Buying to share: How prosumption promotes purchases in peer-to-peer asset sharing

Advocates of the sharing economy cite sharing as a viable alternative to asset purchases and ownership. However, Peer-to-peer (P2P) asset sharing, as a service innovation in the sharing economy, enables consumers to capitalize on their asset ownership by providing others with access to those assets for a fee. These prosumers acquire and consume the asset but also provide it as a service sold to others. In exploring the connection between prosumers and asset manufacturers, this study particularly notes the implications of prosumption for initial asset acquisition. A review of existing P2P asset sharing initiatives, three focus groups


Introduction
In the globally growing sharing economy, manifold manifestations of sharing services have emerged in recent years (Eckhardt et al., 2019).In the United States alone, an estimated 80 million consumers engage in sharing-almost a 100% increase compared with previous years (Statista, 2019).This trend also has been fueled by the increasing availability of access-based services in industries such as carsharing (e.g., Share Now), fashion (e.g., Bag Borrow or Steal), and hospitality (e.g., Airbnb).Extant research commonly regards sharing services-such that a provider grants consumers access to an asset for a predefined time and in return for some fee-as an alternative to asset purchases (Lawson et al., 2016) that enable consumers to reduce the financial burdens of ownership (Schaefers et al., 2016).Therefore, sharing might represent a threat to established business models, such as manufacturers' conventional approach to selling assets for private usage (Eckhardt et al., 2019).
In an effort to address this threat, some manufacturers entered the sharing economy with their own access-based services, but they seem to struggle with monetizing their offers and dealing with the costs of owning and maintaining sharing assets.For example, faced with the risk of failure, BMW and Daimler merged their deficient carsharing initiatives in an attempt to achieve profitability (Daimler, 2019).
Peer-to-peer (P2P) asset sharing, as a service innovation in the sharing economy, might provide manufacturers an alternative means to participate.In P2P asset sharing, consumers offer other consumers (i.e., users) access to idle, capacity-constrained assets, with the assistance of dedicated platforms (Wirtz et al., 2019).The underutilized assets remain owned, maintained, and managed by consumers (Benoit et al., 2017), and the sharing platform provider enables and governs the exchange and value creation process (e.g., searching, booking, payment, platform advertisement).Consumers as asset providers usually get reimbursed with a fee, paid by the asset users (Benoit et al. 2017).Thus, the P2P asset sharing platforms empower consumers who acquire an asset for their own consumption to produce a service too, by temporarily providing others access to underutilized assets.This dual role, as both asset provider and consumer, takes the label prosumer (Eckhardt et al., 2019).
Considering the growing popularity of P2P asset sharing, the connection between manufacturers and prosumers also may be critical, because the opportunity to prosume (i.e., acquire, consume, provide) requires some asset purchases, which implies a potential opportunity for manufacturers.In other words, instead of solely reducing asset sales, because it offers alternative access options to users (Bardhi & Eckhardt, 2012;Fritze et al., 2020), the sharing economy might foster additional sales from manufacturers to prosumers.However, we lack any research guidance or insights into the ramifications for asset manufacturers that aim to sell assets to prosumers that participate in the service innovation of P2P asset sharing.
Therefore, this research applies a mixed-methods approach to investigate the connection between manufacturers' and consumers' roles within P2P asset sharing services and the implications for asset purchases.That is, we address whether and how prosumption promotion by manufacturers might affect consumers' purchase decisions.In a first step, we outline prosumption as our theoretical foundation for exploring the effects of P2P asset sharing on asset manufacturers.Existing research offers limited insights into the effect of prosumption on initial asset purchases and the prosumer's associated motives, so exploratory efforts are needed to gather insights for hypotheses development.We conduct three focus groups to explore the relation between prosumers and manufacturers in P2P asset sharing and to unveil how consumers' purchase behaviors shift with opportunities to prosume.Building on the focus group insights, we adopt mental accounting theory to develop hypotheses about the effect of prosumption on asset purchase behavior.Finally, we conduct two experimental studies to test the outlined hypotheses empirically and derive implications for asset manufacturers that seek to leverage asset purchases from prosumers in P2P asset sharing.
Exploring this connection establishes several contributions to research and practice.First, by analyzing manufacturers' and consumers' roles within P2P asset sharing, we add to a dominant, triadic model that encompasses only the sharing platform, the asset provider, and the asset user (Benoit et al., 2017).We introduce the manufacturer as an active contributor.In turn, this study extends current understanding of P2P asset sharing, beyond the provision and use of idle capacities, to encompass active asset purchases (Eckhardt et al., 2019).This novel perspective on the sharing economy goes beyond the current focus on sharing transactions (Bardhi & Eckhardt, 2012;Belk, 2014;Dellaert, 2019;Lamberton & Rose, 2012).
Second, drawing from mental accounting theory (Thaler, 1985), we show empirically that the burdens of ownership represent an important psychological mechanism that links prosumption to asset purchases.This perspective contributes to literature that identifies reduced burdens of ownership as a key motive for forgoing ownership and using sharing services (Moeller & Wittkowski, 2010;Schaefers et al., 2016).We theorize and empirically confirm that prosumption is an alternative way to reduce the burdens of ownership, which simultaneously results in asset purchases.
Third, we provide insights on crucial interdependencies among manufacturers, prosumers, and users (e.g., Benoit et al., 2017;Ertz et al., 2018).Particularly, we identify a potential conflict between the prosumer's varying roles as acquirer, user, and provider of the same asset.The initial purchase decision rests on prosumers' own brand preferences but also on the brand preference of users of the sharing platform.
Thereby, we answer calls by Eckhardt et al. (2019) to assess the impacts of prosumption opportunities for asset manufacturers, and we offer recommendations for how to cater to different consumers' preferences.For managers and asset manufacturers in particular, our findings illustrate that prosumption in P2P asset sharing (1) is a sales opportunity, (2) enables upselling, and (3) requires careful management of the interdependency between prosumers and users to achieve successful asset sales.

Peer-to-peer asset sharing
Previous research also refers to P2P asset sharing as collaborative consumption (Benoit et al., 2017) or peer-to-peer product sharing (Benjaafar et al., 2018).Another stream of literature links P2P asset sharing to lateral exchange markets (Perren & Kozinets, 2018) or consumer co-production networks (Dellaert, 2019).Despite the multitude of terms, a common understanding establishes that in P2P asset sharing, consumers own, use, maintain, and manage assets that fellow consumers (i.e., users of sharing platforms) access for a given period of time (Benoit et al., 2017;Ertz et al., 2018).The sharing platforms mediate the exchange between asset provider and user but do not own the asset (Wirtz et al., 2019).This triadic relationship among asset providers, sharing platforms, and peer users (Benoit et al., 2017) distinguishes P2P asset sharing from access-based services that provide consumers with temporary access to physical assets in return for fees, while legal ownership of the asset remains with the service provider and not another consumer (Bardhi & Eckhardt, 2012;Hazée et al., 2019).We also differentiate P2P asset sharing from microentrepreneurship in the sharing economy, according to usage value and scope.A microentrepreneur professionally buys or owns (multiple, similar) assets for others' usage value (e.g., buys apartments to rent via AirBnB; also referred to as buy-to-let; Cocola-Gant & Gago, 2019).A provider in P2P asset sharing instead amateurishly owns (single, different) assets for their own usage value and offers access only when that usage value is low (e.g., rent out their car when they are not using it; Benjaafar et al., 2018).
Prior studies of P2P asset sharing mainly feature conceptual analyses that delineate it from other consumption forms such as buying, renting, or access-based services (Benoit et al., 2017;Ertz et al., 2018), or else actor-specific analyses of asset providers' (Philip et al., 2015;Wilhelms et al., 2017) or users' (Hawlitschek et al., 2018;Münzel et al., 2019) participation motives.A few platform-focused analyses also assess exchanges between asset providers and users (e.g., Costello & Reczek, 2020;Perren & Kozinets, 2018).These varied research streams offer a common, shared focus on the triadic exchange relationships among asset providers, sharing platforms, and peer users.Our research instead zeroes in on the asset provider's initial decision to acquire an asset, for the twofold purpose of using it themselves and providing it to others at times marked by low own usage value.We thereby expand the P2P asset sharing triad by including the manufacturer as an active player, as illustrated in Fig. 1.We account specifically for the interaction between manufacturers and consumers who acquire an asset to use themselves and also provide to others through P2P asset sharing.This expansion acknowledges the consumers' multiple potential roles, as acquirer, consumer, and provider of the same asset, an interplay that is conceptually close to the underlying idea of prosumption.

The concept of prosumption
The original prosumer definition merges the terms "producer" and "consumer" to describe someone who produces an asset for their own consumption (Toffler, 1980).However, its connotation has shifted, from producing for one's own consumption to also producing value for others (Ritzer & Jurgenson, 2010;Seran & Izvercian, 2014;Xie et al., 2008).With the rise of P2P asset sharing, consumers took on expanded roles in which they offer underutilized goods for sharing purposes (Eckhardt et al., 2019).The prosumer term, applied to this "new" type of consumers, extends to include product or service provision to other consumers in sharing contexts (Barnes & Mattsson, 2016;Costello & Reczek, 2020).In turn, the sharing economy gave rise to two meanings of a prosumer.First, it refers to a person functioning as an asset provider or a consumer at different points in time (e.g., an Airbnb host is a guest at a different house for another occasion; Benoit et al., 2017;Wirtz et al., 2019).Second, and more in line with the original prosumer sense, Eckhardt et al. (2019) cite the example of "ride-sharing providers (that) 'consume' their car and also 'produce' a service for those who ride along" (p.5).This latter version of prosumption, which is the focus of the current study, defines prosumers in P2P asset sharing as asset owners who consume their own assets and also produce a service by temporarily providing others access to their unused assets through a sharing platform.
Despite acknowledging prosumers, current P2P asset sharing research often takes a two-tier approach, exploring consumers as either asset providers (e.g., Costello & Reczek, 2020) or users (e.g., Morewedge et al., 2021).The assets to be shared are mostly designated as preexisting or "idle capacity," such as unused cars (Frenken & Schor, 2017).However, the prosumer role in P2P asset sharing goes further, because the prosumer acts as an acquirer, consumer, and provider of the asset in a sharing ecosystem (Eckhardt et al. 2019).This aspect differentiates prosumers from regular customers that merely consume the asset themselves and microentrepreneurs that merely use assets for service provision.
In turn, we need to understand prosumers' initial decision to purchase an asset that serves their own consumption needs but also is relevant to others, along with any resulting implications for manufacturers.Existing research provides little guidance and no empirical evidence of how prosumption influences asset purchase from manufactures, so we start with an exploratory study to understand the underlying motives, as well as clarify how the dual role, consuming and providing, might inform purchase decisions.We build on these insights to derive and experimentally test hypotheses about the connection between manufacturers' and consumers' roles within P2P asset sharing services and its implications for asset purchases from manufacturers.

Peer-to-peer asset sharing initiatives and links to prosumption
In a first step, we explore the connection between manufacturers and prosumers, based on recent developments in practice.We focus on the automotive industry, which has been significantly affected by sharing considerations (Barnes & Mattsson, 2017).Manufacturers already are conducting tests and pilot programs to reach out to prosumers and connect asset purchases with consumption and provision of the asset for P2P car sharing.Table 1 provides an overview of some recent initiatives.
Many programs highlight the capacity to reduce vehicle ownership costs by engaging in asset provision.They target current vehicle owners but also actively aim to incentivize prospective buyers.Simultaneously, communication efforts by platforms emphasize the economic incentives of engaging in P2P asset sharing as a prosumer, such as by proposing, "Own a car?Share on Getaround and earn $1000 s" (Getaround), "Turo hosts can cover their payments by sharing their cars just nine days per month" (Turo), and "Let your car earn money!" (Drivy).The review of these managerial cases suggests a general belief that (1) acting as a prosumer on P2P asset sharing platforms may foster asset purchase and (2) the opportunity to earn financial benefits could be of relevance to prosumers.However, these tests and pilot projects are still in their infancies and cannot be linked to any significant share of the car sales market; instead, investigations of whether such a strategy can be successful and how prosumers need to be addressed are required.Only two extant modeling studies consider purchase and provision (Benjaafar et al., 2018;Jiang & Tian, 2018), but they focus on welfare creation, not explicit promotional messages.Considering the limited guidance and research on the role of prosumption in P2P asset sharing, we decided to conduct exploratory focus group research.

Focus groups to explore the connection between prosumers and manufacturers
We conducted three focus group interviews to investigate consumer perceptions of a prosumption opportunity in an asset purchase context and answer the question: How do customers assess an offer from a manufacturer to purchase an asset, as a consumer (i.e., acquire an asset for own consumption) versus a prosumer (acquire an asset for own consumption and for provision to others via a sharing platform)?The focus group approach fits well with our research context, because the group interaction (Albrecht et al., 1993) and replication of social forces (Krueger & Casey, 2014) are valuable elements in generating insights in areas where limited prior knowledge exists (Chisnall, 1997;Stokes & Bergin, 2006;Zikmund, 2000).Focus groups can identify overarching themes (Krueger & Casey, 2014) and thus have frequently been adopted for services research (e.g., Davis et al., 2000;McColl-Kennedy & Sparks, 2003).

Sample characteristics
To recruit interviewees for the focus group study, purposeful sampling was applied by asking participants to participate in an academic market research study (Dagger et al., 2007).Specifically, our sample includes 15 German licensed drivers (seven women) with an average age of 41.5 years (SD = 16.5), who were allocated to one of three focus groups.Table 2 provides an overview of the sample.We explicitly sampled car owners familiar with a car acquisition decision but without prosumption experience.None of the participants was a member of any P2P asset sharing organization.Thereby, we aimed at reducing biases from previous experience with P2P asset-sharing services.The focus groups also consisted of car owners of similar ages, which ensured high exogenous homogeneity and low issue homogeneity, to reduce communication barriers and enable a lively discussion (Greenbaum, 1998;Krueger & Casey, 2014).

Data collection
The focus group guide, developed according to guidelines published by Greenbaum (1998) and Morgan (1997), consisted of general introductory information and a list of open-ended discussion questions, grouped into categories, from general to more specific, to aid the flow of discussion (Steward & Shamdasani, 1990;Zikmund, 2000).The topic of the focus group discussions was P2P carsharing, a specific case of P2P asset sharing, where the shared asset is a car.Carsharing represents one of the most advanced and promising markets for asset-based sharing in general (see also Hazée et al., 2019;Lawson et al., 2016) and P2P asset sharing in particular (Barnes & Mattsson, 2017).As our review of managerial cases underlines, carsharing has particular relevance for P2P asset sharing, and this specific focus allowed for a diverse discussion of real-life cases; Eckhardt et al. (2019) similarly refer to the automotive industry as exemplary of the sharing economy.The focus group interviews started with initial information about the process and goal of the academic study.A discussion of participants' daily mobility habits and previous/future car purchases followed, before the concept of P2P asset sharing in general and the specific context of P2P carsharing were introduced.Subsequently, the groups discussed the benefits of P2P asset sharing in general and carsharing in particular, then reviewed the motives for and barriers to acting as a provider.To this end, we asked the participants to evaluate programs by Getaround 1 and vehicle manufacturers that connect asset purchase offers to asset provision, through participation in Getaround's P2P asset sharing platform (Getaround, 2019), to assess the prosumer role.
To steer the discussion and fully observe the participants' interaction, one of the authors and a research assistant acted as moderators.This procedure also ensured that participants had a profound understanding of the research topic (McDonald, 1993;Sim, 1998).These moderators specifically encouraged group interactions and different means of conversation to create a naturalistic flow and generate meaningful insights into interviewees' attitudes toward the research context (Kamberelis & Dimitriadis, 2013;Wilkinson, 2015).Overall, we conducted three focus group interviews lasting 75 min on average, following a single category design with five participants each (Krueger & Casey, 2014;Morgan, 1997).

Data analysis
The focus groups were video-recorded and transcribed.Two independent researchers analyzed the content (Malhotra & Birks, 2003;Morgan, 1997;Papista & Dimitriadis, 2012).We structured the data analysis according to a three-level process (Wolcott, 1994): (1) describe data based on verbatim quotes and assign codes, (2) analyze the data to identify overarching contexts and categories, and (3) interpret the data.To ensure internal consistency, the two researchers discussed and jointly attributed the findings to higher-order categories (Patton, 2002).A third researcher then reviewed the findings to ensure sufficient objectivity.In case of a disagreement, the issues were discussed until a consensus was reached, and the data were refined in accordance with the presented quotes (Dagger et al., 2007).No significantly "new properties, dimensions, or relationships" emerged in the analysis of the third focus group, indicating theoretical saturation, so we terminated the data collection (Strauss & Corbin, 1998, p. 143; see also Patton, 2014).Further triangulation was ensured by constantly comparing the emerging themes with existing literature (Strauss & Corbin, 1998).

Findings & discussion
The focus group participants cited three main benefits of P2P car sharing in general: economic, sustainability, and community.The sustainability and community benefits, such as positive environmental impacts, were mentioned but relegated to secondary relevance, especially in prosumption.This finding aligns with previous research into the sharing economy and P2P asset sharing (e.g., Philip et al., 2015;Wilhelms et al., 2017), and accordingly, these benefits were not included in the hypothesis development.The monetary benefit also was the most predominantly discussed benefit across all three focus groups, and four major subthemes emerged from the analysis: (1) reducing the costs of access and ownership, (2) lowering the acquisition price, (3) increasing purchasing power, and (4) buying a vehicle attractive to other users.
First, the participants acknowledged the opportunity for users to have cheaper access to vehicles and avoid further costs (e.g., maintenance, insurance), reflecting established themes in access-based consumption (Bardhi & Eckhardt, 2012;Schaefers et al., 2016).When the discussion moved to the concept of P2P asset provision and participants reviewed Getaround's P2P asset sharing platform, the focus of the discussion shifted to the opportunity to economize car ownership.That is, participants positioned asset provision as an opportunity to financially capitalize on asset ownership, particularly when compared with a traditional role as an asset owner.One participant stated outright: "By renting out my car, I get money.That means that the costs of owning a car are minimized" (ID11). 2econd, when presented with the explicit offer to buy a car and rent it out through a platform (prosumer offer), participants perceived the potential rental income as an additional value of an asset purchase that could compensate for the perceived losses caused by the financial burdens of ownership.One participant implicitly referred to it as a promotion, stating: "If I were to come across such an offer, yes, I would consider it.The dealer presents me with the cost of a car and an option to reduce it.That's like a discount" (ID9).
Third, participants explicitly linked their provision to an acquisition role and discussed how asset provision would increase their purchasing power and enable them to purchase a more expensive asset than they otherwise could afford.For example, one participant stated "One could choose a better-equipped car or afford ownership of a car in the first place because of the possibility of recuperating those costs by renting it out" (ID4), and another participant imagined "Renting out would allow me to purchase a bigger car, one that would normally be too expensive" (ID3).
Fourth, participants cited the benefits of purchasing assets that are especially desired by future renters.However, these discussions were motivated not by social motives but mainly by the ability to realize higher earnings and reduce the financial burdens of asset ownership to an even greater extent.One participant mused out loud that "If I take users' interests into account, I can simply make more money with my car than if I have a car nobody wants to rent" (ID14), and another went even further by stating that "I could even imagine buying a different car brand and type than the one I would normally buy if it would result in more rentals" (ID7).
Our findings indicate that asset provision in P2P asset sharing creates value for prosumers by representing an opportunity to economize on the costs of ownership and enriching purchasing power, thus making asset purchase and ownership more desirable.Monetary benefits emerge as the predominant driver that differentiates prosumers from traditional acquirers of an asset, because only prosumers anticipate economizing through future asset provision at the purchase stage.This crucial role of monetary benefits is in line with existing research (Philip et al., 2015;Wilhelms et al., 2017) and suggests that consumers seek to reduce the financial burdens of ownership (Moeller & Wittkowski, 2010;Schaefers et al., 2016).Further, it supports some assumptions about provider behavior that have appeared in proposed models of P2P asset sharing The Getraound app allows consumers to find P2P carsharing cars around them, unlock them (either remotely or by handing over the car key) and pay the access fee.For prosumers, it allows to offer P2P carsharing vehicles and charge an access fee from users.All transactions are fully insured by insurance partners (e.g., Allianz) for both parties.In 2021, The company was active in more than 800 cities globally and had more than 6 million registered users (Getaraound, 2021).(Benjaafar et al., 2018;Jiang & Tian, 2018).Building on this economic motive, participants describe balancing out the potential losses linked to buying a car against the potential gains of providing it for P2P asset sharing in the future.Moreover, they justify more expensive purchases by citing the gains that result from providing the acquired asset for monetary returns (i.e., gains from the same account).These considerations are akin to mental accounting processes.Thus, we build on mental accounting theory to develop hypotheses about the effect of prosumption on asset purchases from manufacturers in the following.

Hypotheses development
A core assumption of mental accounting is that consumers have an implicit accounting system in which they keep track of their resources and expenses under different labels (Thaler, 1985).Specifically, consumers code gains and losses, set budgets, and evaluate purchases depending on the mental account in which they occur.This process of mental accounting influences consumers' decision making (Heath & Soll, 1996;Kivetz, 1999).Our focus group findings indicate that participants evaluate the opportunity for prosumption in P2P asset sharing as a means to increase economic gains in the account assigned to asset purchases and ownership, especially when compared with a traditional consumer role.This gain is likely to remain in the same mental account, instead of being allocated to another account, so the asset purchase appears more economical from the beginning (Heath & Soll, 1996;Kivetz, 1999) and may explain prosumers' willingness to reinvest their economic gain in a purchase.
In line with prospect theory, consumers prefer to integrate losses and segregate gains within this accounting system (Thaler, 1985).The opportunity to engage in P2P asset provision segregates the gains related to an asset in two dimensions: value for personal use and value for renting out through a P2P asset sharing platform.This bi-dimensionality makes the purchase even more attractive.Thus, the effect of acting as a prosumer (vs.regular consumer) is likely to be even stronger for a more expensive than for a cheaper asset, in that acquiring more expensive assets can evoke intrapersonal conflict that requires justification and needs to be resolved with the aid of mental accounting (Kivetz, 1999).Reinvesting the additional economic gain from P2P asset provision for asset purchases might provide the desired justification for a more expensive alternative.Accordingly, we hypothesize: Hypothesis 1: Acting as a prosumer (vs.consumer) of an asset increases willingness to purchase.Hypothesis 2: The positive effect of acting as a prosumer (vs.consumer) on willingness to purchase is stronger for more expensive compared with cheaper assets.
Following a process of mental accounting, consumers commonly set the expenses for an account in advance of the actual consumption and track the expenses against their budget.Consumers tend to stick to their budget and resist transferring funds across accounts (Heath & Fennema, 1996;Heath & Soll, 1996).This budget includes all expenses associated with the asset purchase over its lifetime.That is, the budget includes the cost associated with asset ownership.The anticipated expenses linked to an asset purchase represent the financial burdens of ownership (Schaefers et al., 2016), which include the psychological burdens that consumers experience when deciding whether to acquire an asset.The burdens of ownership prevent consumers from buying and lead them to prefer renting (see also Moeller & Wittkowski, 2010), though this view may be one-sided.In P2P asset sharing settings, prosumers also can reduce the burdens of ownership by renting out purchased assets, instead of forgoing purchase completely.Thus, we propose that asset provision for P2P asset sharing increases prosumers' purchase intentions by reducing the burdens of ownership and offer the following mediation hypothesis: Hypothesis 3: The reduction of burdens of ownership mediates the positive effect of acting as a prosumer (vs.consumer) on willingness to purchase.
We also seek to investigate how the dependency between prosumers and users influences the prosumers' purchase decision.As outlined, prosumers use the opportunity to rent out an asset and its associated economic gains as a justification for its purchase (Kivetz, 1999).Because prosumers seek to reduce their burdens of ownership, they likely see asset purchases as investments for the future (Shafir & Thaler, 2006).However, due to the bi-dimensionality of gains (Thaler, 1985), the process of justifying the investment becomes more complex when potential gains depend not only on their own asset usage but also the anticipated economic gains of making the asset available to peer users on a P2P asset sharing platform.For a prosumer purchasing an asset, there are two competing benefits of ownership: (1) satisfaction through their own consumption (i.e., consumer role) and ( 2) financial returns from receiving money back due to prosumption (i.e., provider role).To justify the asset purchase as an investment, in their mental account (Shafir & Thaler, 2006), prosumers need to consider both their gains from personal usage and the likelihood of achieving economic gains through asset provision (i.e., two dimensions of gains; Thaler, 1985).As such, in a specific P2P asset sharing context, purchase decisions should reflect prosumers' own brand preferences but also predictions about which brand promises the most economic gains from P2P asset sharing provision.We thus hypothesize Hypothesis 4: Prosumers have a higher willingness to purchase brands that are preferred by users of peer-to-peer sharing services.

Method
Experiment 1 examines the effect of acting as a prosumer for P2P asset sharing services on consumers' purchase intentions and the mediating role of the burdens of ownership.It applies a 2 (customer role: prosumer vs. consumer) × 2 (asset price: cheap vs. expensive) betweensubject design.A total of 273 licensed U.S. drivers (mean age = 36.20 years, SD = 10.30;54.2% men) participated through an online panel.They were randomly allocated to one of the four conditions.Consistent with our exploratory study, we chose P2P carsharing as the contextual setting.All four groups viewed an advertisement from a car manufacturer, promoting a new car purchase.Participants in the P2P asset sharing conditions also read that they could earn money by renting out their purchased car on a P2P asset sharing platform.Participants in the different price conditions saw either a cheap or expensive car, of the same brand.The stimuli were designed in line with existing P2P carsharing offers (Table 1) and refined in accordance with feedback from three industry experts from the car manufacturing and sharing industries.The prosumption manipulation also was pretested with a convenience sample of 132 participants (M Prosumer = 5.38; M Consumer = 2.66; F(1, 130) = 103.502,p = .000).The appendix lists the final stimuli used.
After having seen the stimuli, participants reported their purchase intentions and completed the manipulation checks.Noting the focus group results, which emphasize the monetary benefits of purchasing an asset for prosumption as the predominant motive, we used the monetary benefit linked to the asset as a manipulation check in the customer role condition (prosumer vs. consumer) and the asset price as a check for the perceived asset price (Suri & Monroe, 2003).We also measured the financial burdens of ownership (Schaefers et al., 2016).All constructs were measured on seven-point Likert scales.Table 3 lists all the measures.

Results
Manipulation Check.Participants reported a higher perceived asset price in the expensive condition (M Expensive = 6.41,M Cheap = 4.17; F (1,271) = 297.355,p = .000)and correctly identified the customer roles, in accordance with the monetary benefit offered by the asset (M Prosumer = 5.32, M Consumer = 2.67; F(1,271) = 165.953,p = .000).
Main and Interaction Effects.In line with our expectations, the main effect of prosumption in P2P asset sharing services on participants' purchase intentions is positive and significant (M Prosumer = 3.47; M Con- sumer = 2.87; F(1,269) = 8.751, p = .003).We also find a significant main effect of price (F(1, 269) = 59.561; p = .000).Notably, we observe a significant interaction effect between the customer role and asset price on purchase intentions (F(1,269) = 4.373, p = .037).That is, the effect of prosumption on the participant's purchase intentions is stronger for the expensive asset (d = 0.60) compared with the cheap asset (d = 0.11), in support of H1 and H2 and as illustrated in Fig. 2.
Mediation Effect.We use a bias-corrected bootstrapping procedure to assess the mediation effect of burdens of ownership (Zhao et al., 2010).We use 10,000 bootstrap samples with a 95% confidence interval (CI) and indicator coding, with the groups "consumer" (vs.prosumer) and "cheap" (vs.expensive) as the baselines.Acting as a prosumer significantly decreases the burdens of ownership (=-0.30,p = .004);the asset price increases them, as expected (=0.96, p = .000).The lower the burdens of ownership, the higher the purchase intentions (=-0.55,p = .000).In line with our expectation, we observe a positive, indirect effect of prosumption through burdens of ownership on purchase intentions.The bias-corrected 95% CI for the indirect effect of the path through burdens of ownership is [0.059, 285] with a point estimate of 0.166.There is no significant direct effect of prosumption on purchase intentions when we include the mediator (=0.15, p > .1).Thus, the burdens of ownership fully mediate the effect of prosumption on intentions to purchase, in support of H3. 3 The indirect effect of purchase price on purchase intentions though the burdens of ownership also is negative and significant, as expected (-0.533, 95% CI [-0.688, -0.384]).

Method
Experiment 2 aims to provide insights into the relationship between prosumers and peer users by testing peer users' influence on prosumers' purchase decision.Specifically, we seek to test whether prosumers have a higher willingness to purchase brands that are preferred by users of P2P asset sharing services.To test it, we present a P2P asset sharing offer that makes the brand preference of peer users known to prosumers or not.To confirm the effect, we must test it across the potential range of a prosumer's brand preferences.That is, are prosumers more willing to purchase brands preferred by users if the brands also are the prosumers' own first, indifferent, or rejected choice?For this purpose, we use a 2 (peer users' brand preference: known vs. unknown to prosumer) × 3 (prosumer's brand preference: first choice, indifferent, rejected) between-subject design.
A total of 400 licensed U.S. drivers (mean age = 35.73years, SD = 10.50 years; 53.3% men) participated via an online panel and were randomly allocated to one of the six conditions.The stimuli were designed in line with Experiment 1, using P2P carsharing as the contextual setting.In the first part, respondents read that they were participating in a survey about cars, and they were asked to report their brand preferences from a list of the 10 most purchased brands in the U.S. market (indicating first, indifferent, and rejected choices).In the second part of the survey, all six groups received an offer to buy a new car, but they were randomly allocated to an asset offer that matched their first choice, indifferent, or rejected brand.Across peer user brand preference conditions, they saw the offer either with additional information that the majority of peer users preferred that particular asset (i.e., "80% of users report that they prefer renting [brand name] over any other brand") or without any peer users' brand preferences.After having seen the stimuli, participants reported their purchase intentions, as in Experiment 1.For the manipulation checks, they indicated whether they received information about peer users' preferences and how well the brand offered matched their brand preference (Rosengren & Dahlén, 2015).
Main and Interaction Effects.The main effect of prosumers' brand

Table 3
Overview of Measures.

Construct Measures Alpha
Purchase intention -How likely would you be to purchase the advertised [brand], given the information shown in the ad?Would you be more or less likely to purchase the advertised [brand], given the information shown in the ad?
Given the information shown in the ad, how probable is it that you would consider the purchase of the advertised [brand]?0.9641 0.9262

Monetary benefit
-Taking this offer gives me the chance… to earn money to pay for my car.
to generate additional income.to make extra money.

Attitude toward product price
-The advertised price for this [brand] was: very low-very high I felt that the [brand] was: very cheap-very expensive I felt that the advertised price for this [brand] was: very low-very high 0.950 1

Burdens of ownership
-I would worry about the cost of purchasing the advertised [brand].
Given the financial expenses associated with purchasing a car, there is a substantial financial risk.
Given the financial commitment, I may regret purchasing the advertised [brand].3).In other words, peer users' brand preferences influence prosumers' purchase decisions.Yet we did not find a significant interaction effect between prosumers' and peer users' brand preferences (F (2, 394) = 2.011, p > .1),so this effect holds independent of prosumers' actual brand preferences.These findings support H4.

Summary
The study objective was to investigate the connection between manufacturers' and consumers' roles for P2P asset sharing.Particularly, we assess the implications of consumer and provider roles, as conceptualized in accordance with the theory of prosumption, for asset purchase decisions.With an exploratory, qualitative study, we explore the motives that drive prosumption within P2P asset sharing settings.Consumers predominantly rely on monetary considerations when they act as prosumers and adopt a psychological mechanism akin to mental accounting processes.On the basis of these insights, we derive our hypotheses and test them with two experimental lab studies.As Experiment 1 illustrates, promoting prosumption (i.e., asset acquisition in combination with asset provision) rather than asset purchases increases consumers' purchase intentions, particularly for more expensive assets.With a mediation analysis, we further find that reduced psychological burdens of ownership can explain this effect.Experiment 2 then addresses the interdependencies between prosumers as asset providers and consumers as peer users; particularly, we show that peer users' brand preferences influence prosumers' purchase decisions.

Theoretical implications
Including the manufacturer as an active player in P2P asset sharing and exploring the purchase consequences advances current views on P2P asset sharing, which are limited to a triadic framework of providers, users, and sharing platforms (Benoit et al., 2017).As the source of assets though, the manufacturer is important to integrate, and by doing so, we reveal a neglected opportunity for manufacturers to capitalize on the sharing economy.To establish the ramifications for manufacturers and their sales, we conceive of prosumers in P2P asset sharing markets as acquiring assets both for their own consumption and for provision to other users.
These prosumers are mainly driven by economic motives (vs.sustainable or communal motives) and calculate gains and losses through a mental accounting process that they use to justify their purchase decision.By exploring and empirically testing this asset purchase decision, we extend current understanding of P2P asset sharing beyond the provision and usage of idle capacities (e.g., Costello & Reczek, 2020;Wirtz et al., 2019).Moreover, considering the prosumer as an asset acquirer offers a novel perspective on asset purchases in the sharing economy, going beyond a transactional focus in sharing research in general (Bardhi & Eckhardt, 2012;Hazée et al., 2019;Lamberton & Rose, 2012;Schaefers et al., 2016) and P2P asset sharing in particular (Benoit et al., 2017;Perren & Kozinets, 2018).
On a more general level, our investigation provides an alternative for manufacturers, which represents a direct response to recent calls to investigate how traditional firms can engage "in business model innovation by participating in the sharing economy" and specifically consider "car manufacturers (e.g., GM, Volvo) [that] have partnered with car-sharing platforms such as Turo" (Eckhardt et al., 2019, p. 9).From this novel perspective, we offer qualitative and quantitative evidence that manufacturers can profit from prosumers' prospective economic gains in P2P asset sharing, because they increase prosumers' willingness to purchase, particularly expensive assets.Thus, our results challenge the conventional notion that sharing is a threat to ownership (Zervas et al., 2017) and support predictions of value creation opportunities for manufacturers in the sharing economy (Jiang & Tian, 2018).On a more general level, the effect of prosumption in P2P asset sharing provides an interesting example of how service innovations can promote asset purchases.This effect applies not to the user but rather the prosumer, who deliberately acquires the asset to co-create service provision in the future.We also find a moderating effect of asset price on the relationship between prosumption and willingness to pay; that is, prosumers are even more inclined to spend anticipated economic gains from asset provision on a more expensive asset.Accordingly, a car buyer would spend the anticipated income from renting out the asset on a better engine, a dress buyer on a matching purse, and a bike buyer on a special saddle.Our result also relates to research on how anticipated reselling prices of assets affect purchase decisions (Chu & Liao, 2010) and offers empirical support for findings from extant analytical frameworks (e.g., Jiang & Tian, 2018).
A reduction in the financial burdens of ownership, due to monetary gains achieved by switching from a consumer to a prosumer role, underlies the positive effect on prosumers' asset purchase decisions.With this evidence, we provide a new perspective on the burdens of ownership in the sharing economy.Existing literature describes reduced burdens of ownership as a key determinant of whether people start to use shared assets (Moeller & Wittkowski, 2010;Schaefers et al., 2016); we further show that when consumers become prosumers, asset ownership can reduce the burdens of ownership instead of increasing them.As a result, we propose prosumption as a means to reduce the burdens of ownership in the consumer role by taking the role of asset acquirer while anticipating a reduction of the burdens of ownership through the provider role.In line with this noting of buying to share, we provide insights into relevant variables that support effective marketing of prosumption (Dellaert, 2019).
Finally, we explicitly investigate the effect of peer users on prosumers' purchase decisions to provide empirical insights on these crucial interdependencies in P2P asset sharing (Benoit et al., 2017;Ertz et al., 2018).Prosumers not only account for their own brand preferences when making a purchase decision but also consider users' brand preferences.The eventual purchase decision thus rests on prosumers' own brand preferences but also on their sense of which brand promises the most economic gains through P2P asset sharing.This novel view reveals the dimensionality of gains that drive the purchase process in P2P asset sharing: both gains from asset usage (consumer role) and gains from sharing the asset with a third party in exchange for a fee (provider role).We thus highlight an interesting and so far unconsidered conflict between expected gains from the consumption versus the provider role the prosumer faces when initially acquiring the asset.This novel interdependency and resulting role conflict can have important implications for managers and suggest interesting avenues for further marketing research.

Managerial implications
Despite initial managerial efforts to use prosumption to increase asset sales, the projects remain in their infancy, and strategic directions regarding how manufacturers can leverage prosumption are missing.This study provides insights for managerial practice; particularly, it highlights new opportunities for manufacturers to capitalize on the sharing economy by leveraging the service innovation of P2P asset sharing.
First, in contrast with current managerial and academic beliefs that "sharing is a form of anti-consumption with regard to possession" (Akbar et al., 2016, p. 4216), we illustrate that asset sharing services are an opportunity as well.To make use of this opportunity, manufacturers should recognize consumers as both service users and as prosumers (acquirer, consumer, and provider), which is integral to P2P asset sharing scenarios.Our empirical results also reveal that consumers regard the opportunity as attractive.To apply these findings, manufacturers could promote and bundle asset offers with P2P asset sharing platforms and provide the infrastructure to help potential buyers realize economic gains from their asset provision.
Second, allowing prosumers to monetize their purchase reflects a new business model opportunity for manufacturers: build a service around underutilized assets owned by consumers instead of providing and maintaining their own assets for asset-based services.By partnering with P2P asset sharing service providers, manufacturers might adjust their current business model, focused on asset sales, and forgo the costly creation of their own sharing service, including asset maintenance and service provision, while still benefitting from linking their brand with the positive connotation of the sharing economy and strengthening their asset sales by marketing to prosumers.Furthermore, leveraging the idea of prosumption, manufacturers could attract new consumers who previously may have been unable to afford assets and increase their selling potential by creating an opportunity for consumers to realize a form of discount, by promoting P2P asset sharing.Overall, the sharing economy does not need to be a threat.It even can be an opportunity for manufacturers to sell assets.
Third, prosumers are even more inclined to spend anticipated economic gains from asset provision on a more expensive asset.Thus manufacturers can upsell these customers by making more expensive purchases seem economically feasible by promoting that asset provision through P2P asset sharing services reduces the financial burdens that come with asset ownership.Advertising strategies could focus on the feasibility of both asset purchases and of getting a better model in the product line, due to the additional income they would earn on this asset.
Fourth, the interdependency between prosumers and peer users might constitute a crucial challenge.Prosumers will take users' preferences into account when making a purchase decision, so manufacturers need to adjust their marketing efforts to cater to the preferences of both.The interests of asset users and providers likely differ though (e.g., reliability vs. features), so manufacturers might need to split these two target customers of P2P services into two distinct target groups for their advertising.

Limitations and avenues for further research
We offer a first investigation of how the opportunity to become a prosumer in P2P asset sharing affects consumers' purchase decisions, which has some limitations that provide avenues for further research.In our exploratory and experimental studies, we capture consumers' reactions to becoming asset acquirers and providers (prosumers) for P2P asset sharing services.Our focus is on the identification of a general effect.However, similar to research on barriers to consumers' adoption of asset-based services, specific barriers might arise in P2P asset sharing settings.Identifying and offering ways to overcome these specific burdens would be a promising avenue for further research, relevant to both manufacturers and service platforms.
We focused on advertising a P2P asset sharing offer as a key stimulus; continued research might provide more nuanced insights into which marketing stimuli should be featured across the customer purchase journey to convince car buyers to embrace the P2P asset sharing offer.In a similar vein, the question of how to reduce the burdens of ownership provides an interesting basis for more nuanced research that could investigate the determinants of whether consumers actively reduce their burdens of ownership by becoming a peer user or a prosumer in the sharing economy.Moreover, our investigation focuses on monetary benefits of prosumption, which have great relevance; some less prevalent motives (i.e., sustainability, community) also might exert influences.Research with a more nuanced view could clarify the role of prosumption further.We also specifically focused on the P2P asset sharing rather than P2P service offerings (e.g., Uber).We are agnostic about whether our results would extend to such service offerings, which require asset purchase, asset provision, and also service provision.It might be of interest to determine if the need for service, not just the asset, changes customers' purchase decisions or if the underlying decision mechanism differs, rather than involving reduced ownership burdens.
Our findings illustrate that peer users' brand preferences affect prosumers' purchase decisions, though more research could clarify this interdependency, such as by examining how prosumers resolve the role conflict that results from buying an asset for their own usage versus seeking increasing economic gains through P2P asset sharing.For example, when do providers prioritize their own or peer users' preferences, and how might asset manufacturers influence this trade-off for purchase decisions?
buy a [brand] next time I buy a car.I intend to keep purchasing [brand] cars.0.960 2 preference on their purchase intentions is significant (M FC = 5.09, M IC = 4.45, M RC = 3.29, F(2, 394) = 56.336,p = .000),as is the main effect of peer users' brand preferences (M Pref = 4.63, M NoPref = 3.93, F(1, 394) = 23.891,p = .000).This positive effect holds independent of prosumers' brand preferences (i.e., first, indifferent, or rejected choice) but varies in size.That is, the lift is greater when peer users prefer a brand rejected by the prosumer (M Pref = 3.80; M NoPref = 2.79; d = 0.65), medium if the peer users' preference is a brand the prosumer is indifferent to (M Pref = 4.81; M NoPref = 4.09; d = 0.52), and minimal if they prefer the brand that also is the prosumer's first choice (M Pref = 5.25; M NoPref = 4.93; d = 0.25) (Fig.

Table 1
Review of Manufacturer Pilot Tests to Sell to Prosumers.Co launched the 01 car in Netherlands and Sweden, equipped with the necessary technology to facilitate peerto-peer carsharing, to allow owners to offset monthly ownership costs.Mini Germany, Netherlands 2021 Mini owners in Germany and Netherlands can retrofit their cars with a Mini share module, enabling them to digitally engage in P2P carsharing.All new production vehicles are equipped with the functionality.Fiat Chrysler -U-Go Spain 2020 Fiat Chrysler subsidiary Leasys launched U-Go in Spain to enable customers to offset ownership costs of leased vehicles by becoming U-Go players and renting their cars on the company's P2P carsharing platform.Mini sharing module can be used for P2P carsharing in all of Spain.The module is available ex-factory, and Mini released an app for usage in Spain.
Club target UK leaseholders to engage in P2P carsharing to generate income by sharing their vehicles, thus offsetting ownership costs.

Table 2
Focus Group Sample Description.