Mark Lett DOI 10.1007/s11002-013-9265-y

Consumer reactions to business-nonprofit alliances: Who benefits and when? Caglar Irmak & Sankar Sen & C. B. Bhattacharya

# Springer Science+Business Media New York 2013

Abstract We investigate the effect of increased company involvement on consumer reactions to companies and nonprofits in business–nonprofit alliances to show that consumer reactions to the two parties in such alliances can, under certain conditions, diverge from each other. Specifically, we show that increased company involvement results in more positive consumer attitudes toward companies with low (but not high) reputation, while it leads to more positive consumer attitude toward nonprofits that partner with companies with high (but not low) reputation. Furthermore, we demonstrate that these effects are independent of the perceived fit between the company and nonprofit forming the alliance. Finally, we show that when consumers elaborate on company motives, the observed effects of increased company involvement are mitigated. Keywords Corporate social responsibility . Business–nonprofit alliance . Nonprofit . Company involvement . Company reputation . Alliance fit

1 Introduction As the nonprofit sector grows faster than both the business and government sectors (Buffett 2013), the concurrent ascent of corporate social responsibility (CSR) in the C. Irmak (*) University of Georgia, 310 Herty Drive, Athens, GA 30602, USA e-mail: [email protected] S. Sen Zicklin School of Business, Baruch College/CUNY, One Bernard Baruch Way, Box 12-240, New York, NY 10010, USA e-mail: [email protected] C. B. Bhattacharya European School of Management and Technology, Schlossplatz 1, 10178 Berlin, Germany e-mail: [email protected]

Mark Lett

corporate sector has resulted in an unprecedented number of alliances between nonprofits and the private sector (i.e., business–nonprofit alliances 1). Today, corporations, including allied foundations, are a key source of support for nonprofits (Gourville and Rangan 2004; Mannell 2010; Seitanidi and Crane 2009), providing over 4 billion USD of support to charitable causes annually (Center for Nonprofit Success 2009). This explosion in business–nonprofit alliances has produced a growing call for a more considered, strategic approach to such alliances on the part of both the corporation and the nonprofit (Andreasen 1996; Kotler and Lee 2004; Porter and Kramer 2002; Ricks and Williams 2005). We respond to this call by providing an understanding of how a critical stakeholder group—the consumer (The McKinsey Global Survey of Business Executives: Business and Society 2006)—reacts to a partnership based on two key characteristics of the company: its reputation and involvement level. Consumer reactions to business–nonprofit alliances are of particular significance to both partners as these underlie both consumers’ patronage of the company (e.g., brand loyalty) and support of the nonprofit (e.g., personal donations of money and/or time). This paper adds to the growing body of research investigating consumer reactions to business–nonprofit alliances (Andreasen 1996; Ellen et al. 2006; Lichtenstein et al. 2004; Sagawa and Segal 2000; Simmons and Becker-Olsen 2006; Yoon et al. 2006) in three ways. First, we focus on the role of an important but thus far unexamined factor—the level of a company’s involvement with the nonprofit—in driving consumer reactions. While intuition suggests that higher levels of involvement would produce more positive consumer reactions to the company and the nonprofit, our research paints a more nuanced picture by demonstrating that such positive effects of increased involvement are jointly contingent on the reputation of the sponsoring company and the alliance member in question. Specifically, while a higher level of involvement helps—in terms of greater consumer regard—a company with a weaker reputation than that with a stronger one, greater involvement from a company with a stronger reputation helps the nonprofit more—in terms of greater consumer support—than involvement from a company with a weaker reputation. Second, in line with the vast literature on inference making (see Kardes et al. 2008 for a recent review), particularly about marketers and their motives (e.g., Campbell and Kirmani 2008), we implicate consumers’ inferences about both the company’s motives guiding its involvement level (Ellen et al. 2006; Sagawa and Segal 2000) as well as the nonprofit’s likelihood of success (Andreasen 1996) as drivers of their aforementioned asymmetric reactions to the alliance members. Finally, we shed some light on the conditions under which a company with a relatively weak reputation can best benefit, in terms of favorable consumer reactions, from a higher level of alliance involvement; consumer focus on the motives guiding such a company not only diminishes their regard for the company but also, more specifically, erases the potential gains from a higher level of involvement. 1 While business–nonprofit alliances can be of four types—business–nonprofit, business–government, government–nonprofit, and trisector (Selsky and Parker 2005)—in this research, we use the term “business–nonprofit” to refer only to alliances between corporations and nonprofits.

Mark Lett

Next, we theorize about the effects of company involvement level and reputation on consumer reactions to each alliance member and test the emerging predictions in two experiments. We end with a discussion of our findings and some thoughts regarding future research directions.

2 Conceptual development Company involvement, defined broadly as the nature and amount of company resources devoted to the business–nonprofit alliance, can range from a small monetary contribution (i.e., philanthropy) to a more substantial, effortful, and sustained commitment of firm resources, with significant integration between the partners’ missions, people, and activities (Berger et al. 2004). Prior research (Andreasen 1996; Ellen et al. 2006; Lichtenstein et al. 2004) matches intuition in suggesting a positive relationship between involvement level and consumers’ company evaluations because the former comprises the basis for consumer inferences regarding a company’s benevolence, sincerity, and commitment to the focal cause, making them more likely to identify with the company. Greater corporate involvement is likely to produce more positive reactions to the nonprofit as well because the involvement level serves as an inferential cue for not only the importance of the nonprofit, including its cause, but also its likelihood of achieving its goals (i.e., perceived effectiveness of the nonprofit; Andreasen 1996). These effects of firm involvement are consistent with the broader work on brand alliances (Park et al. 1996; Rao et al. 1999), which points to the positive synergistic effect of strong alliances on consumer preference, wherein the alliance signals higher overall quality than that of each individual alliance member (i.e., the sum is greater than its parts). Thus, when corporate involvement is higher, the greater strength of the alliance is likely to reflect positively on the nonprofit’s effectiveness and, consequently, value (Long and Chiagouris 2006). More interestingly, these effects of company involvement are likely to vary with the reputation of the company. We discuss the moderating effect of reputation next. 2.1 Moderating role of company reputation A primary benefit to a company of entering into an alliance with a nonprofit is a gain in reputation (Bhattacharya and Sen 2004; Brammer and Millington 2005; Du et al. 2007; Fombrun 2005), defined as stakeholders’ “collective judgments of a corporation based on assessments of the financial, social, and environmental impacts attributed to the corporation over time” (Barnett et al. 2006). As well, a company’s reputation has been well documented to influence consumers’ subsequent reactions to its CSR efforts (Lai et al. 2010; Yoon et al. 2006). In our focal context, we expect a company’s extant reputation to similarly affect the extent to which increases in company involvement alter consumers’ evaluations of both alliance members. In particular, we draw on extant theories of causal inference making (Kelley 1967; see Kardes et al. 2008 for recent review) and, more specifically, the accessibility– diagnosticity framework (Lynch et al. 1988) to argue that when a company’s involvement level increases, consumers will react more favorably to it when it is has a weak

Mark Lett

reputation than when it has a strong one. This is because, given the extant favorable perceptions of a strong reputation company, its inherently positive alliance involvement information is unlikely to be as diagnostic as in the case of a weak reputation company for which the perceptions are not as favorable. More specifically, prior research suggests that consumers are likely to interpret a company’s level of involvement with a nonprofit as an indication of its sincerity and genuine concern for the nonprofit/cause (Sagawa and Segal 2000). Moreover, such appraisals of genuine concern or intrinsic attributions (Fein 1996) are likely to play a driving role in the involvement-based company evaluations made by consumers (Alcañiz et al. 2010; see Campbell and Kirmani 2008 for recent review). A company with a strong reputation is likely to garner favorable intrinsic attributions by sole virtue of its reputation, making consumers’ causal attributions and consequent evaluations of it less sensitive to actual involvement levels. In fact, in line with their prior perceptions about such companies, consumers may actually expect high levels of involvement from them. On the other hand, consumers are more likely to be skeptical about the motives underlying the alliance participation of companies with weak reputations (Elving 2013; Yoon et al. 2006). Importantly, prior research suggests that in the case of such companies, these unfavorable causal attributions may be mitigated by greater involvement, which shifts consumers’ causal inferences toward genuine concern and sincerity (Sanbonmatsu et al. 1989; Silvera and Laufer 2005; Yoon et al. 2006). In other words, for a weak reputation company, the involvement level is likely to comprise much more of a diagnostic cue for its genuine concern for the cause (Kardes et al. 2008; Weiner 1980), making consumer evaluations particularly responsive to the specific level of involvement. In sum, then we expect an increase in a company’s involvement level to be more likely to increase consumers’ company attitudes when the company has a weak reputation than when it has a strong reputation. From a nonprofit’s perspective, however, the extant reputation of the company is an important criterion in deciding who to partner with (Amos 2005; Andreasen 1996; Gourville and Rangan 2004; Heller 2008). This is consistent with brand alliance research which suggests that because strong brands convey higher quality information to consumers, when a reputable brand allies with another, perhaps lesser, brand, the alliance signals unobservable quality to consumers (Rao et al. 1999). More specifically, to the extent that the operations and effectiveness of the nonprofit are largely unobservable to consumers, a more integrative type of partnership with a highly reputable company (i.e., higher investment of a reputable company in a nonprofit) is likely to be a particularly strong signal regarding nonprofit effectiveness, eliciting, in turn, more positive consumer evaluations. Thus, we expect that compared to a lower level of involvement, higher involvement from a reputable company will cause consumers to perceive the nonprofit as more effective, leading to more positive nonprofit attitudes and support. In the case of a company with a weak reputation, on the other hand, such effectiveness inferences are less likely to vary with varying levels of involvement due to consumers’ intrinsic lack of confidence, relatively speaking, in the abilities of such a company to, intentions aside, actually alter the fortunes of its alliance partner. In sum, we expect an increase in a company’s involvement level to be more likely to increase consumers’ nonprofit attitudes and support when the

Mark Lett

company has a strong reputation than when it has a weak reputation. Furthermore, we expect this interactive effect of involvement level and reputation to be mediated by consumers’ inferred effectiveness of the nonprofit. Our research focuses on the joint roles of company reputation and involvement levels in consumers’ reactions to each member of a business–nonprofit alliance. Notably, a large body of extant research also implicates consumers’ perceived fit between the nonprofit/cause and the company, defined as the perceived congruity or relatedness between the nonprofit and the company (Barone et al. 2007; Simmons and Becker-Olsen 2006), as a key driver of consumer reactions to such alliances. In general, greater fit is thought to produce more favorable reactions toward both partners than lower fit (Amos 2005; Andreasen 1996; Gourville and Rangan 2004; Simmons and Becker-Olsen 2006). Because fit is likely to, in general, matter in business–nonprofit alliances, we also examine its effects in our studies. Importantly, however, some recent research (Lafferty 2007) points to the relative irrelevance of fit in consumer reactions to such alliances when they explicitly consider corporate credibility, in terms of, say, company reputation (Lafferty 2007). Therefore, we do not expect consumers’ fit perceptions to interact with the focal drivers (i.e., company reputation, level of involvement) of consumer reactions to business–nonprofit alliances. Next, we report two studies that test our predictions.

3 Study 1 The purpose of this study was to test our expectations regarding the interaction between company reputation and its level of involvement on both nonprofit and company attitudes. We also manipulated fit to examine its role in consumers’ reactions to these alliance members. 3.1 Method Three hundred and one undergraduates participated in the study in partial fulfillment of course requirements. The experiment utilized a 2 (involvement: low vs. high)× 2 (Reputation: Weak vs. Strong)× 2 (fit: low vs. high) betweensubjects design. Participants began the study by reading two different excerpts. The first excerpt, supposedly from the web site of a nonprofit organization, provided information about a nonprofit organization (VIDA) and its focus on AIDS prevention and care. The second excerpt first introduced a computer (low-fit condition) or a pharmaceutical (high-fit condition) company with low or high reputation, then provided information about the company’s involvement level (low or high) with the nonprofit organization. In line with prior research (Simmons and Becker-Olsen 2006), we operationalized fit as the match between the company’s and nonprofit’s domains of operation. Reputation level of the company was manipulated by describing Company X in the strong-reputation condition as a company that has a sterling corporate reputation, making it to the top 10 list of Fortune’s annual survey of the world’s Most Admired Companies 14 times in the last 15 years. In the weak-reputation condition, Company X was

Mark Lett

described as a company that has yet to establish a credible corporate reputation, never having gained mention in Fortune’s annual survey of the world’s Most Admired Companies. Finally, in the high-involvement condition, the company was presented to pledge US \$1.7 million (a sum about half of VIDA’s operating budget) and leverage its marketing expertise to benefit the nonprofit, whereas in the lowinvolvement condition, the company was presented to support the nonprofit only financially and to a far lesser extent (US \$8,000, a sum about 2 % of the operating budget of VIDA). The level of involvement manipulation was based on the annual financial support levels that companies provided nonprofits in the real marketplace. After participants read both excerpts, they provided their attitude toward the nonprofit by rating VIDA on a four-item seven-point scale, which started with “VIDA is an organization…” (1=I do not like at all, 7=I like very much; 1=Does unimportant work, 7=Does important work; 1=I would definitely not support, 7=I would definitely support; 1=Is not at all legitimate, 7=Is very legitimate). These items were averaged to form a composite measure for attitude toward the nonprofit (Cronbach’s α=0.90). Following the attitude measure, respondents provided their donation likelihood (“How likely or unlikely would you be to donate money to VIDA?”, 1=Extremely unlikely, 7=Extremely likely). Next, perceived effectiveness of the nonprofit was assessed using a two-item seven-point scale: “How likely or unlikely do you think it is that VIDA will be able to fulfill its mission?” (1=Extremely unlikely, 7=Extremely likely) and “How successful do you think VIDA will be in helping AIDS prevention and care in the United States?” (1=Extremely unsuccessful, 7=Extremely successful). These two items were averaged to form a composite measure of perceived nonprofit effectiveness (r=0.66, p<0.01). Next, we measured attitude toward the company by asking participants their overall opinion of Company X (1=very unfavorable, 7=very favorable). Then, two items were used as the manipulation check for fit (“In thinking about the alliance between Company X and VIDA, how would you characterize the “fit” between these two organizations?”, 1=Poor fit, 7=Excellent Fit; “Overall, how good is the match between the VIDA’s cause (AIDS prevention and care) and its corporate sponsor, Company X?”, 1 =Very poor match, 7=Very good match). Respondents’ ratings of these items were averaged (r=0.69, p<0.01). The success of the involvement manipulation was assessed using two items (“How would you characterize Company X’s level of involvement with VIDA?”, 1=Very low level of involvement, 7=Very high level of involvement; “Is Company X doing very little or a lot to help VIDA achieve its mission?”, 1=Doing very little, 7=Doing a lot). The ratings of these items were averaged to form the manipulation check measure for involvement (r=0.86, p<0.01). The success of the reputation manipulation was assessed by asking respondents “What type of a reputation does Company X have?” (1=Poor reputation, 7=Excellent reputation). Finally, credibility of the excerpts was assessed using a three-item seven-point scale (“In your opinion, the information you read from the VIDA web site was…” (1=Not at all believable, 7=Highly believable; 1=Not at all credible, 7=Very credible; 1=Not at all accurate, 7=Absolutely accurate). The study ended with questions on past donation behavior and demographics.

Mark Lett

3.2 Results and discussion Before testing our predictions, we conducted ANOVAs with involvement, reputation, fit, and all their interactions as the predictors of all three manipulation check measures. The results showed that involvement had a significant main effect on its manipulation check measure [M low-inv = 3.47, M high-inv = 5.86, F(1,300)=294.05, p<0.01], while no other effect was significant (all p>0.17). Similarly, fit had a significant main effect on its manipulation check measure [Mlow-fit =4.17, Mhigh-fit =4.96, F(1,300)=29.61, p<0.01]. Finally, reputation had a significant main effect on its manipulation check measure [Mlow-rep =3.04, Mhighrep =6.34, F(1,300)=640.87, p<0.01]. While none of the main effects were significant (all p>0.05), the interactive effect of reputation and involvement was significant [F(1,300)=8.26, p<0.01]. Since the effect size of this interactive effect (ω2 =0.0072) was significantly less than the main effect of reputation (ω2 =0.65), we concluded that the manipulations were successful (Perdue and Summers 1986). To test our predictions, we first conducted an ANOVA with involvement, fit, reputation, and all interactions as predictors of attitude toward the company. Supporting our prediction, a significant interaction of involvement and reputation was found on attitude toward the company [F(1,300)=5.13, p<0.05]: attitude toward low-reputation companies was higher when the involvement level was high than when it was low [M high-inv = 4.86, M low-inv = 4.13, F(1,149) = 13.29, p < 0.01]. However, attitude toward high-reputation companies did not change with involvement level [Mhigh-inv =5.75, Mlow-inv =5.60, F(1,150)=0.94, p>0.33]. As well, an ANOVA on nonprofit attitude with involvement, reputation, fit, and all interactions as predictors revealed a significant interaction of reputation and involvement [F(1,300)=4.54, p<0.05]. As expected, when the nonprofit partnered with a high-reputation company, attitude toward the nonprofit was more positive under high involvement (M=6.20) than under low involvement [M=5.70, F(1,150)=11.16, p<0.01]. In the case of the low-reputation company, however, the involvement level did not influence nonprofit attitude [M low-inv = 5.92, M high-inv = 5.95, F(1,149) = 0.06, p > 0.80]. The same pattern was observed in the analogous ANOVA for donation likelihood [F(1,300)=1.79, p=0.18]. When the nonprofit partnered with a high-reputation company, consumers’ donation likelihood was greater in the high-involvement condition (M=4.70) than in the low-involvement one [M=3.92, F(1,0 150)=9.84, p<0.01]. In the case of a low-reputation company, however, the involvement level did not influence donation likelihood [Mlow-inv =3.91, Mhigh-inv =4.20, F(1,149)=1.40, p>0.23]. Finally, we conducted bootstrapping analyses (Preacher and Hayes 2008) which revealed that the interaction of involvement and reputation had a significant indirect effect through nonprofit effectiveness on attitude toward the nonprofit (β=0.22, 95% CI=0.10–0.37), lending support to the predicted mediation. As expected, fit did not influence the focal reputation×involvement interaction for company attitude, nonprofit attitude, or donation likelihood. However, in line with prior research (Menon and Kahn 2003), the ANOVAs on nonprofit attitude and monetary donation revealed significant interactions of fit and involvement [F(1,300)=4.87, p<0.05; F(1,300)=5.48, p<0.05, respectively]. Follow-up contrasts showed that, in the low-fit condition, nonprofit attitude (M=6.21) and donation

Mark Lett

likelihood (M=4.61) were significantly higher when the company had a high level of involvement with the nonprofit than when it had a low level of involvement [M=5.74, F(1,148)=11.69, p<0.01; M=3.66, F(1,148)=15.19, p<0.01, respectively]. In the high-fit condition, on the other hand, there was no effect of company involvement level on nonprofit attitude [Mlow-involve =5.88, Mhigh-involve =5.94, F(1,151)<1, p>0.60] and donation likelihood [Mlow-involve =4.16, Mhigh-involve =4.29, F(1,151)<1, p>0.50]. Furthermore, bootstrapping analyses (Preacher and Hayes 2008) revealed that the interaction of involvement and fit had a significant indirect effect through nonprofit effectiveness on attitude toward the nonprofit (β=−0.26, 95% CI=−0.48 to −0.13), indicating that the interactive effect of increased company involvement level and alliance fit on attitude toward the nonprofit is mediated by perceived effectiveness of the nonprofit. The results of study 1 suggest that low-reputation companies have more to gain from increased involvement in a business–nonprofit alliance. Based on this, low-reputation companies might be tempted to encourage their consumers to focus as much as possible on their strong support of nonprofits, including elaborating on their potential motives for such involvement. Interestingly, however, such a tactic may backfire. Specifically, prior research (Menon and Kahn 2003) has shown that elaboration on a company’s motives for engaging in CSR lowers the positive ratings consumers typically give to such companies. This is due, at least in part, to the negative attributions about the company that surface during such elaboration, something that is even more likely when the company has a weak reputation to begin with (Campbell and Kirmani 2008). Specifically, when the company reputation is weak, elaboration on motives is likely to result in causal attributions that assimilate to the extant reputation (Herr et al. 1991) increasing the salience of extrinsic motives, such as reputation enhancement, and, thus, causing the company to be perceived as more insincere in its involvement with the nonprofit (Yoon et al. 2006). Thus, we expect, based on prior research (Barone et al. 2000; Friestad and Wright 1994; Menon and Kahn 2003; Yoon et al. 2006), that when consumers elaborate on the reasons for a weak-reputation company’s high involvement with a nonprofit, they will, ironically, be less likely to make intrinsic causal attributions regarding the company’s involvement with the nonprofit, negating the previously documented attitude gain from the increased involvement level. We test for this interactive effect of elaboration and involvement level on consumer attitudes toward weak-reputation companies in study 2, described next.

4 Study 2 This study had three objectives. First, we wanted to replicate the findings of study 1 for weak-reputation companies. Second, we wanted to test for the theorized effect of elaboration about company motives on the company involvement—attitude link for weak-reputation companies. Relatedly, we wanted to demonstrate the driving role of intrinsic (i.e., genuine concern) attributions in this effect of elaboration. Finally, we wanted to enhance both the internal and external validity of our findings by examining consumers’ actual donations to two different nonprofits.

Mark Lett

4.1 Method One hundred and forty-eight undergraduates participated in the study and were paid US \$10 for their participation. The study utilized a 2 (involvement: high vs. low)×2 (fit: high vs. low)×2 (elaboration: high vs. low) between-subjects design. The procedure employed, stimuli used, and measures collected were identical to those in study 1, with the following exceptions. First, in line with the objective of the study, we used only low-reputation companies. Second, we manipulated participants’ elaboration on company motives by asking approximately half the participants an open-ended question before they provided their attitudes toward the company and the nonprofit: “Why do you think Company X is supporting VIDA? In the space below, please tell us what you think are the reasons behind Company X’s support of this nonprofit.” The other half of the participants responded to this question at the end of the study. Third, to enhance the internal validity and generalizability of our results, we used a crossed design wherein two different nonprofits were paired with two different companies, both with low reputation. Specifically, in the high-fit condition, we had either a pharmaceutical company partnering with a nonprofit (VIDA) focused on AIDS prevention and care or a computer company partnering with a nonprofit (VIDA) focused on educational opportunities for underprivileged youth. In the lowfit condition, the pharmaceutical company was paired with the nonprofit focused on educational opportunities for underprivileged youth, whereas the computer company was paired with the nonprofit focused on AIDS prevention and care. Because this design involved two causes, we also controlled for differences in their perceived importance. At the beginning of the study, participants rated the importance of several social issues (1=Do not care at all, 7=Care a lot). Embedded in these issues were “HIV/AIDS prevention and care” and “educational opportunities for underprivileged youth.” Participants’ ratings of the issue they were exposed to served as a covariate in all relevant analyses. Fourth, participants responded to several questions pertaining to their attributions of company motives. Specifically, intrinsic attributions was measured by the average of three items [To what extent do you believe each of the following factors guided Company X’s decision to support VIDA: “Company X’s genuine support for HIV/AIDS-related causes (computer literacy for underprivileged youth),” “Company X’s desire to help people living with AIDS (underprivileged youth),” “To what extent do you think Company X supports VIDA because it is truly a socially responsible company?” (1=Not At All, 7=Completely, Cronbach’s α=0.88)]. Finally, we measured actual donation behavior. Participants received their payment in the beginning of the experiment as one US \$5 and five US \$1 bills in order to provide them a virtual donation scale and make donation easier (as expected, the range of donation amount was 0–10). An envelope, attached to the end of the questionnaire, contained a solicitation flyer. Participants were told about the envelope and the flyer after responding to the questions regarding the nonprofit attitude. They were asked to open the envelope, read the flyer, and decide whether or not they would like to donate some or all of the money they received for their participation. If participants chose to donate, they indicated the donation amount in the flyer and put the flyer back in the

Mark Lett

envelope along with the donation amount itself. If they chose not to donate any money, they put a “zero” on the donation amount line and put the flyer back in the envelope. Next, similar to previous studies, we measured attitude toward the company by asking participants their overall opinion of Company X (1=Very unfavorable, 7=Very favorable). Finally, participants sealed and returned the envelope to the researchers along with the completed surveys, at which time they were debriefed about the study and, if they had donated money to VIDA, their donation returned to them. 4.2 Results and discussion Before testing our predictions, we assessed the success of our involvement and fit manipulations. An ANOVA with involvement, fit, elaboration, and their interactions as predictors of the manipulation check measure of involvement showed a significant main effect of involvement [Mlow-inv =3.92, Mhigh-inv =5.11, F(1,147)=35.46, p<0.0001] and a significant main effect of fit [F(1,147)=4.63, p<0.05]. All other effects were non-significant (all p>0.05). Further analyses revealed no significant interactive effects of fit with any of the independent variables (all p>0.17). Since the effect size of the main effect of fit (ω2 =0.018) was significantly smaller than the effect size of involvement (ω2 = 0.18), we concluded that the manipulation of involvement was successful (Perdue and Summers 1986). ANOVA results pertaining to the manipulation check of fit showed a significant main of effect of fit [Mlow-fit = 3.70, Mhigh-fit = 4.86, F(1,147) = 33.46, p < 0.0001], while all other effects were non-significant (p>0.05). Thus, our manipulation of fit was successful. An ANOVA with involvement, fit, elaboration, and all interactions as predictors of attitude toward the company revealed a main effect of elaboration [Mlow-elab =4.64, Mhigh-elab =3.82, F(1,147)=8.08, p<0.01] and a significant interaction of elaboration and involvement [F(1,147)=4.03, p<0.05] such that higher involvement led to more positive company attitude in the low-elaboration condition [Mlow-inv =4.39, Mhigh-inv = 4.89, F(1,71) = 2.38, p < 0.05], but did not alter company attitude in the highelaboration condition [Mlow-inv =3.70, Mhigh-inv =3.94, F(1,75)=0.57, p>0.90]. Thus, our elaboration prediction was supported. To test for the mediating role of intrinsic attributions, we conducted bootstrapping analyses (Preacher and Hayes 2008). In line with our predictions, the results revealed that the interaction of involvement and elaboration had a significant indirect effect through intrinsic attributions on attitude toward the company (β=0.24, 95% CI=0.017–0.51). As elaboration about company motives is likely to shift participants’ focus away from nonprofits (Menon and Kahn 2003), we did not have specific hypotheses about the effect of elaboration about company motives on consumer reactions to nonprofits. Still, we wanted to investigate the effect of such elaboration on consumers’ nonprofit evaluations and conducted separate ANOVAs with elaboration, involvement, and fit on attitude toward the nonprofit and donation amount, respectively. The results revealed a significant three-way interaction of elaboration, involvement, and fit on nonprofit attitude [F(1,147)= 3.25, p = 0.07]. To understand the meaning of this three-way interaction, we conducted separate ANOVAs in the high- and low-elaboration conditions.

Mark Lett

An ANOVA of involvement and fit in the low-elaboration condition on nonprofit attitude revealed a two-way interaction of involvement and fit [F(1,71)=4.13, p<0.05], replicating the finding of study 1. Specifically, in the low-fit condition, attitude toward the nonprofit was significantly more positive when the company had a high level of involvement with the nonprofit (M=5.81) than when it had a low level of involvement [M=5.29, F(1,34)=8.72, p<0.01]. On the other hand, increased involvement did not have a significant effect on nonprofit attitude in the high-fit condition [Mlow-inv =5.30, Mhigh-inv =5.57, F(1,36)=0.03, p>0.86]. The same pattern of results was observed for the donation amount, although the interactions did not reach significance. Specifically, in the low-fit condition, donation amount was significantly higher when the company had high (vs. low) involvement with the nonprofit [Mlow-inv =\$.53, Mhigh-inv =\$1.22, F(1,34)=5.01, p<0.05], but it did not vary significantly with company involvement level in the high-fit condition [Mlow-inv =\$.21, Mhigh-inv =\$.78, F(1,36)=1.74, p>0.19]. Separate ANOVAs of involvement and fit on nonprofit attitude and donation amount, respectively, in the high-elaboration condition revealed no significant main effects (all p>0.06) or interactive effects of involvement and fit (all p>0.40), indicating that when consumers elaborate on company motives, the interactive effect of involvement and fit on consumer reactions to nonprofits diminishes. Thus, it appears that as consumers’ intrinsic attributions deteriorate under high elaboration, they may be less likely to believe that the company’s support of the nonprofit will increase the latter’s ability to achieve its goals. Specifically, they may perceive the company’s involvement, no matter how substantial, to be a temporary endeavor, preventing the nonprofit from fully achieving its goals. While we generally replicated the effects observed in prior studies, we did not find the expected significant mediating effect of nonprofit effectiveness on donation amount in the low-elaboration condition in this study. In part, the smaller sample size in study 2 may be at play, but overall, these results suggest that there may be other significant factors that influence the link between donation intentions and donation behavior in this context, which can be investigated by future research.

5 General discussion Across two studies, we demonstrate that increased company involvement in business–nonprofit alliances may, under certain conditions, result in asymmetric effects on consumer reactions to companies and nonprofits entering into such alliances. Specifically, we show that companies with weak (vs. strong) reputation can elicit a more positive consumer attitude when they increase their involvement with the nonprofit, whereas nonprofits garner relatively more positive consumer attitude and support when they have high-involvement partnerships with companies that have a relatively strong (vs. weak) level of reputation. These findings build on the emerging research on business–nonprofit alliances (Lichtenstein et al. 2004; Simmons and Becker-Olsen 2006) and, more broadly, add to the literature on CSR (Sen and Bhattacharya 2001) by investigating how changes in company involvement level interact with another relevant dimension of the alliance, the company reputation, to influence consumer reactions to each alliance member. Importantly, we find that the process through which increased company involvement works differs for the

Mark Lett

company and the nonprofit, leading to asymmetries in its effect on consumer reactions to the two parties. Specifically, while increased involvement by a strong-reputation company has a positive effect on consumer attitude toward a nonprofit by increasing perceived effectiveness of the nonprofit, it improves consumer reactions to a weakreputation company by heightening consumers’ intrinsic attributions of the company’s motives in supporting the nonprofit. Future research may further investigate why two different processes underlie consumer evaluations of companies and nonprofits. One potential explanation is that consumer inferences are influenced by the immediate goal of the actors (Campbell and Kirmani 2000; Reeder et al. 2004): because the goal of the nonprofit is to help the cause it works for, it is evaluated based on how the alliance helps its likelihood of achieving those goals (i.e., its effectiveness); on the other hand, the company’s immediate goal is maximizing profits, and an alliance with the nonprofit may make consumers think about the company’s motives in partnering with the nonprofit. Prior work on the role of company reputation in CSR (Yoon et al. 2006) showed that CSR initiatives undertaken by companies with a negative, rather than merely weak, reputation can hurt a company’s image as consumers perceive such initiatives as patently insincere. Building on past work, our work suggests that for companies with weak rather than negative reputations per se, increased company involvement can help overcome consumers’ suspicion about their motives for alliance formation. Importantly, consumers’ elaboration on company motives comprises a boundary condition for this positive effect of increased company involvement level on consumer attitude toward weak-reputation companies. In fact, these findings dovetail with the findings of Yoon et al. (2006) given that companies with bad, as opposed to weak, reputations are more likely to trigger suspicion and make consumers elaborate on company motives. On the other hand, we found that nonprofits can receive greater consumer support when they form high-involvement partnerships with companies with strong (vs. weak) reputations. Together with the results regarding the effect of alliance fit on consumer reactions to nonprofits, these results suggest that nonprofits are likely to increase consumer support as long as they increase their perceived ability to achieve their intended goals. Future research may investigate factors other than alliance fit, company reputation, and involvement that may affect consumers’ perception of nonprofit effectiveness. One interesting factor may be the type, rather than the level, of company involvement with the nonprofit. For instance, given the findings from this research, we expect that nonprofits that need human resources may partner with companies that can provide volunteering personnel, whereas those that have sufficient workforce but need financial resources may ally with corporations that donate money. Such partnerships may redefine the alliance fit around the needs of the nonprofit to achieve its intended goals and, with the right communication to the public, may receive greater consumer support. More broadly, this research focused on the effect of company reputation on consumer reactions to business–nonprofit alliances. An interesting avenue for future research would be how nonprofit reputation, rather than company reputation or in interaction with it, affects consumer reactions to business–nonprofit alliances. As the nonprofit sector develops its own brands (e.g., Livestrong), sooner or later, companies will strive to partner with reputable nonprofits (Quelch et al. 2004). In fact, this

Mark Lett

new phase of business–nonprofit alliances may prove to be more beneficial to companies than to nonprofits.

References Alcañiz, E., Cáceres, R., & Pérez, R. (2010). Alliances between brands and social causes: the influence of company credibility on social responsibility image. Journal of Business Ethics, 96(2), 169–186. Amos, N. (2005). The new world of corporate giving. Corporate Responsibility Management, 2, 34–37 (August/September). Andreasen, A. R. (1996). Profits for nonprofits: find a corporate partner. Harvard Business Review, 74(6), 47–59. Barnett, M. L., Jermier, J., & Lafferty, B. A. (2006). Corporate reputation: the definitional landscape. Corporate Reputation Review, 9(Spring), 26–38. Barone, M. J., Miyazaki, A. D., & Taylor, K. A. (2000). Does one good turn deserve another? Examining the influence of cause-related marketing efforts on consumer choice. Journal of the Academy of Marketing Science, 28(2), 250–264. Barone, M. J., Norman, A. T., & Miyazaki, A. D. (2007). Consumer response to retailer use of cause-related marketing: is more fit better? Journal of Retailing, 83(4), 437–445. Berger, I. E., Cunningham, P. H., & Drumwright, M. E. (2004). Social alliances: company/nonprofit collaboration. California Management Review, 47(Fall), 58–90. Bhattacharya, C. B., & Sen, S. (2004). Doing better at doing good: when, why and how consumers respond to corporate social initiatives. California Management Review, 47(Fall), 9–24. Brammer, S., & Millington, A. (2005). Corporate reputation and philanthropy: an empirical analysis. Journal of Business Ethics, 61(1), 29–44. Buffett, P. (2013). The charitable–industrial complex. The New York Times, July 26, Opinion pages. Retrieved from http://www.nytimes.com/2013/07/27/opinion/the-charitable-industrial-complex.html?_r=0. Accessed 13 Oct 2013. Campbell, M. C., & Kirmani, A. (2000). Consumers’ use of persuasion knowledge: the effects of accessibility and cognitive capacity on perceptions of an influence agent. Journal of Consumer Research, 27(June), 69–83. Campbell, M. C., & Kirmani, A. (2008). I know what you’re doing and why you’re doing it: the use of persuasion knowledge model in consumer research. In C. P. Haugtvedt, P. M. Herr, & F. R. Kardes (Eds.), Handbook of consumer psychology (pp. 549–571). New York, NY: Lawrence Erlbaum. Center for Nonprofit Success (2009). Cause marketing: building profitable relationships with corporate partners. Retrieved from http://www.cfnps.org/Chicago_Fundraising_Summit_56_16.aspx. Accessed 13 Oct 2013. Du, S., Bhattacharya, C. B., & Sen, S. (2007). Reaping relational rewards from corporate social responsibility: the role of competitive positioning. International Journal of Research in Marketing, 24(3), 224–241. Ellen, P. S., Weber, D. J., & Mohr, L. A. (2006). Building corporate associations: consumer attributions for corporate socially responsible programs. Journal of the Academy of Marketing Science, 34(2), 147– 157. Elving,W. J. L. (2013). Skepticism and corporate social responsibility communications: the influence of fit and reputation. Journal of Marketing Communications, 19, 277–292. doi:10.1080/13527266.2011.631569 Fein, S. (1996). Effects of suspicion on attributional thinking and the correspondence bias. Journal of Personality and Social Psychology, 70(6), 1164–1184. Fombrun, C. J. (2005). Building corporate reputation through CSR initiatives: evolving standards. Corporate Reputation Review, 8(Spring), 7–12. Friestad, M., & Wright, P. (1994). The persuasion knowledge model: how people cope with persuasion attempts. Journal of Consumer Research, 21, 32–54. Gourville, J. T., & Rangan, V. K. (2004). Valuing the cause marketing relationship. California Management Review, 47(Fall), 38–57. Herr, P. M., Kardes, F. R., & Kim, J. (1991). Effects of word-of-mouth and product attribute information on persuasion: an accessibility–diagnosticity perspective. Journal of Consumer Research, 17(March), 454– 462.

Mark Lett Kardes, F. R., Posavac, S. S., Cronley, M. L., & Herr, P. M. (2008). Consumer inference. In C. P. Haugtvedt, P. M. Herr, & F. R. Kardes (Eds.), Handbook of consumer psychology (pp. 165–191). New York, NY: Lawrence Erlbaum. Kelley, H. H. (1967). Attribution theory in social psychology. In D. Levine (Ed.), Nebraska Symposium on Motivation, vol. 15 (pp. 192–240). Lincoln: University of Nebraska Press. Kotler, P. & Lee, N. (2004). When it comes to gaining a market edge while supporting a social cause, ‘corporate social marketing’ leads the pack. Stanford Social Innovation Review Spring, pp. 14–23. Lafferty, B. A. (2007). The relevance of fit in a cause–brand alliance when consumers evaluate corporate credibility. Journal of Business Research, 60(5), 447–453. Lai, C., Chiu, C., Yang, C., & Pai, D. (2010). The effects of corporate social responsibility on brand performance: the mediating effect of industrial brand equity and corporate reputation. Journal of Business Ethics, 95, 457–46 (September). Lichtenstein, D., Drumwright, M., & Bridgette, B. M. (2004). The effect of corporate social responsibility on customer donations to corporate-supported nonprofits. Journal of Marketing, 68, 16–32 (October). Long, M. M., & Chiagouris, L. (2006). The role of credibility in shaping attitudes toward nonprofit websites. International Journal of Nonprofit & Voluntary Sector Marketing, 11, 239–249 (August). Lynch, J. G., Marmorstein, H., & Weigold, M. F. (1988). Choices from sets including remembered brands: use of recalled attributes and prior overall evaluations. Journal of Consumer Research, 15, 169–184. Mannell, J. (2010). Are the sectors compatible? International development work and lessons for a business– nonprofit partnership framework. Journal of Applied Social Psychology, 40(5), 1106–1122. Menon, S., & Kahn, B. E. (2003). Corporate sponsorships of philanthropic activities: when do they impact perception of sponsor brand? Journal of Consumer Psychology, 13(3), 316–327. Park, C. W., Jun, S. Y., & Shocker, A. D. (1996). Composite branding alliances: an investigation of extension and feedback effects. Journal of Marketing Research, 33(4), 453–466. Perdue, B. C., & Summers, J. O. (1986). Checking the success of manipulations in marketing experiments. Journal of Marketing Research, 23(November), 317–326. Porter, M., & Kramer, M. (2002). The competitive advantage of corporate philanthropy. Harvard Business Review, 80(12), 56–68. Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879–891. Quelch, J. A., Austin, J. E., & Laidler-Kylander, N. (2004). Mining gold in not-for-profit brands. Harvard Business Review, 82(4), 24–24. Rao, A. R., Qu, L., & Ruekert, R. (1999). Signaling unobservable product quality through a brand ally. Journal of Marketing Research, 36(2), 258–268. Reeder, G. D., Vonk, R., Ronk, M. J., Ham, J., & Lawrence, M. (2004). Dispositional attribution: multiple inferences about motive-related traits. Journal of Personality and Social Psychology, 86(4), 530–544. Ricks, J. M., Jr., & Williams, J. A. (2005). Strategic corporate philanthropy: addressing frontline talent needs through an educational giving program. Journal of Business Ethics, 60, 147–157. Sagawa, S., & Segal, E. (2000). Common interest, common good: creating value through business and social sector partnerships. California Management Review, 42(2), 105–122. Sanbonmatsu, D. M., Kardes, F. R., & Gibson, B. D. (1989). The impact of initial processing goals on memory based brand comparisons. Advances in Consumer Research, 16, 429–432. Seitanidi, M. M., & Crane, A. (2009). ‘Implementing CSR through partnerships: understanding the selection. Design and institutionalisation of nonprofit–business partnerships. Journal of Business Ethics, 85, 413– 429. Selsky, J. W., & Parker, B. (2005). Business–nonprofit partnerships to address social issues: challenges to theory and practice. Journal of Management, 31(6), 849–873. Sen, S., & Bhattacharya, C. B. (2001). Does doing good always lead to doing better? Consumer reactions to corporate social responsibility. Journal of Marketing Research, 38, 225–243 (May). Silvera, D. H., & Laufer, D. (2005). Recent developments in attribution research and their implications for consumer judgments and behavior. In F. R. Kardes, P. M. Herr, & J. Nantel (Eds.), Applying social cognition to consumer-focused strategy (pp. 53–77). Mahwah, NJ: Erlbaum. Simmons, C. J., & Becker-Olsen, K. L. (2006). Achieving marketing objectives through social sponsorships. Journal of Marketing, 70, 154–169 (October). The McKinsey Global Survey of Business Executives: Business and Society (2006). McKinsey Quarterly 2, pp. 33–39. Weiner, B. (1980). Human motivation. New York, NY: Holt, Rinehart, and Winston. Yoon, Y., Gürhan-Canli, Z., & Schwarz, N. (2006). The effect of corporate social responsibility (CSR) activities on companies with bad reputations. Journal of Consumer Psychology, 16(4), 377–390.