; Uber, the worlds’ biggest taxi group, does not own any vehicles [
54]; Airbnb, the worlds’ biggest provider of accommodation, does not own any real estate; and Facebook, the worlds’ biggest media company, does not create any content [
57]. There are lessons for the fashion industry from these examples that support the CE by maximising the utility of garments’ lifespans and making the most of resources before they end up in landfills. In contrast, the fast fashion model, hailed as new in streamlining production efficiencies and increasing the competitiveness of low pricing, has reduced consumer thresholds of how much they were willing to pay [
23]. Fast fashion is reliant on frequent impulsive consumption to rotate their inventory, as profit comes from cutting production costs and generating sales. Taking inspiration from the four examples provided above in which disruptive business models do not accumulate physical assets, unlike fast fashion, “efficiency and sufficiency” [
42] (p. 189) may be found in the CFE.
Fourthly, exploiting old technologies in new ways provides a means to engage with new customers through new technologies and by creating new business models to disrupt markets [
3]. An example would be when scientists at IBM began to track Sputnik (the first artificial Earth satellite) in space for their own amusement, which then led to the development of the technology now used for GPS [
58]. This system was initially used to locate and track between our planet and space and is now used by consumers in their everyday lives [
58]. Accessible through smartphone technology, from finding locations and tracking their loved ones to checking into social media and being globally connected, smartphones have centralised technology within everyday lives [
58]. The development of smartphone applications encourages consumers to share their lives online and offers personalised marketing [
8]. This technology could support the development of the CFE. Historically, consumers shared garments within informal networks of hand-me-down clothes in families and communities. New business models could be developed and supported by digital technology to widen the geographies of networks for sharing. Expanding on social media as a site for practical sharing, it could also provide visual virtue signalling repositioning social and emotional capital that will shift social norms of fashion acquisition, challenging the DSP.
Lastly, Christensen et al. [
53] assert that to be disruptive, innovation should emerge from the lower end of the market and appeal to consumers through innovative attributes. Therefore, brands such as Tesla, maker of electric automobiles, are not authentically innovative, as they appeal to the high end of the market and therefore are not affordable and accessible to a new market of consumers; rather than disrupting business models and technology, Tesla sustains innovation within the automobile sector. The lower end of the market represents consumers least willing to pay for the product [
51], as represented in
Figure 4, which fast fashion is currently sustaining within the DSP. Therefore, innovation that disrupts this DSP will need to entice low-end consumers by creating new values at a similar price point to move into the mainstream. Collectively, the proposed model of
Figure 4 presents Christensen et al.’s [
53] elements as imbued within social and emotive contexts, combining the conceptual and theoretical frameworks to demonstrate how to disrupt the DSP with the CFE. The next section introduces the case study and will examine how it relates to
Figure 4.
4. Case Study: Examination of an Online Fashion Rental Platform: Wardrobe as a Circular Business Model
Creswell (2007) recommends a case study approach to an inquiry when exploring how culture operates within a bounded system. Indeed, the seminal work of Christensen [
17] was based on a set of case studies. As the purpose of this paper is to examine a CFE business model that competes with fast fashion, a case study provides a methodology for closer investigation of a specific issue in a real-life setting [
59]. The case under examination is Wardrobe (
https://www.joinwardrobe.com), a peer-to-peer fashion-renting platform offering access to borrowing celebrity and influencer wardrobes. Based in New York, Wardrobe was founded by American entrepreneur, philanthropist, and artist Adarsh Alphons in 2018. Although Wardrobe positions itself as “the living archive of fashion sitting at the intersection of sustainability and influence” [
60], the original premise of the business inception was to maximise real estate space by liquidising wardrobe content rather than addressing sustainability or creating a CFE business. For the first three months of operations, prior to the COVID-19 pandemic, garments were collected from and dropped off at drycleaners, who ensured that the garments were clean and in good condition. During the pandemic, exchanges moved online, with delivery and collections from couriers via the drycleaners. Over the last few years, Wardrobe’s operations have grown tenfold, demonstrating the appeal of this business model in surviving the pandemic disruption during their infancy of three months of trading. In conversation with the marketing manager, Nina Rowan, growth was sustained during the pandemic, despite the social restrictions, as young people rented fashion to generate content for their social media. This demonstrates the value of social capital created through social media (
Figure 2) via engagement from Generation Z and suggests that, despite social confinement leading to social media as being their sole external window, engaging with Wardrobe provided hedonistic emotions that alleviated the monotony. Wardrobe enables access to luxury branded fashion such as Louis Vuitton, Prada, Celine, and Chanel, and for around half of the borrowers, it is their first experience with luxury fashion. Borrowers can also rent iconic vintage and celebrity wardrobes, which includes fashion worn in films, TV, and music videos and on the red carpet. For example, Queer Eye star Antoni Porowski’s closet collection is for rent, including the shirt he wore in a Taylor Swift video and the black leather jacket he wore on Queer Eye that was immortalised in Porowski’s Lego figure. Hence, the unique history of the product adds value to luxury brands that cannot be found elsewhere.
Having acquired consent from the CEO to gather data, this paper reports on Wardrobe’s operations over 2021 and 2022, when the average price of a four-day rental was USD33.10; since Wardrobe was an intermediary of the renters’ wardrobe/closets rather than investing in inventory, the price to rent garments was comparable to that of average fast fashion. Engagement navigates between inviting celebrities and influencers to rent their wardrobe on the platform and the renting process. Renters can maximise their inventory by collecting revenue on their assets within 2 h of posting, and the highest earning lender has, to date, earned USD10K+, which is symbolic of the micro-entrepreneurship noted by McFarlane et al. [
36]. There has been less focus on marketing, with the priority on streamlining efficiencies in the users’ experience (both those who rent out their wardrobe and those who rent the garments) and using technology to maximise the turnaround of cleaning and shipping. Marketing is user created [
8], where renters post their activities on social media and use #Wardrobe, with the company then sharing those stories on social media.
Although the research could have adopted interviews or surveys with the users of Wardrobe or the influencers/celebrities who rent their wardrobes on the platform, it was decided that a case study would be more appropriate to analyse the business model as the focal operation of innovative commerce, as was argued by Arrigo [
3]. Extant sustainable fashion literature has found that fast fashion and sustainable fashion are incomparable in terms of price, accessibility, and style from a myriad of methodological approaches across the positivistic and interpretive spectrum, leading to a saturation of understanding how to advance the sustainable fashion agenda with no clear way forward in advancing the CFE. Yet, Wardrobe does not compete on sustainability credentials; rather, it is the social and emotive context that underpins this business model, and this is a novel approach that can be replicated to advance the CFE. Wardrobe also demonstrates that social media plays a pivotal role in co-creating identity narratives between Wardrobe, influencers/celebrities, and those renting the fashion. It is these identity constructions that are of interest in driving new social norms of fashion practices and re-socialising consumers away from the DSP. To our knowledge, the application of marketing tactics to encourage engagement in the CFE has not been examined previously. Further, viewing the CFE through the theoretical framework of DI is also unique, as previously, sustainable fashion and the CE were considered sustaining innovations by competing through sustainable production and appealing to ethical values. Over the last decade, sustainable fashion has struggled to compete with fast fashion, establishing the case for disruptive tactics. This approach aligns with Goffin et al.’s [
18] assertion that a case study is useful for exploratory research as well as “theory-building, theory-elaboration and theory-testing” (p. 595), which begins with explaining how the case was selected for theoretical reasons.
Following Stake’s (cited in [
59]) intrinsic case study procedures, the authors recognised that Wardrobe challenged the DSP of the fast fashion industry with a strategy that could be disruptive, thus adopting what Creswell terms “purposeful maximal sampling” (p. 75). A singular focused in-depth case study offers strategic insight into an issue, and although this does not enable generalisation, it offers an undiluted insight [
59]. Adopting Yin’s [
61] embedded analysis to focus more strategically, the conceptual framework presented above was inductively developed from extant literature in order to understand the context in which fashion operates within society and which factors hinder the progression of a CFE. The case study data (see
Table 1) were deductively considered against the conceptual framework to enable a pathway for disrupting the DSP.
Goffin et al. [
18] advocate that the quality of case study research depends on the development of the theory and the appropriateness of the case to be studied. As researchers with a background in fashion marketing, retailing, social media, and consumer behaviour, specifically through the lens of sustainability, our observations and familiarisation with the relevant literature over the last decade, along with our contributions to the literature, have enabled an insight into the challenges of advancing the sustainable fashion agenda. This has led to our conceptual framework presented above, where we identified a case (Wardrobe) that offered an appropriate lens to examine the research problem [
59] through the conceptual framework (theory building) and testing the theory of DI (theory elaboration) [
18]. Following the advice that Goffin et al. [
18] offer of evaluating the application of case study research in the innovation literature, the details of the research design, data collection, and data analysis provide transparency in the processes. In their study, Goffin et al. [
18] did not identify any variation in the quality of papers that were based on single, two, or three cases; therefore, in our paper, we offer no comparison to Wardrobe to distract from the focus of this unique business model. As a caveat, we could not source another fashion-renting platform that enabled the ability to borrow influencers’/celebrities’ own clothing. Therefore, the value of this paper lies in the conceptual and theoretical development of transferring the approach of the DSP to the CFE in such a way that centralises social and emotional capital by “theory-testing” the innovative model of the case study of Wardrobe.
As
Table 1 displays, data did not emerge from one source, and it is pivotal to observe Wardrobe from the consumer perspective, because despite increased awareness of the impact that fast fashion has on the environment and allegations of exploiting fashion workers, as noted above, consumers continue to purchase fast fashion and apply a myriad of excuses for doing so. Therefore, the website was the first unit for analysis in May 2022. Secondly, to better understand how the business model was designed, an online interview with the founder and CEO was carried out in June 2022, adopting a semi-structured format with questions inspired from the literature. Thirdly, as marketing is often blamed for encouraging frequent impulse consumption using manipulative tactics, we were interested in how the marketing manager approached designing content and campaigns, which was explored in asynchronous interviews. Interestingly, the marketing followed two strategies: encouraging influencers/celebrities to rent out their wardrobes through the platform and encouraging consumers to borrow influencers’/celebrities’ clothing. Therefore, we also had access to the content that was presented to the influencers/celebrities as a business proposal and observed the co-created marketing on social media. Finally, we reviewed the social media network Instagram for Wardrobe to observe “instances” of content using #Wardrobe. These parameters are similar to the main data collection materials from the case studies examined by Goffin et al. [
18].
4.1. Data Collection
The marketing manager of Wardrobe contacted one of the authors in November 2021 after reading their research, and this began an exchange of discussions through email and online meetings, during which the marketing manager presented corporate information that aided the understanding of the business model operations from inception. In between those discussions, observations were made of the website content and the marketing campaigns of new influencers and celebrities who were sharing their Wardrobes to the renting client base were received. Collectively, the importance of capitalising on social and emotive capital—the opportunity for consumers to acquire status fashion to present themselves in their social worlds and in which they could be seen—became apparent; this has been missing from previous literature examining fashion consumers’ perspectives of sustainable fashion. It is also important to note that communications began in November 2021, with the previous two years having been disrupted by the global COVID-19 pandemic. Fashion retailers had struggled due to the forced closure on non-essential retailing, and the restrictions had reduced the kind of social occasions that often trigger fashion consumption [
13]. In contrast, Wardrobe reported tenfold growth during the pandemic years, as Generations Z and Y borrowed celebrity luxury fashion to post selfies on social media wearing the influencer/celebrity outfits in their homes. The novelty of this was amusing, but more importantly, the practice aligned with social and emotive capital that avoided the DSP; rather, this practice endorsed the CFE and highlighted Wardrobe as a disruptive and innovative business model using similar DSP marketing tactics, and as such, it merited further investigation to map out the conceptual and theoretical implications to aid in understanding of how to advance the CFE.
Analysis included an ethnographic evaluation of the sample in its setting [
59], as presented in
Table 1 above. However, the conceptual and theoretical frameworks also inform the setting—the broader societal and commercial constructs within which consumers operate and thus support theoretical development of how the CFE can disrupt the fast fashion paradigm. The frameworks offer a strategic approach to collecting relevant instances that exemplify meaning; therefore, the analysis is deductive as being informed by the conceptual and theoretical frameworks, with themes that were extracted from the literature reviewed. As our previous research has explored facets of fashion consumption and sustainable behaviours, it was recognised that the business model of Wardrobe addressed the social and emotive contexts often missing with the CFE. Therefore, the conceptual and theoretical foundations were examined with the data parameters presented above. Having established the conceptual framework and identified an appropriate theory, the analysis of the case study was independently carried out by both authors to support inter-coder reliability [
18].
Table 2 presents a summary of the data collection, timing, and analysis.
Having established the functionalities of the case study and the rationale for selecting Wardrobe, the paper will examine the implications of Wardrobe as a CFE example and how this responds to the conceptual and theoretical framework presented in
Figure 4 above.
4.2. Wardrobe Opens up New Consumer Categories
New consumer categories are evident in both the influencers/celebrities who rent out their wardrobes and the borrowers who may be renting fashion for the first time, given that the practice is still to infuse into societal norms [
2,
50]. Whereas celebrities/influencers are often used as marketing intermediaries to sell the inventory of fashion brands [
21,
33,
36], Wardrobe is unique in being the intermediary between celebrities/influencers and consumers, who are often admirers of the celebrity. Marketing that depicts fashion-focused lifestyles deepens the emotive connection between the celebrity and the audience. As Alphons recognised, the relationship moves beyond the digital realm to “wear[ing] that jacket in real life, it’s kind of mind blowing in that way [that’s just] not possible in any part of the world, apart from the world we’ve created.” Utilising this social capital not only provides a unique experience but can also generate excitement that is co-created between Wardrobe, the influencer/celebrity, and the consumers renting the fashion, especially through visual presentations on social media where hashtags (#) can be used to connect all co-created stakeholders and provide a visible external indication of an aspirational “good life” [
15,
25,
28,
29,
30].
New consumer categories are also formed, as 50% of Wardrobe’s Generation-Z consumers experienced luxury branded and iconic vintage fashion for the first time, as the pricing is affordable for the average consumer [
2,
10]. Therefore, Wardrobe creates disruption from the lower end of the market [
53], where consumers’ pricing thresholds are reduced [
23]. The opportunity to rent luxury fashion may supersede concerns around ownership found in previous research [
42], especially if Generation Z prefer to not repeatedly wear the same garments [
14,
22], as this is a sector from which many fashion consumers are locked out due to higher pricing [
23]. Consequently, Wardrobe enables access to a variety of luxury and celebrity fashion, which can be used to replicate aspiration and success [
15]. The role of highly desirable luxury brands becomes a prop within the construction of self-identity to project a desired image socially [
26], creating aspirational social capital on social media. This blurring of entertainment and commercial activities on social media platforms [
8,
21] has long been a tactic of fashion marketing [
4]. Yet, within this practice there is also the opportunity to renegotiate the stigma that has historically been found in the sharing of garments, especially in hand-me-down clothes that were passed between families and communities [
23]. As such, this sharing model emerges from the luxury sector and celebrity/influencer capital [
21,
34], supported by digital technology to present this desirable hedonistic experience socially. The model supports widening geographies of networks for sharing and could potentially reduce preferences for ownership, especially when luxury and vintage fashion can be more playful and experiential. This has been enhanced by new technologies.
4.3. New Technologies
Wardrobe benefits from new technologies that have made renting fashion easier to engage in [
3], as well as connecting influencers/celebrities with their fan base [
31]. This is increasingly utilised by online fast fashion marketers to entice frequent impulsive fashion consumption, creating a saturated and highly competitive industry where advantages are found in speed to market and lower price points [
22]. However, this comes at a cost: exploitation of people and the planet [
6]. As consumers are increasingly aware of the impact of fast fashion on the climate crisis and have expressed discomfort with the way in which fast fashion employees are treated [
14,
19,
24], renting becomes a viable alternative that enables the utilitarian maximisation of scarce resources from the cradle to the grave [
40]. Although Wardrobe satisfies those least willing to pay [
51], it also aligns with status and esteem to develop a new purpose of renting, which addresses both the social and emotional context sought by Generation Z [
14,
19,
22]. This has the potential to reshape established social norms of fashion acquisition and provide a USP of sustainable advantage [
55]. To make this more appealing to consumers, Alphons explained how they track data to understand and improve the user experience for both influencers/celebrities who lend their fashion and consumers who borrow the fashion. This includes timelines, convenience, cleaning, marketing, and browsing. Therefore, in the same way in which fast fashion has examined the user experience and offered discounts, free delivery, and payment options to ease the barriers to and risks of consumption [
7,
13,
39], Wardrobe focus on refining this new business model to entice new consumers through new technologies, which reduce the risks and enhance the experience of acquiring fashion through the CFE.
4.4. New Business Models
The new business model created by Wardrobe is similar to that of Alibaba, Uber, Airbnb, and Facebook, who do not invest in inventory and generate economic capital by acting as an intermediary between owners and borrowers. This model replicates fast fashion characteristics of low prices, access to a wide range of fashion, and quick turnaround of fashion items provided by influencers/celebrities by enabling consumers to change wardrobes more frequently [
10] but does so in such a way that it elevates social and emotional capital. Access to a wide range of luxury fashion merchandise and high rental turnover at a low price resonates with Generation Z’s need for fashion acquisition [
13,
22], which is still driven by style, price, accessibility, and branding [
14,
15]. Thus, “Sufficiency” [
42] (p. 189) is evident, as Wardrobe still resembles the fashion hauls depicting variety, but this is underpinned by a circular model that avoids landfill contributions [
22,
43]. The model is augmented by enhancing the user experience through technology to make the process more efficient, user friendly, and sustainable. In addition to recognising the role of social and emotive context, this business model offers alignment with fashion self-identity and sustainability: no sacrifice required. This is an illustration of how Wardrobe is becoming a destination for Generation Z and fashion creators.
Although the delivery of the garments is not instantaneous, Wardrobe is as accessible as buying fast fashion online, with a unique USP of belonging to a celebrity/influencer and social capital that supersedes the rapid style change characterised by fast fashion. Similar technologies that enhanced the fast fashion consumer experience by minimising risk and effort for consumers [
13] can align with the sustainability agenda, as we cannot buy our way out of the climate emergency [
41]. The role of celebrities/influencers is paramount to the success of this business model. As celebrities/influencers are renting their own wardrobes, they are more authentic, especially as there is no sponsored content with brands or advertisers. Authenticity helps to develop a sense of trust with their followers and contributes to the success of the business model. Financial benefits for the celebrities/influencers are achieved through rental, and this platform provides celebrities/influencers an opportunity to communicate their wider values. For example, rental proceeds can be donated to a charitable cause. As the business model moves beyond any one brand and operates on celebrity capital to generate social capital, Wardrobe is exploiting the social capital of fast fashion marketing to direct consumers to the CFE.
4.5. Exploiting Old Technologies in New Ways
Exploiting old technologies in new ways is less demonstratable in a technological sense, as the amalgamation of fashion, identity, and social media is not a new concept. It is ingrained within fashion discourse, and fashion is dependent on gaining social acceptance to gain traction [
8]. Similarly, social acceptance is crucial to advancing rental fashion [
2], which illuminates the opportunity to mimic fast fashion marketing and permeate societal norms. Fashion consumers look to social media for fashion inspiration [
24], and Wardrobe provides a similar online experience that provides the lifestyle experiences that Generation Z crave [
4]. Content is stimulated by with #drops and new content that satisfies the hedonism of impulsive and frequent fashion acquisition [
13]. although social media enables branding to be co-constructed between the brand’s communications and consumer interactions, Wardrobe focus on sharing user experiences as their marketing tool, maximising the word-of-mouth exchange [
24] and embedding authenticity directly from the influencer/celebrity to the consumer. The benefits of this focus on the user experience to encourage more consumers to rent from the CFE and make the experience less risky [
13]. Communicative exchanges through social media increase visibility [
44] and mimic the blurring of fashion entertainment and commerce in which consumers have been socialised [
9,
10]. The model does not require Generation Z to give up any of their values in fashion consumption; rather, it allows them to be enhanced and is responsive to alleviate the guilt experienced from fast fashion consumption [
14,
23]. Generation Z can engage with experimenting with fashion styles and constructing fashion self-identity through continued engagement with Wardrobe. Collectively, this suggests that renting platforms have the potential to disrupt the DSP.
The Wardrobe case study is an illustration of disruptive innovation that can progress the CFE. Despite the model emerging from an economic perspective, it addresses sustainability by minimising the excavation of scarce natural resources for limited use before being disposed in landfills [
4,
22,
43]. Considering the rapid growth of the business in a short time frame, harnessing technology to enhance the service attributes of renting could potentially pave the way for the company to become a market leader. The disruptive and innovative model that exploits the current DSP ensures that Generation Z do not need to sacrifice social and emotional capital value in their fashion practice; if anything, this model enhances their overall self-identity and provides egotistical value, as the model allows them to take account of their sustainable contribution. Additionally, if the rental is delivered by courier, the same way that fast fashion is delivered, then consumers will still experience the hedonistic thrill of receiving a new (to them) garment to wear and be seen in, as well as experiencing the thrill of unpacking [
13,
48].
To map out the ways in which Wardrobe disrupts and innovates the DSP of fast fashion,
Table 3 captures the main points from the conceptual framework that have been presented above through the lens of DI. In the first column, the mechanisms that underpin fast fashion management and marketing are established as a basis of comparison, followed by the components of Wardrobe and how this creates appealing values. Having demonstrated Wardrobe as a CFE business model that can provide added value for fashion consumers, concluding comments, limitations, and recommendations for future research are provided next.
5. Concluding Comments
That the CFE has not yet disrupted the incumbency of the DSP is not an indication that it has failed to do so. As Schmidt and Druehl [
51] assert, disruption often takes time to encroach on market share and embed new consumer practice, and it may never fully displace fast fashion. However, Jain et al. [
2] considered that future research should focus on understanding consumer needs, motivations, and preferences for renting fashion and provide strategies for marketing managers; however, this does not include using the same strategies as fast fashion marketing to appeal to the same fast fashion consumers. Examining the business model of Wardrobe has illustrated that there is potential to for the CFE to disrupt the DSP of fast fashion using a similar approach. Further, Jain et al. [
2] suggest that marketers of fashion rental businesses could promote the utilitarian and hedonistic values that encapsulate the social and emotive capital of being able to access a large quantity of luxury fashion, and our research has identified the benefits of doing so. Value can be co-created between the business, influencers/celebrities, and consumers on social media. Indeed, this seems to supersede social activities to become a way to present oneself on social media to curate social capital. Although Jain et al. [
2] also suggest that the environmental benefits also be included in marketing, extant research has already shown that this alone will not change behaviours; nevertheless, as a secondary consideration, it will lessen the guilt and cognitive dissonance of fashion acquisition and offer a tenant of value [
4,
5].
This novel paper makes four theoretical contributions to the literature. Firstly, the paper is one of the first to argue that development of the CFE is dependent on recognising the importance of addressing social and emotional capital to engage fashion consumers in adapting their fashion practice. Although the sustainability aspect is important, it is a secondary value that appeals to consumers, albeit it will endorse the practices of the CFE, and this will become increasingly important as concerns grow for the climate emergency [
24]. Therefore, combining the conceptual model developed from reviewing the literature with the theory of DI advances both sustainable fashion and innovation literature by illustrating that embodied social capital should be prioritised in the inception and marketing of the CFE. A second contribution is the unique examination of fashion, the conceptual model, and the CFE through the lens of a DI theoretical framework. This has enabled a deeper understanding of how the fast fashion model has dominated the marketplace and opens the debate that similar tactics can be utilised to entice consumers to the CFE. This offers a paradigmatic conceptualisation of how to market the CFE through developing innovative business models and addresses the attitude–behaviour gap that has existed over the last two decades [
16]. The case study of Wardrobe exemplifies that engaging with the CFE does not require a sacrifice of social and emotional capital—rather, it can enhance the experience with egotistical value. Thirdly, examining the CFE through the lens of DI indicates that many of the disruptive models that have evolved seamlessly into society and penetrated everyday lives originally evolved through other agendas—such as Sputnik leading the development of the technology now used for GPS ([
58], as described above). Similarly, Wardrobe emerged from the desire to maximise economic utility and, through this, identified a business model that could disrupt impulsive and frequent fast fashion consumption while also addressing sustainability. Fourthly, the argument that the DSP of fast fashion can be disrupted by using similar management and marketing tactics is a novel idealisation. Previous research has considered fast fashion an oxymoron to sustainability, and as consumption was stimulated by marketing, the link between mimicking fast fashion marketing to encourage CFE acquisition had not been recognised. However, given that consumers are reluctant to sacrifice their social and emotional capital, despite increasing concern for the climate emergency, this unique innovative approach may be pivotal in growing sustainable fashion practices.
Limitations and Future Research
This paper has focused on one case as an example of how the CFE can disrupt the fast fashion industry, and future research could expand to multiple cases for comparison and to illuminate different perspectives [
59]. The approach outlined in this paper is novel, emerging from understanding the sustainable fashion and fashion marketing literature over the last few decades. Although previous research has examined consumer experiences and perceptions of sustainable fashion and considered the sustainability of new business models, to our knowledge, a CFE example has not previously been explored within the conceptional and theoretical frameworks as presented above. Yet, there is much to learn from the application, despite the narrow perspective. As such, this paper does not provide a generalisation of the CFE; rather, it is an example of related concepts examined against the theory of disruptive innovation. It is also recognised that the application of DI is contentious (cf [
62]); however, the debate around disruption provides a framework for a deeper analysis and is not one that usually examines the fashion industry.
The fashion industry has always been reflective of looking back and looking forward for design and material inspiration [
4]. Many fashion actors have been described as disruptive—for example, Vivienne Westwood and Alexander McQueen—and been applauded for innovating ideas supported by technological advances. Christensen et al. [
53] responded to critiques of the theory of DI, citing the numerous dynamic variables that change the way in which the theory can be applied, and welcomed further study. Our paper contributes to this debate and further challenges the fashion industry to respond to the wicked problem of our time, the climate crisis, and by analysing DI alongside extant literature to identify solutions. Therefore, our paper emerges from immersion in the fashion, marketing, and sustainability literature and invites contributions to furthering this conversation. Other opportunities for deepening understanding of the CFE include exploring the consumer experience of the sharing economy, as concluded by Jain et al. [
2], which could be supported by adopting netnography to illuminate this social media window as representing social capital or immersive qualitative methods to reflect the lived experience.
One last aspect of a limitation and how it can be addressed in future research is the reliance on moving garments around using couriers, which creates other sustainability issues and carbon emissions [
2]. If consumers are expressing concern for sustainability, it will be an important issue to address and it will be important to provide transparency; yet, this can also be used as a marketing tool. For example, the Scottish rental company Advanced Clothing Solutions (ACS) offer high street fashion retailers a repair and resell service for returned clothing [
63]. ACS (2023) support the expansion of the CFE, and amongst their many accreditations they are both B Corp Certified and carbon neutral. As such, it is an organisation that can be integrated as a marketing tool to illuminate the efforts made to reduce the carbon footprint of renting fashion, including cleaning, repairing, and transport. Although we can recognise that Wardrobe protects scarce natural resources, maximises the lifespan of garments, and minimises what is sent to landfills, it is also acknowledged that this is part of a solution, and more work is required to implement sustainable principles in all operational aspects. Illuminating those processes aids the issue of transparency.