Collaborative consumption, a buzzword that has gone conceptual: Three shades of the sharing economy

It has been a decade now that research on the collaborative economy and collaborative consumption (CC) has thrived. Tremendous academic research has been conducted into this specific concept. This paper re-evaluates the conceptual framework proposed almost half a decade ago about the conceptual foundations, frontiers and limits of the concept of collaborative consumption. The paper provides a revised definition and assesses to what extent the scope and limits in contrast to other forms of exchange still hold now despite current challenges.


Introduction
Early on, we identified consumers' capacity to switch side from "obtainment" to "provision" or from an "obtainer" to a "provider" role as the key criterion of collaborativity, hence of the collaborative economy (CE) (Ertz et al., 2016). At that time this was much in line with other seminal works such as Botsman and Rogers (2010) and Gansky (2010), while differing slightly from those works by adopting a scholarly perspective (vs. a managerial one) and framing consumers' ability to switch roles from obtainer to provider as the key defining criterion of collaborative consumption (CC). As such, CC can be defined as a set of resource circulation systems which enable consumers to both obtain and provide, temporarily or permanently, valuable resources or services through direct interaction with other consumers or through a mediator (Ertz et al., 2019). Collaborative consumption is therefore a concept which stands in sharp contrast with the notion of conventional consumption. Conventional consumption -which underlies classic marketing thought -is a type of resource distribution system which involves passive consumers (not obtainers), who cannot, or are not given the capacity to, provide any resource or service (not providers). Incapable of engaging either in obtainment or in provision, their role is limited to that of buying -monetary exchange -and consuming organization-made resources or services, and, in the case of tangible resources, to discard them. In contrast, collaborative consumption involves not mere "consumers" but "obtainers" who may also be "providers". In sum, consumers' capacity to switch roles from provider to obtainer and from obtainer to provider in a given resource distribution system constitutes the key distinguishing criterion between conventional consumption and CC.
With regards to the scope and limits of the collaborative consumption, our stance was that it is comprised of two major blocks (Ertz et al., 2017): namely what we call 1) "mutualization" (i.e., renting, leasing, pooling, sharing [Gaiardelli et al., 2014]), in line with Arnould and Rose's (2016) call for integrating the mutuality construct as a substitute for the vaguer "sharing" notion; and 2) "redistribution" which involves the reownership of resources through giftgiving (or donations, passing along, giving tax deductions), swapping and secondhand markets. The rationale being that since our CC-defining criterion is ability to enact the obtainer-provider switchover, reownership markets enable that switchover. This article takes a broader perspective on the matter by conceptualizing the collaborative economy and the sharing economy.
1. Toward a new conceptualization of collaborativity and the collaborative economy 1.1. The definitional conundrum issue One recurrent feature of research in the CE and CC is the declaration that there is a "definitional conundrum" about the CE and CC (Murillo et al., 2017). Some authors do not attempt to define the CE (Plewnia and Guenther, 2018). Other authors tend to use inconsistent definitions throughout their work (e.g., Muñoz and Cohen, 2018). In their study of sharing economy business models, Curtis and Mont (2020) concluded nonetheless that there is a need to use a coherent definition throughout a specific work about the CE and to better "demarcate those practices included or excluded in the authors' definition of the sharing economy" (p. 4). We note however that the CE concept has increasingly been narrowed to one of its specific subcomponents that especially deals with webmediated mutualization practices.
Such conceptualization is becoming increasingly common and prevalent, probably because platforms and applications enabling consumer co-production constitute highly innovative and fast-growing segments of the overall CE. For example, Dellaert's (2019) developed a two-layered conceptual framework of consumer co-production networks in the CE. Given the mesh-like and rhizomatic features of the Internet (Giesler, 2006), web-powered networks were highly prevalent across all four categories of co-production networks. Given the increasingly prevalent dominance of platforms in the CE, many authors have gradually moved to the notions of "gig economy" or "platform economy", which are governed by "sharing platforms" to describe entities that facilitate "sharing practice" (Akbar and Tracogna, 2018;Ciulli and Kolk, 2019;Piscicelli et al., 2018;Hawlitschek et al., 2018). That web-centric focus has created a whole new subfield of research in the broader sharing economy area to the point that "sharing economy" and "platform economy" are often being conflated. In Ertz and Leblanc-Proulx's (2018) co-citation analysis of the collaborative economy, this subfield of research refers to cluster 3, which focuses on mutualization and sharing systems of all types of resources from a technological (platform) perspective. For example, Eckhardt et al. (2019) define the sharing economy as a technologically enabled socioeconomic system with key impacts on traditional beliefs and practices in marketing. Likewise, Curtis and Mont (2020) narrow the "sharing economy" to the "sharing platform" concept. Such platforms are considered as business models that create value by the mediation of an exchange between a provider called "resources owner" and an obtainer called the "resource user" (Curtis and Mont, 2020). These are just a few examples among many others that have adopted a tech-oriented perspective on the sharing economy.

The definitional conundrum issue
Concepts change and evolve. We may move from consumers to co-producers, from providers to resource owners (Curtis and Mont, 2020) or service providers (Andreassen et al., 2018;Benoit et al., 2017) and from obtainers to resources users (Curtis and Mont, 2020) or consumers (Andreassen et al., 2018;Benoit et al., 2017). Yet, that evolving loghorrea appears to refer consistently to the same fundamental principle of the collaborative economy. That is, consumer-to-consumer or peer-to-peer exchange that may be mediated by some kind of "middleman" whether it is a platform, an application or any other business model that creates value through facilitating the matching between provision and obtainment. This conceptualization matches closely the one we developed in 2016 and prolonged in 2019. Curtis and Mont (2020) further suggest that this switchover feature is actually a key condition for improved sustainability performance in the CE. They state: "We suggest SEBMs for sustainability operate as a platform that leverages technology to facilitate a twosided market between a resource owner and resource user. As such, this condition excludes businessto-consumer models that do not operate a two-sided market." (p. 6).
They nonetheless include business-to-peer models that operate as "two-sided markets", also called triadic business models that facilitate (rather than create) value as an outcome of the interaction between the supply and the demand side of an exchange system (Andreassen et al., 2018;Rysman, 2009). The authors mention "Spacious", a co-working platform outfit that offers space in restaurants during the day as an example of a two-sided market that can thus be considered to be part of the sharing economy.
Yet, two-sidedness should not be the exclusive criterion to assess whether a business is collaborative or not. This discussion goes obviously beyond the authors' original purpose since they merely aimed at improving sustainability in collaborative business models. However, what Spacious and other co-working platforms basically offer is to occupy otherwise vacant commercial spaces (e.g., restaurants only open for dinner, retailers closed on specific hours) to create cost-effective working units that generate additional income for landlords (Brown, 2020). If we refer to our original conceptual framework of the CE, Spacious is ambivalent since it may or may not constitute a form of CC depending on its configuration. More specifically: 1) If the platform enables a consumer to lend space to another consumer (C2C) or even a company (C2B), the platform conforms to the defining criterion of collaborativity (Ertz et al., 2016(Ertz et al., , 2019 by being a resource circulation system that enables consumers to switch roles from obtainers to providers. 2) If the platform allows solely private companies to lend space to consumers or businesses, the configuration is B2C and B2B respectively, and aligns with conventional consumption.
Since Spacious like many other sharing platforms allow both, they are intermittently collaborative. In other words, they enable conventional exchange configurations by allowing companies to provide resources to users (B2C) or companies (B2B). Yet, since those platforms also enable consumers to switch over (from user to provi-der), they may simultaneously be deemed collaborative. The same holds for other giants of the sharing economy including Uber, Airbnb, Instacart and so on. Meelen and Frenken (2015) drew the distinction between peer and professional providers. Non-professional or peer providers supply resources that they personally own such as their own home, space, car, tools, or skills to others. In contrast, professional providers supply resources that belong to the commercial entity that they represent. As a concrete manifestation of that tension between collaborativity involving peer supply and non-collaborative configurations referring to professional supply, Gyódi (2019) revealed that only a small minority of Airbnb listings can be classified as what he calls "sharing economy services", in the sense of peer hosting (thus involving collaborativity), while commercial offers, in the sense of professional hosting, constitute a significant share of listings on the platform: "The offers that represent sharing economy services constituted a minority of the offers in all of the analysed cities. Our results confirm the assumption that a significant share of Airbnb offers are listed by professional businesses, and may therefore contribute to gentrification" (p. 537).
It is therefore more accurate to frame what the literature has come to monolithically call the "sharing economy" as a general umbrella term (Hamari et al., 2015;Dreyer et al., 2017;Guyader and Piscicelli, 2019), which comprises three forms of exchange with varying degrees of collaborativity as shown in Table 1.
The remainder of this article will describe each of those three prototypical forms of collaborativity in more details. Importantly, while our work has progressed since our earlier definition of collaborativity, we maintain that the sharing economy in general, and the collaborative economy, in particular, may still be deployed across both purely online or offline channels as well as in a multichannel perspective; while involving diverse forms of compensation including but without being limited to money. This includes therefore flea markets, swapping meets, garage sales, secondhand stores, donation centers, trade-in schemes, and many other similar collaborativity-fuelled exchange schemes.
2. The three shades of collaborativity 2.1. Pure collaboration The notion of "pure collaboration" remains partially unchanged since our initial categorization (Ertz et al., 2016(Ertz et al., , 2019. First, it still refers to direct peer-to-peer (or consum-er-to-consumer) exchanges in which consumers directly exchange resources with each other or access to each other's resources. This we call "direct collaboration". However, in contrast to our past conceptualization in which we stressed that purity occurs "without any intermediary, whether facilitator or mediator" (p. 30) (e.g., garage sales, swap meets, flea markets), we stress that intermediation may in fact occur and is increasingly occurring to facilitate exchanges and make them more efficient (Curtis and Lehner, 2019;Curtis and Mont, 2020). Consequently, the two forms of "facilitated collaboration" and "mediated collaboration", that we framed in addition to pure collaboration, could actually conflate with pure collaboration. Pure collaboration thus entails by order of collaborativity, direct collaboration, facilitated collaboration and mediated collaboration. The latter two are described in more details as follows: -Facilitated collaboration involves P2P exchanges that are facilitated by an intermediary. For example, "the Canadian-based platform TheCarpoolingNetwork.ca enables consumers who wish to obtain a ride to connect with consumers who are willing to offer a ride, and to arrange between themselves the terms and conditions related to that ride, with the website acting merely as a facilitator" (Ertz et al., 2019, p. 30). -Mediated collaboration involves exchanges in which organizations not only facilitate but create the value creation process, thus acting as mediators. The mediation process occurs in two steps. First, through sourcing collaboration, the mediator relies on consumers who provide a resource or service (Ertz et al., 2019). For example, a telecommunication provider may act as a mediator sourcing collaboration when a consumer trades in a used mobile phone. Second, through trading collaboration, the mediator redistributes the resource or service to would-be obtainers, usually by adding some value such as reconditioning, terms and conditions which increase service quality, and so on (Ertz et al., 2019). In our previous example, the telecommunication provider adds value by refurbishing the mobile phone before reselling it to consumers (i.e., obtainers).
The frontier between facilitated and mediated collaboration may often be thin and merge especially in an online context where platforms tend to do much more than merely facilitating exchange between providers and obtainers. For example, Uber not only acts as a matching platform but imposes also a broad range of terms and conditions to its drivers regarding the type of car being used, the offering of water bottles and sweets, or even the behavior etiquette, and the pricing scheme with its surge pricing algorithm, all in an attempt to improve service quality and thus contribute to the value creation process.
Besides, finding an organization or platform working solely on the principle of pure collaboration is challenging especially in an online context where intermittent collaboration tends to prevail.

Intermittent collaboration
Intermittent collaboration refers to resource circulation systems which combine both pure collaboration and conventional consumption. The notion of conventional consumption might be unsettling here but it relies fundamentally on the modernist distinction between the production/distribution world and the consumption one. More concretely, that divide refers to conventional commercial exchanges involving organizations selling to consumers (B2C) or inter-organizational commerce (B2B). Initially, we excluded this second form of our conceptionalization of CC as it is antithetical to collaborativity (Ertz et al., 2016). From an essentialist perspective, the divide implied by modernism renders the possibility of switchover, that characterizes collaborativity, impossible. We maintain that collaborativity refers to the switch. Meanwhile, intermittent collaborativity blends in with the recent advances that have been made in marketing, consumer behavior, and management drawing on anthropological and sociological literature, acknowledging the hybrid or dual essence of the economy. In sum, the format / modernist perspective that makes that clear distinction between the production work and the consumption world (i.e., conventional consumption) is inevitably coupled with a postmodernist perspective in which these distinctions blur, allowing for an intertwining of the classic exchange paradigms of gift-giving, barter and commodity exchange, rather than a strict distinction between those (Scaraboto, 2015;Corciolani and Dalli, 2014;Dalli and Corciolani, 2008;Perren and Kozinets, 2018). Any entity (consumer or organization) has inherently a "deficit" (demand) and a "surplus" (supply) side which constantly interact in the form of active exchange of resources (Breyer, 1934;Shaw, 2014;Ertz and Sarigöllü, 2019). A company's surplus side thus fulfills a consumer's deficit side, while that same consumer's own surplus side may fulfill another consumer's or organization's deficit side through a breadth of exchange configurations including gift-giving, reselling, swapping, and so on (Ertz and Sarigöllü, 2019).
Intermittent collaboration reconciles therefore the consumer-business and commercial-disinterested tensions that have been observed in the sharing economy (Slee, 2015;Morozov, 2013;Murillo, Buckland and Val, 2017) by acknowledging that an intermingling of exchange configurations (i.e., C2C, C2B, B2C, B2B) as well as compensations forms (i.e., without compensation, with money compensation, with other compensations) might occur and actually constitutes the most accurate conceptual mirroring of praxis.
It should however be stressed that intermittent collaboration necessitates the allowing for collaborativity, in the sense of consumers switching over roles from users to providers. Spacious, Uber, Airbnb, Boatsetter, Peerby, Craigslist, indeed allow for consumers to become providers while simultaneously providing space for conventional companies, be they hotels, professional drivers, car lenders, boat lenders, retailers or manufacturers, to do exactly the same. Yet, in many cases the switchover is denied and this leads us to the third and last component of the sharing economy, pseudo collaboration.

Pseudo collaboration
Pseudo collaboration refers to resource circulation systems that do not enable (at all) consumers switching over from their current consumer role to that of a provider. As such, it incarnates the modernist exchange paradigm in its purest form. Many organizations that entered the market at the beginning of the sharing economy boom such as carsharing businesses Car2Go or Zipcar, but also bike-sharing Mobike, Ofo, OBike, or Lime, have been wrongly framed as part of the "sharing economy", while in fact denying consumers to switch over. Seminal authors Botsman, Rogers or Gansky and others classified such organizations as collaborative on the basis that they allow short-term rentals and could thus be equated with "sharing" and "collaborative" business models. Yet, how would such businesses then compare to conventional car lenders such as Avis or Enterprise? The same can be said for bike-sharing service providers. That misclassification has been abundantly discussed in the literature (Slee, 2015;Schor, 2016;Ertz and Leblanc-Proulx, 2018;Frenken and Schor, 2019) and reiterating what has already been said is purposeless and goes beyond the scope of this paper. Other authors in the likes of Bardhi and Eckhardt (2012) adopted a more nuanced position by framing such exchange schemes as "access-based consumption". Lamberton and Rose (2012) called them "marketing-managed sharing systems". These are all more appropriate characterizations of those organizations. This paper takes a broader orientation by using the concept of "pseudo collaboration". Importantly, pseudo collaboration should not be confounded with pseudo sharing. According to Belk (2014), pseudo-sharing refers to commodity exchange wrapped in the more acceptable and appealing terminology of sharing. More specifically, business transactions involving money flows are masqueraded as forms of communal sharing (Belk, 2014;Eckhardt and Bardhi, 2015). Likewise, Martin (2016) denounced the framing of the sharing economy as an economic innovation and a pathway to more sustainability in consumption while actually creating highly unregulated marketplaces. Morozov (2013) talked of neoliberalism under steroids. That disguise of the neoliberal wolf in the altruistic sheep's clothes is pseudo sharing but it differs from pseudo collaboration. In fact, the pseudo sharing construct is especially disturbing for scholars framing the CE as being essentially about sharingrelated practices such as lending, leasing, pooling and obviously sharing, which we summarize under the broader mutualization concept. Table 2 explains the differences between these different mutualization concepts. Yet, our original conceptualization of CC placed mutualization and redistribution as being both key components of the CE (Ertz et al., 2019, p. 33). Besides, exchange can be free through donations (Giesler, 2006), charity donation (Bergadaà, 2006), legacies and bequests (Curasi et al., 2004); for points of other currency such as on Barterquest (Botsman and Rogers, 2010), and trade in programs (vouchers, credits) (Eisend, 2015); for other items through swapping with swap meets (Belk et al., 1988) or local exchange trading schemes, time banks, and community currencies (Seyfang and Longhurst, 2013); subscription-based such as in tool libraries (Lamberton and Rose, 2012); or at a cost such as in secondhand marketplaces (Guiot and Roux, 2010), including auction websites (Chu and Liao, 2007) or private sales by the government.
Because we do not take a restricted approach focused on the sharing notion with all the semantic connotations that come inherently with it, our conceptualization is not hindered by pseudo sharing issues. It might however be surprising to consider pseudo collaboration as part of the sharing economy. This is because several papers including seminal ones conflated such types of exchange (see above for references) with the sharing economy (e.g., Hamari et al., 2015). Therefore, as an attempt to circumscribe the current literature, we posit impure collaboration in the form of pseudo collaboration as a third layer of the sharing economy. Yet, that layer does not overlap with the collaborative economy which remains distinct. As shown in Figure 1, the collaborative economy encompasses pure collaboration, obviously, but also that facet of intermittent collaboration which entails pure forms of collaboration. The sharing economy concept, in contrast, subsumes the three components of pure collaboration, intermittent collaboration and pseudo collaboration.

Perspectives for the future
What has come to be known as sharing platforms relies on a triadic perspective which involves three actors (Hawlitschek et al., 2018): -Suppliers: these may be hosts on Airbnb, drivers on Uber or car owners on Turo. They offer a specific resource, that is generally underutilized or unused; -Users: these may be the guests on Airbnb or the passengers on Uber. They seek to access either temporarily or permanently the proposed resource through various means. -Collaborative platforms: they act as an infrastructure for the mutualization or redistribution of resources between obtainers, as well as the maintenance of trust for the realization of exchanges. -Yet, that triadic equilibrium could be threatened in the future with profound implications for collaborativity and the collaborative economy. These shifts could in fact reshape our views on the future of the collaborative economy and the broader sharing economy.
The most crucial aspect in this regard deals with the diminishing role of peers as providers. While Gyódi (2019) already emphasized that peer providers are marginal in comparison with business providers, at least on Airbnb, these findings signal potentially an underlying trend towards the expelling of the peer as a provider. Technology, which contributed to empower individuals through peer-to-peer exchange, could backlash. With the rapid development of technological innovation, an increasing number of platforms seek to replace providers, be they peers or businesses, by machines. The most telling example in this regard refers to Uber's recent attempts to develop selfdriving systems which include sensors that may view 360 degrees around the car (Uber ATG, 2020). Automation could further impact a host of jobs that burst with the emergence of platforms. For example, the delivery men of restaurant food delivery services and more broadly the Q-commerce (Lange, 2020) such Uber Eats, Foodora, or Deliveroo, but also Postmates and Instacart could well be replaced by cheaper robots or drones, and of course, self-driving cars and trucks. Q-commerce is the abbreviation of "quick commerce" which is being considered as the next generation of e-commerce. It draws on the platform economy business model to bring small quantities of goods to customers almost instantly (Lange, 2020). What value could then peer providers put forth to compete with machines that can work 24/7 at a fraction of their own price? The peer providers may therefore survive in niche markets where relationality in the form of human contact remains essential such as teaching, coaching or training for example. They may also survive in the provision of higher-end premium services. Machines will essentially deliver cheaper services while having the opportunity to deal with a human will be highly priced. It is also worth mentioning that the receding of peer providers will inevitably dampen pure collaboration and the collaborative economy while contributing to the incessant growth of intermittent collaboration and pseudo collaboration.

The disappearing platform?
With all these developments it might seem limpid that platforms will thrive at the expense of peer providers. Yet, nothing is more certain either. As an inherently connected economy, the sharing economy and collaborative economy is prone to integrate the most recent technological advances such as artificial intelligence, the smart grid, big data analysis, augmented reality, as well as blockchain technology. While most of these technologies will essentially contribute to render providers less useful, the blockchain technology might actually make the platform itself redundant (Ertz and Boily, 2019). In fact, the blockchain is a decentralized, peer-to-peer system that gives access to a global audience of obtainers and providers (Beck et al., 2018). As such, it could re-empower peers who feel increasingly minored by qua-si-monopolistic giants such as platforms (e.g., Didi, Uber, Airbnb) and the GAFAM (Google, Amazon, Facebook, Apple and Microsoft) players offering intermittent collaborative solutions such as Facebook Marketplace or Amazon Prime, but also eBay and many others. Such attempts have already been made. For example, in 2017, engineers working on blockchain development for the Ethereum cryptocurrency founded Swarm City, a decentralized CE application which enables any individual to create his or her own brand at home without any intervention from a third party (i.e., a platform) (Beck et al., 2018). While several questions remain regarding the blockchain governance and whether it will remain decentralized and disintermediated as it is today (Campbell-Verduyn, 2018;Arsenault and Ertz, 2019). However, the key features of openness, decentralization and transparency of the blockchain surely make it a competitor platform that consistently struggles with legal, governance, industrial relations, and tax issues. Following the simple marketing rule, that any middleman that does not create value eventually becomes redundant and disappears (Breyer, 1934), it might be worth questioning to what extent the current platform model can sustain the competition of blockchain applications in the CE and SE overall.
In fine, users are the only one component who might not disappear. As stressed earlier, peers and businesses with deficit poles (demand) will always have needs to satisfy and will always exist. However, the recoil of either peer providers or platforms or both altogether signals a potential rearrangement of demand as well. Some obtainers are motivated by authenticity in collaborative exchange (Paulauskaite et al., 2017;Bucher et al., 2018) and may feel resentment when faced with a rising amount of business providers and offerings on supposedly "sharing schemes". The market could thus shrink to some extent but it is very unclear to what extent that shrinkage could be detrimental to platforms. However, since the critical mass remains an important driver of success in the collaborative economy (Tomasello, 2009), anything that makes that mass tinier should be of serious concern.

Conclusion
This paper sought to shed some light on the concept of the collaborative economy which was put in relation to the sharing economy as well as other nomologically-related concepts. Almost half a decade has passed after an initial conceptualization of CC and the collaborative economy as being characterized by the key criterion of the switchover opportunity, meaning that a resource circulation system is collaborative as long as it enables consumers to switch roles and become providers themselves (Ertz et al., 2016). That conceptualization remains valid today as well as the various channels (from purely online to purely offline), practices (from gift-giving to secondhand exchange) and forms of compensations (from free exchange to exchange at a cost) that were associated with CC (Ertz et al., 2019). However, in an attempt to better match the exponential research that has been conducted on the broader topic of the "sharing economy" and to most accurately reflect praxis, this study proposes that the sharing economy notion should be considered as an umbrella term that can be divided into three shades that often overlap.
First, pure collaboration involves genuine peer-to-peer exchange, either directly or facilitated by an intermediary, or completely mediated by an intermediary. Pure collaboration is the closest conceptualization of what was originally conceived of as "sharing economy services" as supplied by peers instead of businesses (Gyódi, 2019). Pure collaboration thus encompasses the subcomponents of direct collaboration, facilitated collaboration and mediated collaboration (which includes sourcing and trading collaboration). Second, intermittent collaboration refers to two-sided markets in which both pure collaboration (C2C and C2B) and conventional collaboration (B2C and B2B) occurs. This form of collaboration is in fact the most prevalent one especially on platforms that adapt well to both consumer-to-consumer and business-to-consumer exchange. Finally, pseudo collaboration is mere conventional consumption as partly enacted in intermittent collaboration. While both pure collaboration and the collaborative component of intermittent collaboration form the collaborative economy, the sharing economy subsumes these three shades altogether.
The "platform" has become the increasingly prevalent embodiment of the sharing economy, both in theory and in practice. In fact, sharing economy equates increasingly with a web-mediated and platform-centric economy. This triadic configuration posits providers, platforms and users as the three founding poles. Yet, providers are less and less peers or even human, while platforms could be substituted for more open and transparent systems such as the blockchain, and thus recede from their central position. This also raises the question as to what will remain of collaborativity when peers will be significantly reduced to the point of becoming marginal, or worse, disappear. However, some niche markets could subsist in specific sectors and for premium or luxurious services. With the ebbing of both providers and platforms, the user base might mechanically decline. The maintenance of the critical mass will need to be checked closely for future sustainability.