Corporate Socially Responsible Initiatives and Their Effects on Consumption of Green Products

Corporate social responsibility (CSR) research has focused often on the business returns of corporate social initiatives but less on their possible social returns. We study an actual company–consumer partnership CSR initiative promoting ecologically correct and conscious consumption of bottled mineral water. We conduct a survey on adult consumers to test the hypotheses that consumer skepticism toward the company–consumer partnership CSR initiative and the moral emotion of elevation mediate the relationship between company CSR motives perceived by consumers and consumer behavioral responses following this CSR initiative. Favorable consumer behavioral responses, in turn, relate positively to consumer support of other green products. The results provide scholars and managers with means of improving their understanding and handling of company–consumer partnership CSR initiatives.


Introduction
At least three approaches to encouraging responsible consumption can be identified. First, public policy constitutes a type of top-down governance where regulations shape the behavior of the public through incentives or penalties (e.g., Albareda et al. 2007). Second, non-governmental organizations (NGOs) provide independent information, advice, and advocacy designed to influence government policies, firm practices, and consumer behavior (e.g., Konga et al. 2002;Arenas et al. 2009). Finally, corporations favor business choices that yield measurable business returns with the belief that what is good for the firm is largely good for the public as well, at least indirectly (e.g., Russo and Fouts 1997).
We consider a novel fourth program for energizing responsible consumption, namely, creating a partnership between corporation and consumer for promoting a shared responsibility to protect the environment. As with any partnership, such an endeavor can be initiated by either corporation or consumer(s). Corporate-initiated programs can involve consumers directly in a bottom-up way to create a common cause for socially responsible consumption. The aim is to forge joint, shared commitments in a way difficult to accomplish with top-down public policy and independent NGO undertakings. Consumer behavior is changed by establishing a kind of social contract between firm and consumer as a means for promoting the common good. Many firm-consumer positive acts can aggregate or accumulate to yield collective societal benefits.
Corporate-sponsored partnerships can be established by providing goods that satisfy both consumer and firm needs yet do so in a way also benefiting the environment. Such initiatives work at the micro level of the individual psychology of consumers by providing frequent opportunities for reinforcement in consumers through their involvement and contributions to the environment when repeatedly purchasing everyday goods from the corporation. The momentum so engendered cultivates communal feelings and is potentially self-sustaining through individual learning and the sense of community that links consumer, firm, and society (Bagozzi et al. 2012). In the spirit of Aristotele's interpretation of ethics, such actions potentially create habits of virtuousness, a kind of learning by doing (Hursthouse 1999).
Many companies leverage their unique capabilities to encourage the adoption of sustainable behaviors in domains relevant for their businesses and markets. For example, Unilever has developed the ''Five Levers for Change'' roadmap (http://www.unilever.com/sustainable-living/our approach/embeddingsustainability/Encouragingbehaviour change/) which makes systematic basic rules to encourage consumers to change their behavior. Starting from its previous experiences in health and hygiene campaigns, Unilever plans to widen the range of application of its approach to the environmental field, encouraging consumers to use less water, emit less greenhouse gases, and produce less waste. Recently, Comfort One Rinse has sought to encourage people to use less water when rinsing clothes (http://www.youtube. com/watch?v=ZDITi7mnj-I); and Suave has tried to communicate how turning off the tap when lathering hair can save money on utility bills and has partnered with the Nature Conservancy in this regard (http://www.suave.com). Beyond saving money for consumers and conserving resources environmentally, such initiatives give the potential for the firms to be beneficiaries of consumer gratitude ) and at the same time create mutual feelings of acting together.
Despite the rise of such corporate social initiatives, corporate social responsibility (CSR) research has focused largely on the business returns of such activities rather than on the social returns, per se (for exceptions, see Du et al. 2008;Gourville and Rangan 2004). Importantly, in addition to their ''primary'' impact on the brand and company (see among others, Du et al. 2007;Sen et al. 2006), CSR initiatives can also influence other ''secondary'' outcomes related to partner organizations (e.g., non-profits) and the main cause or social issue addressed by the company's CSR efforts (Lichtenstein et al. 2004). Moreover, firms can motivate consumers to undergo some forms of behavioral change themselves. For example, as reported by Bhattacharya and Sen (2004), the cross-sector partnership between Home Depot and Habitat for Humanity enhanced consumers' support of related non-profit organizations, as well as positively affected their attitudes toward the issue of housing for underprivileged people. In addition, the marketing of organic food products by several producers (e.g., Stonyfield Farm, Newman's Own) influenced consumers to increase the overall proportion of organic foods in their diet, through a real process of behavioral change (Bhattacharya and Sen 2004).
Little attention has been given until recently to the investigation of these ''secondary '' outcomes (e.g., Lichtenstein et al. 2004;Romani and Grappi 2014) and especially to whether and how corporate support for certain causes or issues can result in consumers' engaging in behaviors that are more consistent with the cause (for a pioneering example in this vein, see Du et al. 2008). However, as Bhattacharya and Sen (2004) point out, companies need a better understanding of these ''additional'' outcomes of CSR if they are interested in increasing the social return on their CSR investments. Moreover, Bhattacharya et al. (2009) report the need for a more precise understanding of different possible responses by consumers to CSR initiatives and the underlying psychological processes driving such responses. In particular, extant models tend to overlook the relationship between company CSR activities and consumer behavioral change, as well as their possible explanations.
Our research addresses this gap by focusing on the social returns of an environmental initiative by a national grocery retailer directed to reduce consumption of bottled mineral water, in terms of target consumer reactions both to water consumption and more generally to support of green products. We assess the program's social return by examining the extent, and the mechanisms by which, it increases the commitment to reduce the consumption of bottled mineral water within the target group (defined below). We also examine the effects of general support to consumer green purchasing. In detail, our research makes three contributions: (a) It documents, in terms of actual consumer behavioral responses, the social outcomes of a real companyconsumer partnership CSR initiative for the environment; (b) It demonstrates the mechanisms by which these responses operate, showing the key role played by company CSR motives, consumer skepticism, and the consumer moral emotion of elevation; and (c) It reports the role of such responses to the specific CSR initiative in strengthening consumer support of other green-related products and practices.
The rest of the paper is organized as follows. We begin by reviewing key research findings and develop hypotheses based on company CSR motives, skepticism, the moral emotion of elevation, and consumer behavioral responses following the company-consumer partnership CSR initiative. Then we present the results of a survey study designed to test our predictions. We end with a discussion of our findings for theory and marketer practices.

Conceptual Framework and Research Hypotheses
Prior research (Forehand and Grier 2003;Sen et al. 2006;Ellen et al. 2006;Du et al. 2007;Parguel et al. 2011) shows that business returns of CSR initiatives are contingent on, but generally occur below the level of awareness of, motives that consumers attribute to the company's involvement in social responsibility initiatives. In general, consumer attributions of a company's CSR motives may be of two kinds: extrinsic, in which the company is seen as attempting to increase its profits or other business returns; or intrinsic, in which it is viewed as acting out of a genuine concern for the social issue. Stronger attributions of intrinsic motives lead consumers to react positively toward the company, while perceptions of predominantly extrinsic motives lead to ambivalent or less-favorable attitudinal and behavioral responses.
We believe that consumer attributions can play an important role also in consumer behavioral responses following company support of a specific social cause. However, we propose that the effect of these extrinsic and intrinsic motives on consumer behavioral responses will be mediated, respectively, by skepticism of, and the moral emotion of elevation toward, the company-consumer partnership CSR initiative.

Company CSR Motives, Skepticism, and Consumer Behavioral Responses
Skepticism refers to a person's tendency to distrust or disbelieve. Although some consumer studies consider skepticism as a personality trait (e.g., Boush et al. 1994;Obermiller and Spangenberg 1998), most of consumer research, particularly in contexts related to CSR (Forehand and Grier 2003;Singh et al. 2009;Vanhamme and Grobben 2009;Pirsch et al. 2007), focuses on situational skepticism, which is a consumer state, induced independently from one's traits, and varies depending on the context. Here skepticism is localized to the perception of specific marketing actions or messages of the company under study. Consumers utilize their knowledge and information to interpret and evaluate such actions and messages, and in some cases skepticism emerges with subsequent effects on consumer evaluations and behaviors (Friestad and Wright 1994).That is, people acquire general knowledge about goods and tactics of companies and their persuasive attempts in this regard to influence the public, and at the same time or subsequently, consumers develop coping styles to adapt and respond to the persuasive tactics. Over time, consumers acquire knowledge and coping skills, such as judged skepticism, that are used to evaluate specific claims made by companies (Friestad and Wright 1994).
Recently, Skarmeas and Leonidou (2013), applying attribution theory (Kelley and Michela 1980;Weiner 2000), provided specific evidence on the role of consumer causal inferences for company's CSR practices in the development of skepticism toward CSR. Their findings show that extrinsic motives (in their paper termed, egoistic and stakeholder-driven motives, based on Ellen et al. 2006) contribute to the development of consumer skepticism toward CSR, while intrinsic motives (termed, value-driven motives) inhibit its formation. Moreover, skepticism exerted a direct negative effect on consumer-based brand equity, consumer resistance to negative information, and word of mouth. It is important to note that past research shows that consumers attribute both intrinsic and extrinsic motives to firms and both influence consumer responses (e.g., Ellen et al. 2006).
Here we propose a mediating role of skepticism toward CSR on the relationships between the motives inferred by consumers for the company-consumer partnership CSR initiative and consumer behavioral responses in reactions to this CSR initiative in terms of their consumption of green products. Perceptions of company CSR motives influence the development of skepticism. Consumers who believe that the company is taking advantage of the social cause and trying to capitalize on it to promote its own branded products will respond to the CSR campaign with high levels of skepticism (Ellen et al. 2006;Forehand and Grier 2003;Skarmeas and Leonidou 2013). Conversely, low levels of skepticism will be present in consumers who perceive authentic benefits for the environment as a result of company CSR actions. Skeptical consumers then doubt the company's reasons for contributing to the well-being of society and are wary of its ethical standards and social engagement. These reactions undermine consumers' motivation to re-orient their responses toward more sustainable practices. From the point of view of social exchange theory (e.g., , skepticism devalues persuasive claims made by the firm and weakens the likelihood that one would respond positively to the initiative attached to the claim. This happens in at least two ways. First, skepticism is a threat to joint gains via mutual interdependence and to perceptions of fairness undergirding reciprocity, both of which devalue social exchange and weaken the prospects for positive ongoing relationships. Second, as implied by the affect theory of social exchange (Lawler 2001), people respond to actual and anticipated social exchanges with positive and negative emotions that serve to increase or decrease the value of the relationship to the parties and shape the nature and course of social exchange. The emotional outcomes influence shared responsibility (Lawler et al. 2008), commitment of parties to each other , and trust (Zaheer et al. 1998). Such shared consequences emerging between consumer and firm influence the responses of each toward the other; in our case, consumers react skeptically to perceived motives of firms, and this, in turn, leads to favorable or unfavorable responses to firm-initiated CSR, depending on whether the motive is seen to be extrinsic or intrinsic. Consequently, we hypothesize the following: H1 Consumer skepticism toward CSR will mediate the relationship between perceived company CSR motives and consumer behavioral responses following the companyconsumer partnership CSR initiative: H1a: The greater the perceived extrinsic motives, the greater the felt skepticism; H1b: The greater the perceived intrinsic motives, the lower the felt skepticism; H1c: The greater the felt skepticism, the less favorable consumer behavioral responses following the companyconsumer partnership CSR initiative.

Company CSR Motives, the Moral Emotion of Elevation, and Consumer Behavioral Responses
Elevation, one of the ''other praising'' moral emotions, is elicited by acts of charity, kindness, self-sacrifice, or other displays of virtue on the part of others (Haidt 2003). Elevation consists of felt elation and intense global feelings of inspiration and well-being. Several studies in psychology (see among others, Aquino et al. 2011;Cox 2010;Schnall et al. 2010) provide evidence that witnessing altruistic behaviors elicits elevation, which, in turn, promotes a prosocial orientation and leads to increased altruism. Recently, Romani and Grappi (2014) used mixed methods evidence to show the important role of this particular emotion in explaining consumer behavioral responses to CSR toward the local community (e.g., support by the retailer of the community after an earthquake disaster). Elevation is elicited by CSR initiatives directed to prevent or relieve harm to people, and this emotion motivates consumers to react morally or pro-socially themselves.
Here we hypothesize a role of consumer perceived motives for company CSR practices in the elicitation of elevation. Intrinsic motives refer to consumer attributions that the company engages in CSR actions largely because of its moral and societal ideals and standards. In this case, consumers believe that the company cares about the cause and has authentic concerns about social problems. Such attributions of altruistic motives to firms engender feelings of respect and admiration toward firms in consumers and convey an aura of goodness and idealism that is uplifting in the perceiver. Consequently, in line with the research illustrated above, we can expect consumers to experience felt elevation in response to such benevolent, altruistic initiatives. Conversely, extrinsic motives refer to the attribution that the company is exploiting rather than supporting the cause. When consumers assign opportunistic motives to CSR initiatives, they become suspicious and perceive CSR as a deliberate attempt to mislead them into false conclusions about the company. Consumers conclude that the company is preoccupied with its own interests. Beyond skepticism, such attributions engender feelings of disgust and of being deceived in consumers leading them to experience negative and critical moods and to negatively react toward the company CSR activities (Du et al. 2010). Thus, when opportunistic motives are revealed, consumers are likely to question and doubt the CSR effort, and we can expect that the elicitation of elevation will be lessened. In the presence of elevation, consumers are likely to embrace the company's social initiatives, feel the urge to reciprocate with supportive actions (Lawler 2001;, and be motivated to react morally or pro-socially themselves toward the firm. In the presence of diminished elevation, consumers are likely to react negatively to the company's lack of forthcomingness, to experience an urge to not patronize the company, and be unmoved to act morally to support the company. Consequently, we hypothesize the following: H2 Consumer felt elevation toward CSR will mediate the relationship between perceived company CSR motives and consumer behavioral responses following the companyconsumer partnership CSR initiative: H2a: The greater the extrinsic motives, the lower the felt elevation; H2b: The greater the intrinsic motives, the greater the felt elevation; H2c: The greater the felt elevation, the more favorable consumer behavioral responses following the companyconsumer partnership CSR initiative.

Consumer Behavioral Responses Following the Company-Consumer Partnership CSR Initiative and Consumer Support of Other Green Products
Finally, changes in reactions following a CSR initiative may lead to readiness to adopt other environmentally beneficial purchasing activities. Based on the evidence of positive spillover effects in relation to pro-environmental behaviors (e.g., Lacasse 2013; Thøgersen and Olander 2006;Whitmarsh and O'Neill 2010), we expect that consumers who react favorably to the CSR campaign directed at consumption of a specific green product will express greater environmental support of the purchase of other green products. For example, Thøgersen and Olander (2006) argue and find that different environmentally beneficial choices spring from a common motivational root. Further, they point out that broad, cross-situational goals reside in abstract attitudes or values. We build on this rationale and claim that attributions of extrinsic and intrinsic motives constitute common motivational causes of a general nature, as analyzed by Thøgersen and Olander (2006). We add to their rationale by proposing that these general motivations produce their effects on support of the company-consumer partnership CSR initiative under study and broader green purchases through the specific mediating emotions of felt skepticism and elevation. The similarity among green product categories favors positive spillovers (Thøgersen 2004). Such spillover effects have been shown to depend as well on one's own pro-environmental selfidentity (Whitmarsh and O'Neill 2010). Consequently, we hypothesize the following: H3 Consumer support of other green products will relate positively to favorable consumer behavioral responses following the company-consumer partnership CSR initiative.
Figure 1 presents our theoretical model. In addition to the hypothesized relationships between variables (solid arrows), we also control for other possible explanations (dashed arrows). The direct influence of extrinsic and intrinsic motives on consumer behavioral responses following the company-consumer partnership CSR initiative for a specific green product will be conducted in a way to detect and control for possible unmediated effects of these two variables on the dependent variable, under the hypothesized mediation by skepticism and elevation. Moreover, we also consider the possible direct paths between felt elevation and skepticism and consumer support of other green products, to test for partial or full mediation and rule out alternative explanations. These tests of hypothesized and rival hypotheses will be done while controlling for the degree of consumer knowledge of, and expertise with, green products, as a covariate (dotted arrows; in Fig. 1). The particular company under study provided various green products in assortment, thereby making it easier for consumers to engage in other green purchasing behaviors. Effects on Consumption of Green Products 257 of promoting ecologically correct and conscious consumption of bottled water. The basic idea was to protect the environment by encouraging consumers to use tap water or at least mineral water from nearby springs (''Km 0 water''), reducing their bottled water consumption that has a very high environmental impact (Botto 2009). To this end, Coop engaged in multiple initiatives with special emphasis on direct action at the point of sale (e.g., aisle displays of mineral bottled water changed in ways promoting tap water and its purification, eco bottles, and local sources), communicating these activities in an integrated way through television, print advertising, and the creation of a special community on the company web site. Finally, public water systems provided the information on the high quality of water coming from taps in the home. The Coop CSR initiative generated broad public discourse in various national media where its motives behind the initiative were initially questioned. For example, Coop was accused of capitalizing on the public water social movement to improve its reputation and its positioning compared to competitors and was criticized for exploiting sales of its private label bottled water which came from close-in locations and yielded considerable margins.

Respondents and Procedures
We surveyed Coop customers exposed to the CSR campaign using customer email addresses provided by the company. The study began in mid-2012. We contacted 1288 Coop customers and obtained 356 complete questionnaires (i.e., a 27.64 % response rate). However, 10 responses were eliminated because of incomplete data on the key measures and another 16 because respondents reported being unaware of the CSR initiative. Thus, the final sample consisted of 330 respondents, 43 % of whom were male. We checked the demographic characteristics of our sample to ensure the presence of different respondents and avoid the risk of focusing exclusively on specific respondent categories. Respondents varied considerably in terms of age (18-

Questionnaire Development and Measures
A review of the literature and interviews with Coop managers and Coop customers helped specify the conceptual domain of each construct and initial operationalization. The questionnaire was then refined through independent evaluations by three professors and three PhD students.
Whenever possible, well-established measures were identified from existing research and adapted to better suit the context at hand. The sponsor insisted that the questionnaire not be too long, which required using sub-scales from larger scales drawn upon. The items measuring intrinsic and extrinsic CSR motives were derived from the work of Ellen et al. (2006), Du et al. (2007), and Parguel et al. (2011). Participants expressed their level of agreement or disagreement on a 7-point Likert scale. Two items assessed intrinsic motives introduced by the words ''express your level of agreement with the following statements regarding the Coop CSR initiative''; ''I think that Coop feels morally obligated to help environment and society''; and ''I think that Coop has a real, authentic long term interest in the environment.'' Two items assessed extrinsic motives: ''I think that Coop is taking advantage of environmental causes to help its own business'' and ''I think that Coop seeks actually to get publicity.'' A three-item measure of skepticism was developed based on Babin et al. (1995) and Taylor et al. (2010). The items were ''skeptical,'' ''suspicious,'' and ''distrustful,'' with scale anchors of ''not at all'' (1) and ''very much'' (7). For elevation, we used the measure provided by Grappi et al. (2013). The items were ''touched'' and ''inspired,'' with scale anchors of ''not at all'' (1) and ''very much'' (7) (for details, see Tables 1, 2).
The measure for consumer behavioral responses following the company-consumer partnership CSR initiative was derived mainly from interviews with Coop managers and Coop customers. Several qualitative interviews were conducted with customers to help in the development of options representing responses to the CSR initiatives and to ensure content and face validity for the construct. Then, Coop managers tested the resulting pool of five options for self-reported behavior to check for consistency, clarity of wording, and response format. Respondents were asked to select one of the following options for self-reported behavior: (1) ''I did not change my behavior and I continue to buy bottled mineral water''; (2) ''when purchasing bottled mineral water I buy brands from local springs''; (3) ''I'm more inclined to use filtration pitchers and water from public fountains, although in some cases I continue to consume bottled mineral water''; (4) ''I've given up completely with the purchase of bottled mineral water and I consume only purified tap water or water coming from public fountains''; (5) ''I've given up completely with the purchase of bottled mineral water and I consume only tap water without filtering systems.'' We followed the encoding rule that assigns a score from 1 to 5, depending on the answers given by respondents (from 1, if the respondent selected the option, ''I did not change my behavior and I continue to buy bottled mineral water'' corresponding to no response to the CSR campaign, and up to 5, if the option selected was ''I've given up completely with the purchase of bottled mineral water and I only consume tap water without filtering systems,'' corresponding to the maximum response to the CSR campaign). In this way, consumer behavioral responses were measured by a single index along a continuum going from no response to maximum Finally, the measure for consumer support of other green products was adapted from Lee (2009). We used the following three items: ''after this CSR initiative I have started'': ''to be more aware of the environment in my buying behavior considering organic products or environmentally safe products''; ''to worry about environmental issues in my purchases considering for example products that use recycled/recyclable packaging or products that contain no or fewer chemical ingredients''; ''to feel guilty when I don't consider environmentally responsible products.'' The measures thus refer to specific behaviors in support of green product purchases.

Validation of Measures
We ran a confirmatory factor analysis (CFA) using LISREL 8.80 on the measures of all the variables in the model to assess convergent and discriminant validity. The model showed an excellent fit: v 2 (51) = 61.43; RMSEA = 0.03; SRMR = 0.03; NNFI = 0.99; CFI = 1.00 (See Table 1). We also assessed common method variance (CMV) by controlling for the effects of an unmeasured latent method factor (see Podsakoff et al. 2003). We added to the CFA model a new first-order factor explaining all our measures (v 2 (38) = 34.63; RMSEA = 0.01; SRMR = 0.02; NNFI = 1.00; CFI = 1.00). The analysis showed that CMV is not a major problem in our model (see Table 1). None of the loadings of our measures on the common method factor (CMF) were significant (all t values between -0.15 and 1.90). On the contrary, all loadings of the measures on the 6 theoretical latent variables remained positive and significant (p values \0.001). Given these results, it is appropriate to move to tests of our main hypotheses.

Results
In order to test hypotheses, we ran the model illustrated in Fig. 1 1 using structural equation modeling (LISREL 8.80). Table 2 shows the results of the proposed model, which achieved an excellent fit: v 2 (88) = 181.89; RMSEA = 0.06; SRMR = 0.06; NNFI = 0.97; CFI = 0.97. The extrinsic motives positively related to skepticism (0.33, t = 6.71), supporting H1a, while the intrinsic motives negatively related to skepticism (-0.24, t = -3.96), supporting H1b. In turn, felt skepticism was negatively related to consumer behavioral responses following the company-consumer partnership CSR initiative (-0.21, t = -3.60), supporting H1c. Thus, the findings show that consumer skepticism toward the specific CSR initiative mediates the relationships between perceived company CSR motives and consumer behavioral responses, supporting the general hypothesis H1.
Concerning the hypothesized mediating role of felt elevation, we found that intrinsic motives positively related to affective elevation (0.30, t = 4.49), supporting H2b, whereas extrinsic motives did not relate to elevation (-0.09, t = -1.69), hence H2a is not supported. In turn, felt elevation was positively related to consumer behavioral responses following the company-consumer partnership CSR initiative (0.28, t = 4.76), supporting H2c. Thus, findings show that consumer elevation toward the specific CSR initiative mediates the relationships between company CSR intrinsic motives and consumer behavioral responses. This relationship did not hold for the link of CSR extrinsic motives and elevation. Consequently, with one exception, the general hypothesis H2 is supported.
The direct paths linking intrinsic (0.07, t = 1.00) and extrinsic (0.11, t = 1.76) motives to consumer behavioral responses following the specific CSR initiative were nonsignificant. Thus, we can conclude that skepticism and elevation fully mediate the relationship between company CSR motives and consumer behavioral responses, as proposed.
The relationship between consumer CSR responses following the water initiative and consumer support of other green products was tested as well. Consumer behavioral responses positively related to consumer support of other green products (0.17, t = 3.20), supporting H3. We also controlled for the possible relationships between felt elevation and skepticism and support of green products. Findings showed a significant positive relationship of elevation and support of green products (0.28, t = 5.05), while no direct path occurred from skepticism (-0.08, t = -1.47), as expected.
Finally, the influences of the control covariate variable, that is, knowledge of and expertise with green products, on elevation (0.14, t = 2.41), skepticism (-0.17, t = -3.29), and consumer support of green products (0.22, t = 4.15) are significant. By contrast, knowledge of and expertise with green products do not affect consumption of the water initiative (0.01, t = 0.26).

General Discussion
Our research examines the extent and the mechanisms by which a CSR retail grocery initiative to reduce consumption of bottled mineral water promotes both responsible water consumption and support of green purchasing behaviors in general. Specifically, by using measures for self-reported behavioral change, we proposed and empirically demonstrated the positive effect of a company-consumer partnership CSR initiative on two pro-social outcomes; the first strictly related to the environmental cause promoted by the company (i.e., reduction of bottled mineral water consumption), the second related to support of consumer green purchasing behaviors more broadly. In so doing, our research not only advances theoretical understanding of consumption in the area of responsible environmental behavior but also provides insights for 1 We also included as a control variable the ''knowledge of and expertise with green products'' of respondents to take into consideration the possible influence of different levels of individual consciousness about the environment on the variables of our model. We measured this variable using the following 7-point items ''I have a great deal of knowledge about green products'' and ''I have a great deal of experience about green products'' (inter-item correlation = .90).
Effects on Consumption of Green Products 261 marketers interested in maximizing consumer welfare emanating from CSR initiatives.

Theoretical Implications
Our findings provide valuable theoretical and empirical insight into previous research that called for the uncovering of so-called ''secondary'' social outcomes of CSR activities (Bhattacharya and Sen 2004) and the mechanisms underlying such returns. Firstly, our research highlights the role of felt skepticism toward the companyconsumer partnership CSR initiative, as an important obstacle in the adoption of the new sustainable behaviors. Our results, extending prior research (Skarmeas and Leonidou 2013), suggest that the minimization of consumer skepticism is not only a prerequisite for reaping CSR's business benefits but also social ones. Consumers who are skeptical, in fact, react less favorably to CSR initiatives, finding it difficult to undergo certain forms of virtuous conduct. Secondly, our findings add to a growing body of research (e.g., Bratanova et al. 2012;Laham 2009;Reed and Aquino 2003) that implicates moral issues as important motivational sources for engagement in environmentally beneficial activities, and more generally in broad moral pro-social acts. Corporate perceived virtuous support of environmental causes activates the moral emotion of elevation and motivates individuals to support virtuous conduct. It is also worth noting here that our findings show that witnessing good environmental actions by a company triggers the moral emotion of elevation, which can change people's thought-action repertoire and increase the likelihood that they not only show behaviors in line with the CSR initiative, but also engage in similar sustainable behaviors in related contexts with different products. Findings show that mitigating the negative effect of skepticism and enhancing the positive effect of moral elevation come through a proper management of consumer perceptions about perceived company CSR intrinsic and extrinsic motives. Thus, persuading consumers that company CSR actions are disinterested, even altruistic, and not motivated solely by profit is essential to produce positive consumer behavioral responses. These findings therefore confirm the evidence provided by Parguel et al. (2011) about the central role of intrinsic motives for evaluation of corporate CSR initiatives and their general effects on corporate brand evaluations. Finally, our research also contributes to our general understanding of CSR in the social domain by suggesting that a comprehensive picture of the secondary social impact of a company-consumer partnership CSR initiative includes not only the first-order effects strictly related to the initiative (i.e., consumer bottled mineral water consumption in our case), but also second-order effects motivating consumers to more broadly engage in green purchasing behavior beyond advertised initiatives. More specifically, our research provides empirical evidence for positive spillover effects, showing the impact that commitment to a new CSR initiative in terms of water consumption may have on the likelihood of supporting other green products. This finding therefore reinforces the importance to broaden the way CSR initiatives are currently viewed and to go beyond the idea of business returns of such initiatives to include social outcomes. Thus, company-consumer partnership CSR initiatives can really inspire and increase virtuous acts in consumers in its broadest sense.

Managerial Implications
Our findings have important implications for business. Firstly, knowledge about negative consequences of consumer skepticism toward CSR initiatives reveals the importance of companies undergoing efforts to minimize this phenomenon. Especially in company-consumer partnership CSR initiatives, such as the one studied herein, where there exists a predominant focus on formulation and communication of environmental issues (rather than on the company and its products), consumers are more likely to be suspicious of different motives behind company decisions (Du et al. 2010). Accordingly, to build credibility and trustworthiness, companies should emphasize the importance of environmental issues and thus possible CSR impact; also, it should communicate to target audiences the absence of hidden motives. This can be done by strategically choosing issues that are not strictly related to the company business, per se, but rather have broader social implications.
More generally, by properly communicating actual virtuous practices, companies can foster positive returns for themselves, and at the same time, positive changes in the marketplace and environment. Companies can become ''models of CSR excellence,'' proposing new, responsible social projects that consumers can identify with, finding inspiration for defining new life plans and desirable personal commitments. These effects can be amplified by good corporate reputations and, in turn, can build longterm positive company images and thus create a foundation for consumer-company identification and symbiosis.
Our findings suggest one direction for future research. Company efforts to convince consumers that their CSR engagement is authentic and genuine can have important effects, through the moral emotion of elevation, not only for companies implementing positive CSR strategies, but also for society at large. The role of management in facilitating and coordinating CSR company efforts deserves further consideration.

Limitations and Future Research
Our research has some methodologies and other limitations related to the context of the study. Given the fact that the company-consumer partnership CSR initiative was already in place at the time of our study, we cannot infer causality as would be possible by use of a classic controlled experimental design. We tried to reduce the extent of this weakness by controlling for common method bias, employing a methodology correcting for measurement error, testing for rival hypotheses, and controlling for knowledge of and expertise with green products. However, we cannot rule out all alternative explanations for our results. Further research should focus again on an actual company-consumer partnership CSR initiative using a pretest/post-test experimental design or introducing a control group, if possible, to overcome this limitation. At the same time, contextualizing the proposed theoretical model in alternative CSR domains (i.e., where other ethical issues occur) could strengthen the generalizability of our findings. Longitudinal studies of the effects of CSR initiatives are needed to better untangle the direction of causality between variables found to be empirically associated in our research. Comparative studies between firms conducting CSR initiatives and those not doing so, or between firms conducting different degrees of CSR initiatives, could also provide tests of external validity for our hypotheses.
A related shortcoming concerns our measures of consumer behavioral responses following the company-consumer partnership CSR initiative and support of other green purchases. Our measures were self-reports of behaviors. Future research should employ measures of actual behavior if possible.
Our research focused on perceived intrinsic and extrinsic motives of the company under study. But Forehand and Grier (2003) showed that the effects of such attributions can be more nuanced in the sense that consumer skepticism will be reduced when consumers recognize that strategic benefits accrue to a firm, but less-favorable evaluations of a firm result when the firm claims that their self-serving motives were solely public serving. Future research could investigate these and other discrepancies between claims made by firms and consumer perceptions and attributions. Our research did not examine such nuances but rather studied only reactions to perceived motives of the company under study.
Future research should also control for skepticism as a personal trait in the analyses, in order to certify unambiguously the effect of skepticism raised by CSR initiatives from that of the personal trait. Other important elements that can be considered in future studies concern constructs that can limit the effects of consumer skepticism and/or leverage the effect of moral elevation in CSR contexts, at both company and individual levels. For example, at a company level, the role of company reputation or company CSR positioning could affect consumer skepticism and consumer moral elevation. Companies should attempt to undergo actions that explicitly contribute to a positive corporate reputation in the CSR area so as to counter possible invalid consumer skepticism and provide reasons to experience elevation. At an individual level, the role of morally relevant traits as moderators (e.g., empathy, moral identity, relational and collective self) could be considered to complement the role of the two variables we investigated. Morally relevant traits can make consumers more sensitive to intrinsic motives and enhance the effects of elevation. For example, to the extent that empathy is strong, in the sense of a person identifying with CSR and having empathetic concern for people experiencing the harm done to the environment, elevation might be accentuated and have stronger effects. Moreover, considering that our study focused on positive spillover effects closely related to the CSR initiative, future studies should consider the boundary conditions of these effects, investigating for example pro-social behaviors in unrelated domains.