Understanding how virtuous lenders encourage support for peer-to-peer platforms ’ prosocial initiatives

P2P platforms are, therefore, advised to rely on virtuous lenders and strengthen users ’ moral principles to increase the persuasiveness of their prosocial appeals.


Introduction
Peer-to-peer (P2P) services are a subset of sharing economy activities whereby consumers can temporarily use a variety of private possessions (e.g., houses, cars, etc.) on the allowance of those objects' legal owners-hereafter "lenders" (Botsman & Rogers, 2010 a,b;Hamari et al., 2016).These services operate through digital marketplaces, commonly known as "P2P platforms", that match needs (i.e., demand) with haves (i.e., supply).P2P service platforms often attract people with similar interests willing to share their assets with others to save or gain money and form a community of like-minded people.People can use these platforms as borrowers, lenders, or even fill both roles.
The rise of these P2P services is intertwined with the discourse on public welfare.While critics argue that these services may exacerbate inequality (Schor, 2017) and problems such as discrimination (Edelman et al., 2017), supporters argue that such services facilitate the development of small entrepreneurial activities based on the efficient use of idle or under-utilized resources and reduce humans' impact on the natural environment (Stofberg & Bridoux, 2019;Tussyadiah, 2016).Advocates of P2P services believe, in particular, that they can contribute to prosocial goals, such as improving social equality and environmental sustainability, as well as increasing community cohesion and development (Botsman & Rogers, 2010 a,b;Perren et al., 2019;Piscicelli et al., 2018).Accordingly, emerging P2P services are making prosocial conduct part of their design, by offering second-hand goods (e.g., Freecycle, Facebook Marketplace), time (e.g., Timerepublik; Vaish et al., 2018), training programs (e.g., WellSquad), and maintenance services (e.g., Repair Café).Furthermore, many P2P service platforms engage in prosocial and pro-environmental initiatives-e.g., by donating part of the rental proceeds to prosocial organizations and/or soliciting additional donations from users.To illustrate, the P2P fashion rental platform Nova Octo supports charities and education programs, while the P2P carsharing platform Turo helps would-be entrepreneurs with crowdfunding loans.However, these efforts can meet resistance from consumers, who often mistrust requests for help to benefit the wider society (Alhouti et al., 2016;Choi & Park, 2021).For this reason, marketing scholars and practitioners continuously seek to increase the effectiveness of prosocial marketing communications (Pomering & Dolnicar, 2009).
Past studies have identified a variety of factors that may affect consumers' reactions to prosocial appeals: from victim facial emotions (Small & Verrochi, 2009) and nostalgic emotions (Ford & Merchant, 2010), to new technologies (Moriuchi & Murdy, 2022), language style (Pfeiffer, et al., 2023), and exposure to "virtuous" individuals, namely moral exemplars that, through their goodness, integrity, compassion, trustworthiness (Freeman et al., 2009), pursue what society considers intrinsically good (Brewer, 2009;Bright et al., 2006;Meyer et al., 2019).Because people tend to consider such individuals as role models, charitable organizations often deliver their prosocial appeals through "virtuous" endorsers (e.g., celebrities with a righteous public image that people perceive as moral exemplars; Ho et al., 2022;Karlan & List, 2020).Such virtuous individuals facilitate the formation of positive opinions, intentions and behaviors toward the endorsed organizations (Kang & Choi, 2016;Kim et al., 2014;Park & Cho, 2015) and positive reactions to their appeals.
However, considering that peopel generally learn about positive and negative behaviors through observing others (cf.Bandura's [1977] Social Learning Theory), even common individuals (e.g., one's "peers") who behave virtuously may act as role models (Gretemeyer, 2022;Jung et al., 2020) and promote prosocial behaviors (Bruhin et al., 2020;Kessler, 2017).This suggests that exposing P2P platform users to "virtuous" lenders-i.e., lenders who convey virtuousness through their profile descriptions-could motivate users to engage in desirable behaviors, such as supporting a P2P platform's prosocial initiatives.
Despite the potential value of this prosocial modeling effect (Gretemeyer, 2022;Jung et al., 2020), the literature on P2P services has overlooked this phenomenon.As a result, very little is known about lenders' social influence on P2P platform users.Furthermore, the mechanisms through which prosocial models influence others are still under-investigated and vary from context to context.To address these gaps, the present research hypothesizes and demonstrates that users' exposure to virtuous lenders elicits positive reactions to P2P platforms' prosocial appeals due to the aforementioned prosocial modeling effect.In particular, this research shows that virtuous lenders lead platform users to extend their ethical judgments to the platforms, consider these organizations as genuinely ("intrinsically") committed to social and environmental problems, and thereby react favorably to their prosocial appeals.According to the so-called "Meaning Transfer" theory (Biswas et al., 2006;McCracken, 1989), individuals tend to transfer one object's meaning to another through associations.Accordingly, we demonstrate that lender virtuousness fosters the perception that P2P service platforms have an "intrinsic" motivation to pursue prosocial goals (Kelley & Michela, 1980;Parguel et al., 2011), i.e., an authentic commitment to such goals.Consistent with past studies (Fajardo et al., 2018;Lee et al. 2019), this perception, in turn, elicits the so-called "donation efficacy" (Bekkers & Wiepking, 2011)-i.e., it leads users to believe that they can make a positive difference through their own contribution-and ultimately persuades users to support P2P platforms' prosocial initiatives.However, we also found that users' moral disengagement, i.e., the set of psychosocial mechanisms that they use to disable moral self-sanctions (Bandura, 2001), neutralizes the positive influence of lender virtuousness.
Our findings make three major contributions to the literature.First, they expand the stream of research on the positive influence of virtuous models.This stream of research has mostly concentrated on the direct influence that said models exert on individuals' prosocial behaviors (see Table 1 and Jung et al., 2020 for a meta-analytic review).Instead, we disentangle the direct and indirect mechanisms through which exposure to virtuous models prompts prosocial behavior.As regards the latter mechanism, our research builds on the aforementioned "Meaning Transfer" theory and, by focusing on P2P services, demonstrates that lenders' virtuousness extends to P2P platforms, engenders positive attitudes toward such organizations, enhances users' trust in the relevance of their own prosocial behavior, and hence increases their willingness to engage in such behavior.Second, our results bolster knowledge about how users relate to lenders in P2P platforms.Past research mainly noted that users tend to embrace a collaborative mindset and engage in empathy-driven relationships with lenders (Huang & Savary, 2022;Pera et al., 2019;Shuqair et al., 2019).However, our results demonstrate that lenders can also inspire prosocial actions if users perceive them as virtuous individuals.Third, our research expands the literature on the effects of moral disengagement on individuals' behavior.While past studies have mostly investigated unethical behaviors (Shepherd et al., 2013; see also Detert et al., 2008;Hinrichs et al., 2012), we show that moral disengagement also affects individuals' willingness to engage in prosocial behaviors.Accordingly, we found that virtuous lenders influence less individuals prone to disengagement from moral self-sanctions.

User reaction to lender virtuousness
Virtuous facts or actions-henceforth, "virtuous cues" (e.g., charitable donations, acts of kindness, self-sacrifice, altruism, or moral goodness)-elicit a warm, uplifting feeling that generally motivates individuals to act altruistically and engage in prosocial behaviors (Aquino et al., 2011;Haidt, 2000).This phenomenon can be typically seen when an individual is exposed to an organization or others who display virtuousness or are known for their virtuousness.Evidence, for instance, shows that virtuous organizations such as reputable charities and donation campaigners increase donations (Beldad et al., 2014;Brown et al., 2017;Purwandari et al., 2022; see Table 1).In the same way, virtuous models (e.g., virtuous endorsers) trigger favorable behavioral reactions to pro-social (Karlan & List, 2020;Wymer & Drollinger, 2015) and pro-environmental (Ho et al., 2022) appeals.
Even inanimate objects may convey virtuousness: past research has documented that music lyrics that refer to moral goodness increase the salience of prosocial thoughts, elicit empathy, and promote prosocial behavior (Greitemeyer, 2009a); videogames oriented around benefiting game characters encourage prosocial behaviors (e.g., helping others after a mishap; Greitemeyer, 2009b;Greitemeyer & Osswald, 2010;Shoshani, 2023); and green products marketed as "virtuous" (i.e., endowed with moral qualities) elicit feelings of moral reward and stimulate additional virtuous acts (e.g., charitable donations; Spielmann, 2021).Interestingly, past studies have also found that witnessing one's peer displaying virtue triggers a spontaneous desire to be altruistic, to cooperate with others, to donate resources, etc. (Jung et al., 2020).For example, exposure to co-workers' visible sign of support for a charity fundraising campaign (e.g., a pin with a charity logo) galvanizes further contributions (Kessler, 2017).Similarly, individuals living on the same street tend to engage in blood donation when they learn that their neighbors have enacted this behavior (Bruhin et al., 2020).
These findings can be explained by the theory on "prosocial modeling" (Gretemeyer, 2022), which claims that individuals are prone to engage in altruistic and cooperative behaviors when they learn about other persons' altruistic actions-i.e., when they are exposed to virtuous others.The prototypical situation addressed by this theory involves a virtuous individual (i.e., a prosocial model), someone who receives this person's help (the model target), someone who witnesses the virtuous individual's behavior (the observer), and someone who receives the observer's help (the observer target).This theory postulates that, after witnessing a prosocial model's display of virtue, the observer feels motivated to imitate this person when he/she is presented with an opportunity to engage in virtuous behavior-a phenomenon known as "downstream reciprocity" (Jung et al., 2020).The most feasible explanation for this effect is that individuals tend to develop normative beliefs through observational learning processes that lead them to adopt other people's attitudes and behaviors (Bandura, 1977;Busching & Krahé, 2020).
Extending this to P2P platforms, past research has found that mere exposure to lenders' socially desirable characteristics (e.g., care for others) might prompt socially desirable behaviors: For instance, the "superhost" badge, which signals outstanding host service quality in P2P accommodation platforms (Liang et al., 2017), motivates users to be friendly and conform to social norms (Newlands et al., 2019).Moreover, interactions with virtuous lenders may motivate users to adopt a cooperative mindset and engage in participative pricing decisions, which result in a higher willingness to pay than conventional payment methods (Huang & Savary, 2022).
Therefore, based on the above, it is plausible that exposing P2P platforms' users to virtuous lenders increases the likelihood that users favorably react to the platform's prosocial appeals and engage in charitable giving.Formally: H1.Exposure to virtuous lenders increases users' responsiveness to P2P platforms' prosocial appeals.
To better understand how exposure to lenders' virtuousness translates into favorable responses to prosocial appeals, we elaborate on past studies that investigated individuals' attributions of intrinsic (i.e., unselfish) versus extrinsic (i.e., selfish) motivations to other agents (e.g., other individuals, companies, or other organizations; Campbell & Kirmani, 2000;Kelley & Michela, 1980;Parguel et al., 2011;Skarmeas & Leonidou, 2013).This literature suggests that individuals tend to associate virtuous cues (e.g., acts of altruism) with intrinsic motives.As such, lender virtuousness may signal that a P2P platform is intrinsically motivated to support prosocial causes-i.e., it is led by voluntary commitment rather than external (e.g., monetary) rewards (Osterloh & Frey, 2004).The so-called "Meaning Transfer" theory (Biswas et al., 2006;McCracken, 1989) explains this effect because it postulates that individuals tend to transfer the meanings they attach to a company's representatives (e.g., a leader, an endorser, etc.) to the company itself.
Hence, it appears reasonable that platform users might transfer the meanings they attach to virtuous lenders to P2P platforms, thus perceiving them as intrinsically motivated to support prosocial causes.As the following sections explain, this effect, in turn, bolsters users' belief that their contribution can have an impact (donation efficacy; Bekkers & Wiepking, 2011).While this mechanism ultimately results in positive reactions to P2P platforms' prosocial appeals, moral disengagement-i.e., individuals' tendency to suspend moral self-sanctions (Bandura, 2001)-likely curbs the positive influence of lender virtuousness.Past research indeed noted that moral disengagement can significantly limit prosocial behavior (Sharma & Lal, 2020;Wu et al., 2021).

Mediating role of platforms' intrinsic prosocial motivation
According to the so-called "Attribution Theory" (Kelley & Michela, 1980), individuals generally infer motives from other people's or organizations' actions.Concerning companies, consumers might perceive their motives either as "extrinsic" (i.e., "self-serving") or "intrinsic" (i.e., "public serving").Consumers believe that companies' actions primarily aim to benefit the companies themselves; with the latter, companies' actions aim to benefit people outside the company (Campbell & Kirmani, 2000;Foreh & Grier, 2003).Accordingly, when faced with messages concerning prosocial or pro-environmental initiatives, consumers may infer that the company is opportunistically exploiting this form of communication (extrinsic motives) or genuinely committed to the cause (intrinsic motives).Only in the latter situation do consumers tend to perceive companies as public-oriented organizations (Edinger-Schons et al., 2018) whose benevolent actions stem from an authentic desire to do good for society (Marín et al., 2016;Skarmeas & Leonidou, 2013) and a selfless concern for others (Howie et al., 2018).
Consumers normally infer intrinsic motives from a company's established reputation for being socially responsible (Bhattacharya & Sen, 2004) or from third-party organizations' favorable evaluations of the company's commitment to prosocial causes (e.g., high sustainability ratings; Parguel et al., 2011).However, company representatives, such as founders, leaders, or endorsers also shape consumers' perceptions of such motives.The afore mentioned Meaning Transfer theory (Biswas et al., 2006;McCracken, 1989) indeed postulates that the meanings consumers associate with a representative of a certain organization carry over to the organization itself.Thus, because consumers normally extend virtuous endorsers' characteristics (e.g., ethicality, trustworthiness) to the endorsed brands (Kang & Choi, 2016) and companies (Kim et al., 2014), this theory explains why celebrities or CEOs who convey virtuousness through their honesty or trustworthiness enhance a company's reputation (Sohn & Lariscy, 2015;Zakari et al., 2019).The same principle also explains why celebrities who appear altruistic elicit favorable attitudes toward the charitable organizations they endorse (Park & Cho, 2015).By the same token, spreading negative information about endorsers and business leaders normally leads to negative perceptions of the related companies (Amos et al., 2008;Langmeyer & Shank, 1993;Zhu and Chang, 2013).
Globally considered, these findings indicate that individuals tend to transfer endorsers, CEOs, or other representatives' characteristics to the related organizations.Therefore, based on the transfer effect (Biswas et al., 2006;McCracken, 1989), this research hypothesizes that the meanings that platform users associate with virtuous lenders carry over to P2P platforms.Due to this effect, platform users perceive some virtuousness in the platforms, particularly an intrinsic motivation to pursue prosocial goals.This perception, in turn, promotes favorable reactions to the platforms' prosocial appeals.The following section delves into the mechanism underlying this effect.

Downstream consequences of intrinsic motivation: The role of donation efficacy
Consumers' perception of a company's behavior as extrinsically (i.e., self-oriented) versus intrinsically (i.e., public-oriented) motivated differently affects their intentions and behaviors toward said company (Becker-Olsen et al., 2006;Leonidou & Skarmeas, 2017).Extrinsic (i.e., egoistic) motives normally engender negative intentions and behaviors toward a company, whereas intrinsic (i.e., altruistic) motives engender more favorable reactions.Therefore, when consumers view a company's motives as extrinsic (i.e., selfish, manipulative, or exploitative), they tend to react less favorably to its prosocial appeals, whereas when consumers view such motives as intrinsic (i.e., unselfish and publicoriented), they might perceive companies' involvement in prosocial initiatives as benevolent actions (Romani & Grappi, 2014;Romani et al., 2016).In the latter situation, consumers tend to be more responsive to companies' prosocial appeals (Howie et al., 2018;Hwang & Kandampully, 2015;Lichtenstein et al., 2004) because intrinsic motives persuade consumers to believe that the company is actually striving to solve social or environmental problems; therefore, their donations will be useful.In other words, consumers who trust a company's genuine commitment to a social cause tend to form positive opinions about the outcome efficacy of their donations-hereafter "donation efficacy" (Bekkers & Wiepking, 2011)-and believe that the cause is worth supporting (Sargeant et al., 2006).
Information concerning organizations that deliver prosocial appeals normally affects message recipients' perception of the efficacy of their own donations (Fajardo et al., 2018).In particular, messages from organizations that appear intrinsically committed to prosocial causes boost message recipients' trust in the efficacy of their own contributions (Lee et al., 2019).With specific reference to charitable giving, past research has noted that many donors care about whether their prosocial behavior can actually benefit the cause they are supporting (White et al., 2019) and feel more motivated to donate if they believe that their contribution might have a significant impact (Gneezy et al., 2014;Sharma & Morwitz, 2016).Consistent with this reasoning, past research has found that individuals' confidence in charities is positively related to donation efficacy (Carroll & Kachersky, 2019).Donation efficacy, in turn, exerts a positive influence on charitable giving (Bekkers & Wiepking, 2011;Cheung & Chan, 2000;Labban et al., 2023): an individual's perceived ability to produce changes through his/her actions is, in fact, a key precursor of said actions (Bandura, 1997).
The literature on donation efficacy still lacks a comprehensive view of its determinants.While it appears intuitive that organizations' commitment to ethical causes will enhance potential donors' conviction in their donations' impact and spur more donation behavior, scholars have yet to assess this effect empirically.To close this knowledge gap, the present research proposes the following hypothesis: H2.Users' perceptions of P2P platforms' intrinsic motivation and user donation efficacy sequentially mediate the positive effect of lender virtuousness on users' reaction to prosocial appeals.
In short, this study hypothesizes that users' positive attributions about virtuous lenders transfer to P2P platforms; those perceptions then elicit donation efficacy, thereby triggering favorable reactions to the platforms' prosocial appeals.However, personality-related factors, such as moral disengagement, may adversely influence users' reactions to prosocial appeals.Thus, the following section hypothesizes that lender virtuousness does not affect users who exhibit a chronic tendency to disengage from moral standards.

Moderating effect of moral disengagement
Individuals can deactivate their moral self-regulation through a series of mechanisms (e.g., moral justification, displacement or diffusion of responsibility, disregarding or distorting the consequences of their own actions) that, taken together, constitute "moral disengagement" (Bandura, 2001).In broad terms, moral disengagement is the process through which people cognitively restructure the amount (or the cause) of harm generated by unethical behaviors (Detert et al., 2008).For instance, people may cognitively misconstrue reprehensible behavior to increase the moral acceptability of unethical behaviors.As such, moral disengagement drives unethical decision-making (Chowdhury & Fernando, 2014) by helping individuals avoid related feelings of guilt or distress (Hinrichs et al., 2012).In contrast, individuals with a strong tendency to respect ethical principles can resist the temptation to disengage from moral self-sanctions (Peeters et al., 2019).Such individuals are less likely to disengage from moral standards and are inclined to engage in prosocial behaviors (De Caroli & Sagone, 2014).
Individuals prone to moral disengagement tend to think that pursuing ethical principles is not worthwhile.Such individuals might exhibit low self-efficacy (i.e., a low ability to resist peer pressure to misbehave; Bandura et al., 2001) and may not engage in prosocial behaviors.Indeed, past studies have found that moral disengagement weakens individuals' intention to support animal protection organizations (Lim et al., 2019), to donate blood (Chen et al., 2022), to help others (Obermann, 2011;Paciello et al., 2013), and to engage in charitable giving (Baberini et al., 2015;Hardy et al., 2015) and environmental protection (Wu et al., 2021).
With specific reference to P2P services, the literature has highlighted that moral disengagement promotes misbehaviors (e.g., littering or misusing lenders' resources, even to the point of damage) that often compromise the normal functioning of such services and decrease consumer willingness to use shared resources (Schaefers et al., 2016).Against this background, the present research hypothesizes that moral disengagement offsets the positive influence of lender virtuousness on user willingness to support P2P platforms' prosocial initiatives.Formally: H3.Moral disengagement weakens the positive effect of lender virtuousness on user responsiveness to P2P platforms' prosocial appeals.

Table A1 in Appendix
A summarizes the studies that support the development of H2 and H3.

Overview of studies
We tested our research framework and hypotheses (Fig. 1) through three empirical studies that assessed consumers' reactions to the prosocial appeals delivered by three real P2P platforms.We note upfront that such platforms do not actively showcase lenders who convey ethicality and unselfishness; thus, virtuous lenders generally have limited visibility in the selected platforms' online environment, even if those people could favorably influence these platforms' image.Study 1 focused on Airbnb, a popular P2P service platform that supports people needing emergency housing.This study tested whether exposing platform users to virtuous lenders fosters favorable behavioral reactions to the platform's prosocial appeals (H1).The study assessed this reaction in terms of the likelihood that users will click on a hyperlink with instructions on how to make a donation.Study 2 (which included two experimental studies) focused on a real P2P car rental platform to explain the effect investigated in Study 1.To this end, the study tested whether lender virtuousness generates favorable reactions to the platform's prosocial appeals via increased user perception of the platform's intrinsic motivation, increasing user donation efficacy (H2).The study also controlled for a possible effect of the lender's gender.
Finally, Study 3 focused on a real P2P fashion rental platform and assessed whether users' moral disengagement moderates the positive influence that lender virtuousness exerts on users' reactions to the platform's prosocial appeals (H3).This study further validated lender virtuousness's direct and indirect effects on users' reactions to such appeals.

Pretest
We wrote a short text, seemingly written by an Airbnb host who described her experience of hosting a family of refugees (see the text in Table B1 in Appendix B).We assessed the extent to which the text made readers perceive the host as a virtuous person-i.e., dedicated to doing good for others-through an online pretest that involved a sample of 50 US consumers (mean age = 36.40,SD = 8.85, 42% female) recruited from Amazon's Mechanical Turk.Participants read the text and then rated the host's virtuousness using a three-item seven-point scale (α = 0.90) derived from Cameron et al. (2004;see Appendix C).A one-sample t-test-using the scale mid-point of (4) as the test value-confirmed that participants perceived the host as a virtuous person (M = 5.34, SD = 1.09, t(49) = 8.72, p < 0.001).

Main Study: Procedure
Study 1 was a two-cell (lender virtuousness: absent vs. present) between-subjects online experiment that involved 216 US consumers (mean age = 38, SD = 10.09,52% female), recruited as above.Participants were presented with information concerning Airbnb's prosocial commitment (Table B1).About half of them were also shown the pretested information describing the virtuous host, whereas the other half did not receive such information.All participants then read that they could support Airbnb with their donations and were shown a link that they could click to learn about the donation procedure.Our online questionnaire recorded participants' clicks on the link.Then, those participants who received information about the Airbnb host rated this person's virtuousness, intended as her dedication to others, using the same scale as in the pretest (α = 0.91).Finally, all participants provided their gender and age, and indicated how frequently they used Airbnb on a seven-point scale (1 = never, 4 = about twice a year, 7 = more than four times a year).

Main Study: Results
A one-sample t-test, conducted as above, confirmed that participants exposed to the virtuous lender perceived this person as virtuous (M = 5.37, SD = 1.10, t(1 0 7) = 12.94, p < 0.001).Next, we performed a logistic regression, which expressed participants' click-through behavior (0 = link not clicked, 1 = link clicked) as a function of lender virtuousness (-1 = absent, 1 = present), while controlling for participants' age, gender, and frequency of using Airbnb.This test returned a significant positive effect of lender virtuousness: Participants exposed to the virtuous lender were significantly more likely to click on the donation link than their unexposed counterparts (b = 0.52, Wald χ 2 = 4.09, p = 0.043).Thus, Study 1 offers initial evidence for H1.

Pretest
We wrote a brief profile description of a car lender, seemingly taken from SnappCar.nl, a Dutch P2P car rental platform (see the text in Table B2 in Appendix B).In this description, the lender presented himself as a person who was happy to cooperate with SnappCar and support its pro-environmental initiatives.We assessed the extent to which the written description made readers perceive the lender as a person dedicated to doing good for the environment (i.e., as a virtuous person) through a test that involved 50 US consumers (mean age = 36.5,SD = 9.35, 32% female) recruited from Prolific Academic.The study used a sample of US consumers in order to reduce the likelihood that participants would have prior knowledge of the selected European platform.To rule out a possible influence of prior knowledge of the selected platform, participants were asked to indicate whether they had used this service platform before by answering a dichotomous (Y/N) question.All participants confirmed that they had never used SnappCar before.Then, based on the profile description, pretest participants rated the lender's virtuousness using a three-item scale similar to the scale that we used in Study 1′s pretest (α = 0.92; see Appendix C for details).A one-sample t-test-using the scale mid-point (4) as the test value-confirmed that participants perceived the lender as a virtuous person (M = 5.58, SD = 1.14, t(49) = 9.78, p < 0.001).

Main study: Procedure
Study 2a was a two-cell (lender virtuousness: absent vs. present) between-subjects online experiment involving 179 US consumers (mean age = 43, SD = 12.69, 57% female) recruited via the same method as the pretest.As in the pretest, to rule out the possible influence of prior knowledge of the platform, participants indicated whether they had used SnappCar before by answering a dichotomous (Y/N) question.Then, a brief written scenario instructed them to imagine that they were interested in renting a car through SnappCar.The scenario provided some background information about this Dutch P2P car rental platform and its commitment to environmental problems.Then, about half of the participants were presented with information about a virtuous lender (i.e., the same profile description from the pretest).In contrast, the other half did not receive such information.
The scenario informed all participants that SnappCar welcomed monetary donations for environmental research institutions and asked participants to choose one of nine possible donation amounts, ranging from USD 0.00 to USD 10.00 (see Table B2 for details).We considered participants' choice as an indicator of the extent to which they favorably reacted to SnappCar's prosocial appeal.Next, all participants completed a three-item, seven-point scale measuring SnappCar's intrinsic motivation (e.g., "I think that SnappCar …is really conscious of environmental issues"; α = 0.92; see Appendix C), drawn from Parguel et al. (2011).
They also answered a three-item, seven-point measurement of perceived donation efficacy (e.g., "To what extent do you think that your donation could make a positive difference?";α = 0.94), which was drawn from Sharma and Morwitz (2016) and Mainardes et al. (2017; see Appendix C).Those participants who had received information about the lender also rated his virtuousness using the same scale as in the pretest (α = 0.82).Finally, all participants provided their gender and age.

Convergent and discriminant validity assessment
All participants confirmed that they had never used SnappCar before.We assessed the dimensionality of the employed multiple-item measures (i.e., platform intrinsic motivation and donation efficacy) by entering the corresponding items in a principal component analysis with oblimin rotation.The analysis returned a two-factor solution that explained over 80% of the variance-all items loaded on the corresponding factors, with factor loadings higher than 0.87.Next, we checked our multipleitem measures' convergent and discriminant validity by running a confirmatory factor analysis with the maximum likelihood estimation method.The analysis returned adequate fit statistics: χ 2 /df = 1.293;Goodness of Fit Index (GFI) = 0.981; Comparative Fit Index (CFI) = 0.998; Normed Fit Index (NFI) = 0.991; Standardized Root Mean Square Residual (SRMR) = 0.009.For each latent construct, factor loadings and composite reliability coefficients were higher than 0.90, while Average Variance Extracted (AVE) indicators (AVE Platform intrinsic motivation = 0.89; AVE Donation efficacy = 0.85) were higher than the squared inter-construct correlation coefficient (r = 0.28).Because each multiple-item measure exhibited satisfactory convergent and discriminant validity, we averaged the scores obtained on each scale to produce an aggregated measure of the respective construct.

Validating the indirect effect of lender virtuousness
We used Hayes' (2022) PROCESS macro to test the hypothesized serial mediation effect of platforms' intrinsic motivation and users' perceptions of their donation's efficacy (H2).To this end, we ran a linear regression analysis that considered lender virtuousness as a dichotomous independent variable (-1 = absent, 1 = present), platform intrinsic motivation and donation efficacy as first-level and second-level mediators, respectively, and participants' donation as the dependent variable.The analysis (see Table 2) returned the following: a significant positive effect of the independent variable on platform intrinsic motivation (b = 0.66, p = 0.001); a significant positive effect of platform intrinsic motivation on user donation efficacy (b = 0.68, p < 0.001), controlling for lender virtuousness (p = 0.923); and a significant positive effect of donation efficacy on user donation (b = 1.09, p < 0.001), controlling for both lender virtuousness (p = 0.238) and platform intrinsic motivation (p = 0.447).Consistent with H2, lender virtuousness exerted a significant positive indirect effect on user donation via the sequential mediation of platform intrinsic motivation and donation efficacy (b = 0.49, 95% CI: 0.21, 0.81).We re-ran the analysis with participants' gender and age as covariates, but the results did not substantially change.

Study 2b. Controlling for a possible effect of the lender's gender
To control for a possible effect of the lender's gender, we re-ran Study 2a using new samples of consumers and information about a female lender.

Pretest
We administered the adapted version of the same questionnaire as in Study 2a's pretest (see Table B3 in Appendix B) to a new sample of 80 US consumers (mean age = 45, SD = 12.96, 52% female) recruited from Prolific Academic.As in Study 1a, all participants confirmed that they had never used SnappCar before, thus ruling out a possible influence of prior knowledge of this platform.Then, they completed the same scale as in the previous pretest to assess the lender's perceived virtuousness (α = 0.93).A one-sample t-test-using the scale mid-point (4) as the test value-confirmed that participants perceived the lender as a virtuous person (M = 6.15,SD = 0.98, t(79) = 19.49,p < 0.001).

Main study: Procedure
Study 2b involved 202 US consumers (mean age = 44, SD = 12.59, 50% female), recruited as above, who confirmed that they had never used SnappCar before.As in Study 2a, about half of the participants received information about the female lender and indicated how much they were willing to donate to SnappCar.They then completed the same scales as in Study 2a to measure SnappCar's intrinsic motivation (α = 0.91) and donation efficacy (α = 0.95).Those participants who had received information about the lender also rated her virtuousness (α = 0.91).Finally, all participants provided their gender and age.

Convergent and discriminant validity assessment
A principal component analysis returned a two-factor solution that explained over 88% of the variance.All items loaded on the corresponding factors, with factor loadings higher than 0.80.Then, as in Study 2a, we ran a confirmatory factor analysis and obtained adequate fit statistics: χ 2 /df = 1.308;GFI = 0.984; CFI = 0.998; NFI = 0.991; SRMR = 0.029.Factor loadings were higher than 0.80, composite reliability coefficients were higher than 0.90, and AVE indicators (AVE-Platform intrinsic motivation = 0.87; AVE Donation efficacy = 0.77) were higher than the squared inter-construct correlation coefficient (r = 0.25).Because the employed measures exhibited satisfactory convergent and discriminant validity, we averaged the scores obtained on each of the two multi-item scales.

Indirect effect of lender virtuousness
As in Study 2a, we used Hayes' (2022) PROCESS macro to assess whether lender virtuousness indirectly affected participants' donation through the serial mediation of platform intrinsic motivation and participants' donation efficacy.The analysis (see Table 3) returned the following: a significant positive effect of lender virtuousness on platform intrinsic motivation (b = 0.22, p = 0.010); a significant positive effect of platform intrinsic motivation on user donation efficacy (b = 0.63, p < 0.001), controlling for lender virtuousness (p = 0.101); and significant positive effects of donation efficacy (b = 1.29, p < 0.001) and platform intrinsic motivation (b = 0.40, p = 0.021) on user donation, controlling for lender virtuousness (p = 0.758).Consistent with Study 2a, lender virtuousness exerted a significant positive indirect effect on user donation via the sequential mediation of platform intrinsic motivation and donation efficacy (b = 0.18, 95% CI: 0.04, 0.33).Including participants' gender and age as covariates did not alter these results.

Pretest
As in the previous studies, we wrote a brief profile description of a female lender, seemingly taken from Tulerie.com, a French P2P fashion rental platform that donates part of the rental proceeds to recycling organizations (e.g., Rewearable; see Table B4 in Appendix B).The lender presented herself as a person who was happy to cooperate with Tulerie and support its pro-environmental initiatives.Through a pretest involving 50 US female consumers (mean age = 35.84,SD = 10.55)recruited from Prolific Academic, we assessed the extent to which this description made readers perceive the lender as a "virtuous" person who is dedicated to doing good for the environment.Like before, we focused on a sample of US consumers to reduce the possible influence of prior knowledge about the selected platform.To rule out prior knowledge's possible influence, we asked participants to indicate if they had used Tulerie before by answering a dichotomous (Y/N) question.All participants confirmed that they had never used this service before.They read the description and then rated the lender's virtuousness using the same scale as in the previous pretests (α = 0.87).A one-sample t-test-using the scale mid-point (4) as the test value-confirmed that participants perceived the lender as a virtuous person (M = 5.07, SD = 0.98, t(49) = 7.73, p < 0.001).

Main Study: Procedure
Study 3 was a two-cell (lender virtuousness: absent vs. present) between-subjects online experiment that involved 135 US female consumers (mean age = 40, SD = 11.61)recruited as above.As in the pretest, all participants confirmed that they had never used Tulerie before, thus ruling out a possible influence of prior knowledge of the platform.A brief scenario (see Table B4) instructed them to imagine that they were interested in renting a luxury dress through this service.All participants received some background information about this French P2P fashion rental platform and its commitment to environmental problems.As in the previous studies, about half of the participants received information about the virtuous lender (in the form of the pretested profile description), whereas the other half did not receive such information.
The scenario informed all participants that Tulerie welcomed monetary donations for recycling organizations and asked them to choose one of nine possible donation amounts, ranging from USD 0.00 to USD 10.00 (see Table B4 for details).All participants completed the same scales used in Study 2 to measure the platform's intrinsic motivation and user donation efficacy.Participants then engaged in a distraction task aimed at assessing their online browsing habits and preferred websites.Afterward, they completed a three-item, seven-point scale of environmental moral disengagement (e.g., "Considering the pollution caused by big enterprises, it's hardly a sin for people to perform environmentally harmful behaviors sometimes"; α = 0.84), which was drawn from Wu et al. (2021) and Moore et al. (2012).Those participants who had received information about the lender also rated her virtuousness using the same scale as in the pretest (α = 0.89).Finally, all participants provided their gender and age.

Convergent and discriminant validity assessment
As in the previous studies, we assessed the dimensionality of the employed multi-item scales by running a principal component analysis.The analysis returned a three-factor solution that explained 85% of the variance; all items loaded on the corresponding factors with factor loadings higher than 0.85.Next, we assessed the convergent and discriminant validity of the scales through a confirmatory factor analysis, which returned adequate fit statistics: χ 2 /df = 1.490;GFI = 0.947; CFI = 0.988; NFI = 0.965; SRMR = 0.041.Factor loadings were higher than 0.67 and composite reliability coefficients were higher than 0.80, thus supporting convergent validity.AVE indicators (AVE Platform intrinsic motivation = 0.86; AVE Donation efficacy = 0.80; AVE Moral disengagement = 0.67) were higher than the squared inter-construct correlation coefficients (which were all lower than 0.25), thus supporting discriminant validity.Therefore, we averaged the scores obtained on the three multi-item scales to produce aggregate measures of the respective constructs (i.e., platform intrinsic motivation, donation efficacy, and moral disengagement).

Main effects of lender virtuousness
A one-sample t-test confirmed that participants exposed to the Note: N = 135; X was coded as: − 1 = Absent; 1 = Present; Me = Mediator.

Validating the indirect effect of lender virtuousness
To obtain further evidence for the serial mediation effect of the platform's perceived intrinsic motivation and donation efficacy, we used Hayes' (2022) PROCESS macro to test a linear regression model.Lender virtuousness was the dichotomous independent variable, platform intrinsic motivation and donation efficacy were first-level and secondlevel mediators, respectively, and user donation was the dependent variable (Table 4).This analysis yielded a significant positive effect of lender virtuousness on platform intrinsic motivation (b = 0.18, p = 0.035).In turn, platform intrinsic motivation exerted a significant positive effect on user donation efficacy (b = 0.67, p < 0.001) when controlling for the effect of lender virtuousness (b = 0.33, p = 0.008).Finally, donation efficacy exerted a significant positive effect on user donation (b = 0.81, p < 0.001), controlling for the effects of both lender virtuousness (p = 0.287) and platform intrinsic motivation (p = 0.745).Consistent with H2, lender virtuousness exerted a significant positive indirect effect on user donation via the serial mediation of platform intrinsic motivation and donation efficacy (b = 0.10, 95% CI: 0.01, 0.21).The inclusion of participants' gender and age as covariates did not alter the results.

Moderation effect of moral disengagement
To assess the predicted moderation effects of moral disengagement, we used Hayes' (2022) PROCESS macro to run a serial mediation model that tested the above relationships, alongside the possible moderating effect of moral disengagement.The analysis returned a significant positive effect of participants' moral disengagement on platform intrinsic motivation (b = 0.18, p = 0.037; Table 5), as well as significant positive effects of lender virtuousness (b = 0.33, p = 0.009) and platform intrinsic motivation (b = 0.67, p < 0.001) on donation efficacy.The analysis also returned a significant positive effect of donation efficacy (b = 0.81, p < 0.001) on user donation and a significant negative interaction effect of lender virtuousness and moral disengagement on user donation (b = -0.34,p = 0.034).
In line with H3, the conditional direct effect of lender virtuousness on user donation was significant and positive at a lower level of moral disengagement (M− 1SD; b = 0.72, 95% CI: 0.08, 1.36), but nonsignificant at a higher level of moral disengagement (M + 1SD; b = -0.27,95% CI: − 0.90, 0.36).Therefore, the results confirmed that virtuous lenders enhance potential donors' responsiveness to P2P platforms' prosocial appeals, but only when potential donors are less likely to disengage from moral self-sanctions.Again, including participants' gender and age as covariates did not alter the results.

General discussion
Past studies have documented a positive relationship between exposure to virtuous objects or persons and willingness to engage in prosocial behaviors, such as charitable giving (e.g., Jung et al., 2020) or helping others (e.g., Greitemeyer, 2009a, Greitemeyer, 2009b).Although this effect plays an important role in prosocial communication, its mechanisms are still far from being understood.For companies involved in prosocial actions, like P2P service platforms, a better understanding of such mechanisms can support the development of effective communication strategies.However, marketing scholars have so far overlooked these companies' prosocial appeals and the communication potential of virtuous lenders.Therefore, little is known about the factors motivating users to cooperate with these platforms and support prosocial projects.
The results of this research advance knowledge on this topic by establishing, first of all, that lenders who convey a sense of virtuousness through their profile descriptions do not only lead users to make positive attributions about P2P platforms' intrinsic motivation to pursue prosocial goals, but also inclines users to think that their donations can make a positive difference (donation efficacy).These findings, which ultimately result in positive reactions to P2P platforms' prosocial appeals (e.g., greater willingness to support their prosocial initiatives through charitable giving), highlight a persuasive indirect effect determined by mere exposure to virtuous individuals which has been neglected so far.Furthermore, although it may seem intuitive that the persuasive influence of virtuousness carries over to online environments, the literature on ethical behavior and online communication has only recently started to investigate this topic (Leban et al., 2021).
Second, our findings expand the emerging stream of research on how P2P service users relate to lenders.Past studies make clear that P2P service users are inclined to collaborate (Huang & Savary, 2022) and empathize with lenders (Pera et al., 2019).For this reason, users are tolerant of lenders' service failures (Shuqair et al., 2019).However, our results indicate that lenders can inspire prosocial actions when conveying virtuousness.From this perspective, virtuous lenders' role appears comparable to the role played by moral exemplars (e.g., Hattke & Hattke, 2019;Jung et al., 2020;Nook et al., 2016;Shao et al., 2008).
Finally, this research increases the field's comprehension of the effects of moral disengagement on individuals' behavior.Extant research has often investigated this construct with reference to behaviors that violate ethical standards (Detert et al., 2008;Hinrichs et al., 2012;Shepherd et al., 2013).However, moral disengagement may also affect individuals' propensity to enact (or not enact) prosocial behaviors (Wu et al., 2021).Accordingly, our third study demonstrates that lender virtuousness positively influences users' willingness to support a social cause, but only when users do not exhibit high levels of moral

Table B1
Study 1: Materials used in the main study.disengagement.Thus, our findings to research on the negative effects of moral disengagement on prosocial behavior, which remains under-developed.

Managerial implications
From a managerial point of view, this research offers guidance to marketing managers of P2P platforms supporting prosocial initiatives.Our findings indicate that such platforms could identify virtuous lenders-for instance, by checking their profile pages or even contacting them to assess their virtuousness (e.g., in terms of sensitivity to social and environmental problems)-and bolster their visibility in the online environment to increase users' willingness to support the afore mentioned initiatives.To this end, the platforms could devote specific sections of their websites, blogs, and social media to virtuous lenders.Furthermore, since virtuous lenders can boost the platforms' perceived intrinsic motivation, platform managers could seek to involve virtuous lenders in their prosocial communication programs.To illustrate, virtuous lenders may serve as endorsers in P2P platforms' prosocial communications.This tactic would strengthen users' perceptions of the platforms as genuinely committed to social and environmental issues, motivating users to contribute to those initiatives.In particular, considering that this perception of genuine commitment to prosocial causes enhances platform users' confidence in the relevance of their own contribution, P2P platforms could further reassure users about the impact of their donations.Toward this end, they should seek to communicate this impact in a transparent manner.
P2P platforms should be aware that these communication strategies might be ineffective when users are prone to moral disengagement.Thus, P2P platforms should also seek to buffer (or even neutralize) the negative influence of moral disengagement.To achieve this goal, platform managers could deploy strategies to increase users' selfidentification as moral individuals-i.e., their moral identity (Aquino & Reed, 2002).In this respect, the literature suggests that internalizing ethical principles reduces individuals' likelihood of disengaging from such principles and keeping a moral identity (Aquino et al., 2007;Vitell et al., 2011).Therefore, platform managers and virtuous lenders could remind users of the importance of respecting ethical principles (e.g., using others' possessions in a responsible way) and encourage them to adhere to such principles.Moreover, platform managers and virtuous lenders could share this information through their communication channels if a certain prosocial project receives support from many users.By highlighting the normative value of charitable giving, this tactic could reduce users' likelihood of disengaging from moral standards and enhance users' willingness to support P2P platforms' prosocial initiatives.

Limitations and future research directions
The limitations of this research pave the way for future studies on the effect of lender virtuousness in P2P platforms.First, our studies manipulated lender virtuousness using "static" stimuli: brief self-

Table B2
Study 2a: Materials used in the main study.presentations based on written text.However, it would be interesting to check whether richer and more interactive stimuli (e.g., brief video presentations) can enhance the positive influence of lender virtuousness on users' reactions to prosocial appeals.Second, our three studies considered platforms offering services with some hedonic components (i.e., accommodation, car rental, and fashion rental services).However, assessing the identified effects hold when the platform offers more utilitarian services (e.g., workspace or tool rental services) might be worthwhile.Third, this research identified that a platform's intrinsic motivation and participants' donation efficacy mediate the effect of lender virtuousness on participants' donations.However, future research could investigate alternative mediators, such as the enhanced feelings of empathy activated by virtuous lenders.Finally, it would be interesting to test whether situational factors such as users' sense of social closeness (Liberman et al., 2007) to lenders or lenders' popularity (which can be signaled by the number of users' reviews; Tang et al., 2023) magnify our identified effects of lender virtuousness.

Appendix C. . Main measures used in the research
Platform intrinsic motivation (Parguel et al., 2011): I believe that the platform… (1 = strongly disagree; 7 = strongly

Table B4
Study 3: Materials used in the main study.

Table 1
Key studies on the effect of virtuousness on prosocial behavior.

Table 5
Study 3: Moderated serial mediation analysis.= 135; X was coded as: − 1 = Absent; 1 = Present; Me 1 = First level mediator; Me 2 = Second level mediator; Mo = Moderator.Studies supporting the downstream effects of lender virtuousness investigated in the present research.
G.Pino et al.

Table D1
Study 2a: Differences between the study's conditions.

Table D3
Study 3: Differences between the study's conditions.