SESSION OVERVIEW

How consumers connect with brands remains at the forefront for consumer researchers and managers alike. A substantial body of research examines positive consumer-firm connections manifested in consumer-brand relationships (e.g., Fournier 1998), brand engage-ment in consumers’ self-concept (e.g., Sprott, Czellar and Spangen-berg 2009), attachment (e.g., Park, Eisingerich and Park 2013; Park et al. 2010), and love (e.g., Batra, Ahuvia and Bagozzi 2012; Yim, Tse and Chan 2008). On the contrary, work on negative consum-er-brand connections is comparatively scarce, possibly reflecting a

“positivity bias in brand relationship research” (Fournier and Alvarez 2013, p. 260). The focus on positive consumer-brand experiences and connections is somewhat surprising given that negative expe-riences, connections, and emotions tend to be more impactful than their positive counterparts (Baumeister et al. 2001).

Consequently, the central purpose of this session is to shed light on ‘negatives’ in consumer-brand connections. To this end we have collected four exciting papers that look at frictions in consumer-brand relationships from different perspectives: first, this session highlights that consumer-brand relationships are not as positive as portrayed in prior research once measurement of these consumer-brand con-nections accounts for the social context which they are embedded in (Khamitov, Thomson, and Johnson). Second, this session shows that individual differences predict whether, how, and why consumers may remain in negative and troublesome brand relationships (Dom-mer, Swaminathan, and Gürhan-Canli; Consiglio and van Osselaer).

Third, the final paper in the session shows that negative experiences and feelings towards one specific brand are likely to affect competing brands as well (Boegershausen, Klesse, Hoegg, and Dahl).

All four papers provide multifaceted insights into the negative spectrum of consumer-brand connections. Collectively, they address the following questions: (i) which types of consumers are most likely to remain in negative consumer-brand connections and (ii) whether and why do negative consumer-brand connections affect competing brands. In addition, this session reinforces the need to overcome the prevalent positivity bias in measuring and studying consumer-brand connections. A full appreciation of the potential frictions between consumers and brands can enrich our theories about consumer-brand relationships. This session is of interest to a diverse audience as it appeals to those interested in brands, consumer-brand relationships, and service failures.

Interpersonal Influences on Consumer-Brand Relationships: Exploring the Effect of Providing

Relationship Reminders on Brand Evaluations EXTENDED ABSTRACT

The current standard approaches of measuring consumer-brand relationships (CBRs) strength involves some variant of asking con-sumers to provide evaluations of researcher-identified brands (e.g.,

“To what extent do you feel personally connected to Apple iPod?”, Park et al. 2010, p. 6), self-selected brands with no category con-straint (e.g., “Think about a brand to which you have some degree of emotional attachment”, Thomson, MacInnis and Park 2005, p.

7), or self-selected brands within a particular category (e.g., “brand you love” in the consumer electronics category,Batra, Ahuvia and Bagozzi 2012, p. 7). Yet, consumers are typically involved in an ar-ray of relationships, many involving interpersonal dyads such as with friends and family. It has been argued that only by looking at the

“broader context of the consumer’s life” (Fournier 2009, p. 5) can a marketer accurately gauge the functioning and importance of brand relationships. But the standard measurement approaches outlined above do not do this: they assess CBR strength in a vacuum and provide no acknowledgement of this relational array. We argue this approach may systematically overstate the vitality and meaningful-ness of CBRs. To this end, we propose that contextualizing CBRs by providing interpersonal relationship reminders will lead consumers to rate CBRs as less agentic and strong. We argue this effect will be driven by the activation of relational schemas and will result in weakened consumer ratings of brand performance.

Study 1 establishes the phenomenon that providing interperson-al relationship reminders reduces consumer ratings of brand perfor-mance by lowering perceptions of agency (i.e., proximal mediator) and CBR strength (i.e., distal mediator). This study uses self-selected brands and manipulates relationship reminders both explicitly and implicitly. We find that providing relationship reminders reduces brand performance (attitudes, WOM, price premium, and willing-ness to defend) through our hypothesized mediators.

Study 2 replicates this effect with a different means of eliciting brands and a different measure of CBR strength. Instead of using self-selected brands, study 2 asked respondents about two pretested brands (Apple and Abercrombie & Fitch). The results of Study 2 gen-erally replicate study 1, offering additional support for the hypothesis that respondents who receive an interpersonal relationship reminder perceive brands to be have less agency, resulting in weaker CBRs and ratings of brand performance.

Study 3 demonstrated that the effect operates with interpersonal but not other types of relationship reminders (e.g., other brands, ce-lebrities), thus replicating the mediational pathway observed in stud-ies 1 and 2. In addition, Study 3 reinforced the role interpersonal re-lationship reminders play in influencing perceptions of CBR strength by revealing an important asymmetry: interpersonal relationship strength is not affected when immediately preceded by brand ques-tions but the reverse is not true. This result helps allay concerns that the prior results are confounded by a perceptual contrast effect.

Study 4 showed that the weakening effects of interpersonal relationship reminders on CBR strength are not driven by reduced materialism or enhanced perceived social support (alternative expla-nations). It also investigates if the negative effects of relationship reminders on CBR strength can be mitigated by prompting consum-ers to think about brands as people (Aggarwal and McGill 2012).

We expect that consumers who are made to think about brands in human-like terms will be more likely to integrate the brand into their relational schemas, thereby mitigating the effect of providing inter-personal relationship reminders on perceived brand agency and CBR strength. Replicating results of previous studies, consumers who re-ceived (vs. did not receive) an interpersonal relationship reminder reported weaker scores on brand agency and strength. However, among consumers who were situationally induced to think of brands as people and thus integrate brands into relational schemas, the re-minders had no little effect.

This work contributes to the CBR literature by answering the call of relationship theorists to investigate the role played by con-sumer’s social connections in influencing CBRs (Fournier 2009), helping to draw a finer line between brand and interpersonal rela-tionships (Swaminathan and Dommer 2012) and enhancing under-standing of the way interpersonal relationship reminders operate in a branding context (Cavanaugh 2014).

To the best of our knowledge, our research is the first to dem-onstrate how and when reminding consumers of their interpersonal relationships impacts CBRs. While previous research referencing consumers’ interpersonal relationships has suggested that market-ers should improve ratings of brand performance by appealing to consumers’ interpersonal relationships, such as through use of nos-talgic advertising (Muehling and Pascal 2011; Zhou et al. 2012), friendship-driven brand communities (Bagozzi and Dholakia 2006), brands as facilitators of interpersonal relationships (Goode, Khami-tov and Thomson 2015) and community appeals (Cutcher 2008), our results appear divergent and show that this is not necessarily the case. Our results offer insight into the role of agency in CBRs and extend a more nuanced understanding of the relationship metaphor into the brand domain (Fournier 1998), suggesting a more pervasive and complex psychological phenomenon whereby brand relation-ships can become affected by placing them in the broader context of consumers’ lives. Indeed, this is consistent with an increasing body of social psychological research on relational schemas that finds that activating interpersonal relational schemas exerts influence on sub-sequent perceptions of other social experiences (Andersen and Chen 2002; Baldwin 1992; Chartrand, Dalton and Fitzsimons 2007). Thus we contribute to this research by showing that the influences of rela-tional schemas activation extend to perceptions of consumer-brand interactions.

Who Blames but Forgives When Brands Err? Applying Attachment Theory to Explain Consumer Responses to

Brand Failures and Recovery Efforts EXTENDED ABSTRACT

In 2010, after Toyota announced a recall of cars due to a faulty gas pedal some customers swore off the brand entirely, while oth-ers went to dealoth-erships to express their support for the brand (Audi 2010). Although transgressions weaken the consumer-brand rela-tionship for some, others may forgive and forget.

Using a framework built on attachment theory (Bartholomew and Horowitz 1991), we examine how consumers’ attachment styles affect attributions following a transgression and subsequently sponses to transgressions and recovery efforts. Attachment styles re-fer to internal working models of interpersonal relationships based on two dimensions: anxiety and avoidance (Bartholomew and Horowitz 1991). Anxiety represents an individual’s fear of abandonment in an interpersonal relationship due to a negative view of self. Avoidance represents a fear of abandonment because of a lack of trust in the relationship partner. We believe the avoidance dimension will affect consumers’ stability attributions, and subsequently their brand atti-tudes, following a brand transgression. Because high avoidance indi-viduals believe relationship partners are untrustworthy, they should believe in the stability of brands’ transgressions and consequently lower brand attitudes.

A desire to punish the brand by spreading negative word-of-mouth (WOM) following a transgression is likely associated with both severity and controllability perceptions (Folkes 1984; Mc-Cullough, Fincham and Tsang 2003). Among individuals low in avoidance, those low in anxiety have a positive view of the self, which they can protect by attributing more control to the brand fol-lowing a transgression (Burger 1981). In contrast, low avoidance-high anxiety individuals tend to blame themselves following inter-personal rejections because doing so allows them to maintain their positive views of others (Bartholomew and Horowitz 1991). High avoidance-high anxiety individuals perceive greater severity in re-lational transgressions (Horan 2012). High avoidance-low anxiety individuals, however, avoid relying on relationships to define who they are (Collins 1996) and likely simply withdraw from the rela-tionship rather than punish the brand. In sum, we believe that both low avoidance-low anxiety and high avoidance-high anxiety individ-uals will demonstrate a greater willingness to spread negative WOM following a transgression. Furthermore, we expect controllability attributions to mediate the effect of anxiety on likelihood of spread-ing negative WOM among low avoidance individuals, but severity perceptions to mediate the effect of anxiety among high avoidance individuals.

In study 1, participants (n = 198) completed manipulations of anxiety and avoidance (Swaminathan, Stilley and Ahluwalia 2009) before reading information about a recall from GAP and responding to measures of brand attitude and likelihood of spreading negative WOM. Participants in the high avoidance conditions had signifi-cantly lower brand attitudes compared to those in the low avoidance conditions (ps < .06, one-tailed test). Those in the high avoidance-high anxiety and low avoidance-low anxiety conditions were more willing to spread negative WOM than those in the high avoidance-low anxiety and avoidance-low avoidance-high anxiety conditions (ps < .03, one-tailed test). Study 2 began with participants (n = 185) complet-ing Brennan et al.’s (1998) measures of anxiety and avoidance before reading information about a recall by Nike. We measured brand at-titude and likelihood of spreading negative WOM as well as severity, control, and stability (Folkes, Koletsky and Graham 1987).

Avoid-ance significantly predicted brand attitude (b = -.32; t(181) = 3.08, p < .01) and stability mediated this effect (bootstrapped confidence interval: -.1003, -.0026). The interaction of anxiety with avoidance significantly predicted likelihood of spreading negative WOM (b = .27, t(181) = 3.59, p < .001). The Johnson-Neyman point estimate (Spiller et al. 2013) indicated that when avoidance (anxiety) scores averaged at or above .7678 (.8020) from the mean or at or below -.8841 (-1.0841) from the mean, the effect of anxiety (avoidance) was significant. Severity mediated the effect of anxiety at high levels of avoidance (indirect effect at +1 SD of avoidance: .0595, .2658) and controllability mediated the effect of anxiety at low levels of avoidance (indirect effect at -1 SD of avoidance: -.2564, -.0243). In study 3 we considered two types of recovery efforts: apology and blame redirection. Apologies are associated with interpersonal bonds (Mikulincer and Florian 1998) and thus likely unappealing to highly avoidant individuals. Blame redirection is seen as a way of shift-ing responsibility for the transgression incident to external causes and subsequently reducing stability and controllability attributions.

While we believe a blame redirection will be successful in getting high avoidance-low anxiety individuals to restore their brand atti-tudes, we do not believe this will be the case for high avoidance-high anxiety individuals. Forgiveness requires a decrease in motivation to maintain estrangement from the offender (Mikulincer and Florian 1995), which these individuals are unlikely to have because they tend to avoid getting close to others (Bartholomew 1990). Partici-pants (n = 405) completed the measures of anxiety and avoidance and then read a Nike transgression scenario before reporting brand attitudes. One week later 176 participants returned to complete the second study and were exposed to a recovery effort and reported their brand attitudes again. Greater avoidance was again associated with lower brand attitudes before the recovery effort (b = -.21; t(172)

= 2.17, p < .04). The effect of recovery effort was significant among those with high avoidance-high anxiety (b = .64, t(168) = 3.39, p <

.001) and marginally significant for those with high avoidance-low anxiety (b = -.69, t(168) = 1.67, p < .10). High avoidance-high anxi-ety individuals lowered their brand attitudes even further when the brand redirected blame compared to when they offered an apology.

When high avoidance was coupled with low anxiety, a blame redi-rection increased brand attitudes.

This research examined how interpersonal attachment styles af-fect responses to brand transgressions. By illuminating the role of stability, severity, and controllability as processes governing these effects, we provide a more nuanced understanding of how inter-personal relationships of consumers impact their relationships with brands. Although recovery strategies have been shown to be influ-ential in guiding consumer–brand relationships following transgres-sion, our research sheds light on the moderating role of consumers’

attachment styles in enabling such a recovery.

The Devil You Know:

Service Failures, Self-Esteem and Behavioral Loyalty EXTENDED ABSTRACT

Why do some consumers switch to available alternatives fol-lowing service failures whilst others do not? In this research, we investigate one of the factors that might explain behavioral loyalty in face of service failures. Research suggests that individuals with low self-esteem who experience relational transgressions develop an avoidant attachment style, which impairs their interpersonal func-tioning and their willingness to take further interpersonal risks (Park and Maner 2009), and in particular to engage in other long-term relationships (Walker 2009). Drawing on this research, we propose

that low self-esteem (LSE) consumers who experience service fail-ures become unwilling to commit themselves to alternative brands, even when they have the opportunity to do so, thus – paradoxically – they remain trapped in their current brand relationship. High self-esteem (HSE) consumers, instead, are more likely to switch to other available service providers when they experience service failures, as compared to when they do not experience failures. We also predict that LSE consumers who experience service failures tend to avoid new commitments in general, thus favoring transactions relative to long-term contracts, even in consumption domains that are unrelated to the service failure. We tested these predictions in one survey and three experiments.

In study 1a, participants completed a measure of self-esteem (Rosenberg 1989; e.g., “I have a number of good qualities”), and reported the quality of their internet connection (“Considering your Internet usage over time, what percentage of the time does your In-ternet connection work perfectly?” 0% = InIn-ternet never works per-fectly, 100% = Internet always works perfectly.) These questions were interspersed among others in order to disguise the purpose of the study. Finally, participants indicated how likely they would be to switch to a competitor of their current Internet provider, if cancel-ing their current contract were free of charge and the new provider did all the paperwork. As the quality of their internet connection decreased, HSE consumers were more likely to switch to a differ-ent provider. However, frequency of failures did not have an effect on LSE consumers’ likelihood to switch. In study 1b, we replicated these findings in an experimental setting: HSE consumers who imag-ined to use an extremely faulty Internet connection were more likely to switch to an available provider as compared to their counterparts who imagined to use a perfectly functioning Internet connection;

LSE consumers did not express different switching intentions be-tween conditions.

Since we had hypothesized that the loyalty of LSE consum-ers in face of service failures is driven by their avoidance of new long-term relationships, in study 2 we manipulated the length of the contract offered by an alternative Internet service provider. When an alternative service provider offered a long-term contract (1 year) we replicated previous results: the two-way interaction between self-esteem and service quality was significant—HSE consumers were more likely to switch to this alternative service provider as the qual-ity of their internet connection worsened, but LSE consumers were not. Instead, when an alternative service provider offered a short-term contract (1 month, renewable) LSE consumers were as likely as HSE consumers to switch to this provider, as revealed by the absence of an interaction between self-esteem and service quality in this con-dition. In fact, following a service failure, LSE consumers seemed to prefer a short-term contract relative to a long-term contract, which suggests an increased fear of committing to alternative brands. In study 3, we demonstrate that the fear of new commitments induced by service failures extends to unrelated domains. LSE consumers who imagined to use an extremely faulty Internet connection ex-pressed a greater preference for buying a magazine at the newsstand relative to subscribing to this magazine, as compared to their coun-terparts who imagined to use a perfectly functioning Internet connec-tion and participants in a negative mood condiconnec-tion. The preferences of HSE consumers, instead, were not affected by service failures.

In sum, these studies suggest that following service failures, HSE consumers are more likely to switch to a competitor compared to their counterparts who do not experience failures, whilst LSE con-sumers are not (studies 1a, 1b, and 2). Consistent with the hypothesis that service failures (vs. no failures) make LSE consumers wary of relational risks, thus rendering them avoidant of commitment with

alternative brands, LSE consumers increase their preference for low-commitment offers from competitors and decrease their preference for high-commitment offers (study 2), even in unrelated consump-tion domains (study 3).

This research has significant practical import for marketing and policy-making. Consumers who experience service failures can more easily escape a negative brand relationship if low-commitment alter-natives are available; thus, marketers who wish to lure dissatisfied consumers from their competitors should consider diversifying their offerings and include short-term/low-commitment offers in their portfolio. Moreover, our results suggest that service failures might create generalized distrust in long-term relationships that extends to different consumption domains. In the long run, this could hurt marketing efforts to build trust and committed relationships. Finally, policy makers should take into account that behavioral loyalty is not determined only by switching costs or apathy. Lowering switching barriers to favor competition might not be enough to protect consum-ers’ rights, but motivating brands to right their wrongs – for example by broadening the scope of class actions – might be an important additional protection for LSE consumers.

The Category Spillover Effects of Brand Hatred EXTENDED ABSTRACT

In a recent campaign by the budget airline Spirit Airlines, more than 30,000 consumers expressed their hatred for Spirit and other American airline brands (Spirit 2014). The likely managerial motive behind Spirit’s campaign is that a brand benefits when consumers hate competing brands. Such category spillover effects of brand ha-tred are the focus of this paper. We examine how haha-tred for a brand

In a recent campaign by the budget airline Spirit Airlines, more than 30,000 consumers expressed their hatred for Spirit and other American airline brands (Spirit 2014). The likely managerial motive behind Spirit’s campaign is that a brand benefits when consumers hate competing brands. Such category spillover effects of brand ha-tred are the focus of this paper. We examine how haha-tred for a brand