The influence of contradictory CSR messages between the
corporate and product brand level on consumers’ attitude
towards the product brand
Name: Julia Popma
Student number: 10277404 Date of submission: Jan 29th 2016
Version: Final Master Thesis
Statement of originality
This document is written by Julia Popma who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Table of Contents
1. Introduction ... 5
2. Literature review ... 10
Part one: CSR ...10
2.1 Corporate social responsibility (CSR) defined ...10
2.2 Consumers’ responses to positive CSR ...12
2.3 Consumers’ responses to negative CSR (Irresponsibility) ...14
2.4 Consumers’ responses to inconsistent CSR ...18
2.4.1 Backfire effect ...20
2.4.2 Buffering effect ...23
Part two: Branding ...28
2.5 Spillover effects between corporate and product brands ...28
2.6 Hypotheses development ...34 2.7 Conceptual model ...42 3. Methodology ... 43 3.1 Research design ...43 3.2 Stimuli requirements ...44 3.3 Measures ...45 3.3.1 Dependent variable ... 45 3.3.2 Independent variables ... 46 3.3.3 Manipulation check ... 46
3.3.4 Control- and other variables ... 47
3.4 Pre-tests ...48 3.4.1 Pre-test one ... 49 3.4.2 Pre-test two ... 52 3.5 Main study ...56 4. Results ... 57 4.1 Reliability analysis ...58 4.2 Correlations ...58 4.3 Assumption of normality ...59 4.4 Hypotheses testing ...61 4.5 Additional analyses ...65
5. Discussion and conclusion ... 67
5.1 General discussion ...67
5.2 Theoretical implications ...71
5.3 Managerial implications ...72
5.4 Limitations and future research ...73
References ... 76
Appendices ... 82
Appendix A Method ...82
Appendix B1: Pre-test one ...83
Appendix B2: Pre-test two ...86
Appendix C: Final survey ...87
Appendix D: Sampling check ... 103
Appendix E: Reliability analyses ... 105
Appendix F: Normality ... 107
Abstract
Present research aimed to provide insights in how consumers respond to contradicting CSR messages across different brand levels. More specifically, it is examined how a positive (negative) product brand CSR message followed by a negative (positive) corporate brand CSR message influences consumers’ attitude towards the product brand. It was expected that the product brand message would be more diagnostic and would therefore predominate the corporate message. Furthermore, either a backfire or a buffering effect was expected. None of these two effects were significantly supported. However, the results indicate that a negative corporate brand CSR message significantly lowers consumers’ attitude towards the product brand. Furthermore, the positive product message mitigated the effect of the negative corporate CSR message. These findings reflect on implications for brand managers.
Keywords: Corporate Social Responsibility, contradictory, inconsistency, branding,
1. Introduction
Perhaps you remember the parody of the KitKat commercial released by Greenpeace in 2010. The well-known ‘have-a-break, have-a-KitKat’ moment is depicted, but instead of the actual chocolate fingers, the man in question is eating a bleeding Orang–Utan finger. Greenpeace was aiming to draw attention to Nestlé’s use of palm oil as a replacement for the expensive cacao butter, in its chocolate brand KitKat. This palm oil is extracted from deforested areas in Indonesia, which jeopardizes the lives of the Orang-Utans (The Guardian, 2010). The message of extracting palm oil is contradicting to Nestlé’s corporate social responsible message of being environmentally sustainable along their value chain (Nestlé, 2013). The phenomenon above highlights the occurrence of inconsistent corporate social responsibility (hereafter: CSR) messages.
Nowadays, consumers demand companies to behave in social responsible ways (Sen & Bhattacharya, 2001). Firms approve on the importance of behaving responsible and recognize the positive outcomes such as a better reputation (Cone, Feldman & DaSilva, 2003) and better consumer product performance perceptions (Chernev & Blair, 2015). The growing mass media coverage concerning transparency cuts both ways; on the one hand consumers will be more easily aware of the company’s social responsible activities, meantime there is also more light shed on the firm’s irresponsible actions (Wagner, Lutz & Weitz, 2009). The notice of irresponsible information can harm companies in severe ways and emphasize the fragility of a firm’s reputation. Likewise, Warren Buffet states: “it takes 20 years to build a reputation and five minutes to ruin it” and Henry Ford asserts that “a good
name is achieved by many actions but can be lost by just one” (Helm & Tolsdorf, 2013, p. 144). Most research up to date has only focused on either the notion of corporate responsibility or its counterpart, irresponsibility. Little is known about the impact of both socially responsible and socially irresponsible reporting and how consumers evaluate a brand characterized by the two simultaneously. Wagner et al. (2009) were the first to explore this phenomenon and introduced the term corporate hypocrisy.
However, no research yet has focused on such CSR inconsistencies in the context of different brand levels (i.e. the corporate level and the product brand level). Therefore this research aims to investigate the effect of contradictory (inconsistent) CSR messages between the corporate brand and the product brand. More specifically, this research will focus on how consumers cope with contradictory CSR messages: when the corporate CSR message is inconsistent with the product brand CSR message and vice versa.
For analyzing the effect of CSR inconsistency, consumers’ attitude will be measured for the current research instead of purchasing behavior, because only a positive relationship between CSR and purchase behavior exists when several conditions are met. Mostly, consumers are found to be unwilling to pay more even though they maintain a positive attitude (Sen & Bhattacharya, 2004). The current research is measuring the effect of inconsistent CSR on consumers’ attitude toward the product brand. There is opted for measuring attitude toward the product brand in lieu of the corporate brand for two reasons. First, with regards to social responsibility activities, the corporate level is aimed at multiple stakeholders such as employees, shareholders, and the community etc. (Grohmann & Bodur, 2014) while the brand level is more concerned with consumers as the primary stakeholders. Second,
reasoning from a more practical approach, consumers are rarely in direct contact with corporate brands. Meanwhile, consumers use product brands on a daily basis. In my view, consumers mostly identify with the product brand and are sometimes even ignorant of the existence of the corporate brand.
Research question: To what extent do contradictory CSR messages between the corporate brand (on the corporate level) and the product brand (on the product level) influence consumers’ attitude toward the product brand?
To provide an answer to this main research question, the following sub-questions will need to be examined:
What is CSR?
How do consumers respond to positive CSR?
How do consumers respond to negative CSR (irresponsibility)? How and why do consumers respond to inconsistent CSR?
How and why do perceptions spill over between different brand levels within a business portfolio?
What is the role of directionality of spill over effects between corporate and product brands?
Theoretical contribution
The concept of inconsistent CSR messages is something that has started being researched only recently. Wagner et al. (2009) are the first to explore the effect of corporate hypocrisy when consumers are exposed to inconsistent CSR information
(i.e. when the firms’ CSR policies are inconsistent with actual behavior (Wagner et al. (2009)). In addition, they examine different communication strategies that firms can use to diminish this effect. However, they have only focused on corporate hypocrisy and neglected the possible halo effect. Furthermore, they measure CSR inconsistency on the same brand level; either the corporate- or the product brand level, not a combination of the two. Finally, CSR inconsistencies are only tested while the messages were in the same CSR domain. More specifically, Wagner et al. (2009) tested inconsistent CSR messages that were both concerned with either environmental, employee or national economy issues. It is of interest whether the same findings hold for testing inconsistent messages that are unrelated. Hence, the current research aims to contribute to existing literature by testing CSR inconsistencies concerning unrelated CSR issues along different brand levels.
Other CSR studies were directed at the CSR halo effect in brand or product evaluations and suggest that CSR can be used as a buffering effect (Chernev & Blair, 2015; Klein & Dawar, 2004). However, these studies have only aimed at the halo effect and therefore neglected corporate hypocrisy. In addition these effects were only studied at the same brand level, which creates new avenues for the current research.
In order to study the inconsistent CSR messages at different brand levels, the branding literature should be scrutinized to understand the concept of spill over in brand portfolios. Previous studies have examined the spill over effects of the corporate brand to the product brand (Biehal & Sheinin, 2007; Brown & Dacin, 1997) and between product brands within one brand portfolio (Lei, Dawar & Lemmink, 2008). These studies found spill over effects of positive corporate CSR associations to product brands; however, the product brand was characterized by not implementing CSR activities, i.e. revealing neutral CSR information. Hence, this current research
will investigate the spill over effects when the CSR messages are contradicting i.e. when the corporate CSR message is positive (negative) and the product brand CSR message is negative (positive). This study therefore aims to provide new academic insights by combining two different literature streams, CSR and branding, to investigate CSR inconsistency at different brand level.
Managerial contribution
Today’s managers are facing complex business portfolios with challenging brand structures (Aaker & Joachimsthaler, 2000). Brand portfolios contain multiple brands and managers are pressured to leverage brand assets due to high costs associated with creating new brands (Aaker & Joachimsthaler, 2000). In the meanwhile consumers demand companies to address CSR issues (Sen & Bhattacharya, 2001). Managers believe that CSR activities enhance and strengthen the company’s reputation and therefore spend much effort in involvement in these CSR activities. However, the product brand managers cannot influence what happens outside the product brands’ scope; elsewhere in the business portfolio or on the corporate level. According to the branding literature, spill over effects between corporate- and product brands exist, which questions the effectiveness of a product brand’s CSR message. For product brand managers it is therefore of interest to know whether the CSR investments are beneficial if negative spill over from the corporate brand can instantly tackle the positive CSR associations on the product brand level. The same applies for managers at the corporate level. Is it useful for corporate brand managers to invest in CSR activities when the product brands do not live up to this standard? How do consumers respond to contradictory CSR messages between the corporate- and product level?
Current research aims to address these issues and to provide managers with interesting insights regarding CSR communications in brand portfolios.
The remainder of this paper is structured as the following. Section two, the literature review, will set out the existing CSR and branding literature and proposes the hypotheses and the conceptual framework. Section three covers the data and research method. Section four reflects on an in depth analysis of the results. Finally, section five elaborates on the discussion of the findings and provides an answer to the research question.
2. Literature review
This chapter provides an overview of the relevant literature for the current research. As mentioned in the introduction, two different streams of literature will be scrutinized for the sake of this research. To provide clarity and distinguish between the two different literatures, part one covers the CSR literature and part two comprises the branding literature and theories. In section 2.6 ‘hypotheses development’ the two literature streams are set up against each other in order to establish the hypotheses.
Part one: CSR
2.1 Corporate social responsibility (CSR) defined
Corporate Social Responsibility (CSR) is a term, which is omnipresent in today’s business world. Different definitions are found among existing literature, however most definitions overlap. According to van Marrewijk (2003) CSR is defined as “the
companies’ activities – voluntary by definition – demonstrating the inclusion of social and environmental concerns in business operations and in interactions with stakeholders” (p. 102). Mohr, Webb, and Harris (2001) term CSR as “a company’s commitment to maximizing or eliminating any harmful effects and maximizing its long-run beneficial impact on society” (p. 47). Brown and Dacin (1997) term CSR as “status and activities with respect to its perceived societal obligations” (p. 68). Dahlsrud (2008) analyzed 37 CSR definitions and found that all definitions are consistently referring to at least some of five dimensions: stakeholder, social, economic, voluntariness, and environmental. However, he stresses that the challenge is not about how CSR is defined, but how CSR is socially constructed in a specific context and that this guidance should be taken into account when developing business strategies.
Grohmann & Bodur (2014) were the first to study CSR at the product level and identified a distinction between social responsibility at different brand levels. They introduced the concept of Brand Social Responsibility (BSR): “consumers’ perceptions of the extent to which a brand reflects the human values related to social responsibility” (p. 3). In their view BSR is more concerned with brand anthropomorphization. In the end CSR and BSR are centralized around the same concept of social responsibility and accordingly for the purpose of this research the distinction between social responsibilities at different brand levels will be clarified by ‘corporate brand CSR message’ and ‘product brand CSR message’. Of interest for this research, the finding regarding predictions of consumers’ responses to product brands is most important. BSR is found to be different from CSR and is most influential in driving consumers’ responses to product brands (Grohmann & Bodur, 2014).
Taking into account the five dimensions identified by Dahlsrud (2008), this paper adopts the CSR definition from the Commission of the European Communities (2001) as “a concept whereby companies integrate social and environmental concerns in their business operations and in their interaction with stakeholders on a voluntary basis” (as cited in Dahlsrud, 2008, p. 7).
2.2 Consumers’ responses to positive CSR
The CSR strategies help both the social cause by providing financial rewards and at the same time benefits the company by enhancing their reputations (Cone, Feldman & DaSilva, 2003). Engaging in CSR results in favorable consumers’ evaluation of the company and purchase intent (Mohr & Webb, 2005). Several studies show that CSR results in positive attitude outcomes/evaluations. Other more relational behaviors as a result of CSR include loyalty, positive word of mouth and resilience to negative company information (Sen & Bhattacharya, 2004).
An important determinant in the relationship between CSR information and company evaluation is consumers’ perceived congruence between their own characters and that of the company (i.e. C-C congruence) (Sen & Bhattacharya, 2001). Consumer-Company identification is a form of social identification and is driven by individuals’ needs for self-definition and social identity that provokes them to connect as a member of a particular social entity; organizations (Bergami & Bagozzi, 2000). This social identification finds its roots in social identity theory.
In addition, the C-C congruence is in turn influenced by consumers’ support for the CSR domain. Hence, the effect of CSR on consumers’ purchase behavior is depending on consumers’ personal support for the CSR issue and their perception of congruence between the company’s character and their own (Sen & Bhattacharya,
2001). Put differently, the company’s character is positioned by its CSR (Brown & Dacin, 1997), this leads to consumer identification with the company, and leads to better company evaluations (Sen and Bhattacharya, 2001). This stresses the importance of selecting the CSR initiative that attracts the highest support of the company’s target segment.
Consumers’ responses to positive CSR are also linked to consumers’ attributions about the companies’ underlying motives behind CSR. Attribution theory is primarily conducted in social psychology and focuses on perceived causes of another person’s behavior (Kelley & Michela, 1980). It is conceptualized as “people interpret behavior in terms of its causes and that these interpretations play an important role in determining reactions to the behavior” (Kelley & Michela, 1980, p. 458). Whereas the traditional literature (Webb & Mohr, 1998) assumed that the motives consisted of a continuum between self-or other centered, more recent research identified four motives (Ellen, Web & Mohr, 2006): self-centered motives comprising strategic and egoistic, and other-centered motives consisting of values driven and stakeholder driven. Interestingly, most consumers attribute a combination of both; consumers attributing both values-driven and strategic motives were found to have the most positive response to the firm’s CSR efforts (Ellen, Web & Mohr, 2006). In line with different attributions, consumers view proactive companies that adopt a proactive stance concerning CSR initiatives more favorably versus companies that adopt CSR initiatives as a defensive and competitive strategy to keep up with the industry (Sen & Bhattacharya, 2004). The source informing consumers about CSR activities also plays a role in attributions. According to Yoon, Gürhan-Canli, and Schwarz (2006) consumers learning from CSR from the company source itself hurts
the perceived sincerity and suspicion, especially combined with high benefit salience. This occurrence is also referred to as the backfire effect of CSR (Yoon et al, 2006).
Concluding, several studies claim that engaging in CSR results in positive company outcomes/evaluations. Creyer (1997) emphasizes that the ethicality of a firm’s behavior (i.e. engaging in CSR) is important for consumers during purchase decisions. Consumers expect companies to behave in an ethical way and are willing to pay higher prices for that firm’s product. Interestingly, their findings indicate that consumers may still buy from companies that do not focus on ethical behavior, but they indicate to be willing to pay lower prices. The same positive effect was found by Mohr and Webb (2005). Their results indicate that CSR (in both the environmental and philanthropy domain) impacts company evaluation and purchase intent positively. However this contrasts the findings of Sen and Bhattacharya (2004) who stress the importance of the difference between attitude and actual purchase behavior in the event of price premiums. Yes, consumers will have a more favorable attitude towards companies that engage in CSR, but their research points out that only a small percentage of consumers is actually willing to pay price premiums. Hence, this denotes the reluctance to compromise on core attributes such as price and quality. CSR will only lead to increased purchase behavior when the CSR initiative is not at the expense of price and/or quality.
2.3 Consumers’ responses to negative CSR (Irresponsibility)
In the CSR debate, Corporate Social Irresponsibility (CSI) is the counterpart of CSR and is not being extensively researched yet. A first attempt was made by Armstrong in 1977. He defines a CSI act as “a decision to accept an alternative that is thought by the decision maker to be inferior to another alternative when the effects upon all
parties are considered” (p. 185). Incidents of irresponsible behaviors are rising due to firms’ globalization and these behaviors do not stay unnoticed by consumers (Wagner et al, 2009). Consumers nowadays have access to CSR activities from the firm due to growing mass media coverage, but in the meanwhile also CSI is revealed more easily (Wagner et al., 2009).
Focusing on irresponsible behavior within the company’s management, managers involved in CSI seem to behave in a certain irresponsible way because they feel that it is proper behavior for their role (Armstrong, 1977). Findings from his stakeholder role research point out that when board members are informed to represent all interest groups (stakeholders) in a decision, the choice to opt for an irresponsible decision is reduced.
More interesting are the expectations and responses to CSI from the consumers’ perspective. After all, consumers are the primary stakeholder group directly impacting the corporation (Sweetin, Knowles, Summey & McQueen, 2013). Past literature has studied consumers’ response to CSI in terms of emotions (Grappi et al., 2013), company evaluations (Yoon et al., 2006) and consumers’ attitude (Wagner et al., 2009). Consumers’ responses to CSI include behavioral reactions such as “negative word of mouth, complaining, boycotting companies and taking legal actions” (Grappi, Romani & Bagozzi, 2013, p. 1814).
Grappi et al., (2013) focus on two specific responses; negative word of mouth and protest behaviors. Firstly, negative word of mouth is the dissemination of dislike, dissatisfaction and detraction about CSI actions from companies. One of the objectives of negative word of mouth is to punish the company that is involved in CSI. Sweetin et al. (2013) found that consumers are more likely to punish and less likely to reward socially irresponsible corporate brands compared to socially
responsible, environmentally friendly and no social responsible brands. Secondly, protest behavior is the act against CSI with the aim to stop the company from continuing behaving harmfully (Grappi et al., 2013).
Corporate irresponsible behavior (CSI) is considered as moral transgressions (Grappi et al., 2013). Research has identified two categories of moral transgression; (1) ethical transgression, i.e. harm done to other people (e.g. workers and consumers), and (2) social transgressions, i.e. harm done to a community (Grappi et al., 2013). Ethical transgressions provoke negative emotional reactions including contempt, anger, and disgusts. The impact of these negative emotional reactions on efforts to punish the company or get them to change their behavior, is regulated by other-regarding virtues (i.e. justice, beneficence, equality, and communal cooperation) consumers hold (Grappi et al., 2013).
Being perceived as social irresponsible entails a serious risk; all consumers respond negatively to negative CSR information. On the contrary, only the consumers who are most supportive of the CSR issues react positively to positive CSR information as suggested by the moderating role of consumers’ support for the CSR domain as mentioned in section 2.2 (Sen & Bhattacharya, 2001). Furthermore, negative CSR (i.e. CSI) information is found to have a stronger effect than positive CSR information (Sen and Bhattacharya, 2001). This observation finds its roots in negativity bias. Negativity bias derives from the psychology literature and was first researched in the scope of individuals’ ethical behaviors (Skowronski & Carlston, 1987) and proved that negative information causes impressions ratings to differ from predictions of the equal-weight averaging model in the mathematical approach to impression formation (Skowronski & Carlston, 1989). It reveals that an individual’s ethical and unethical behavior influences one’s attitude towards this individual
asymmetrically, such that unethical behavior harms the attitude more than ethical behavior enhances it (Skowronski & Carlston, 1987). An explanation can be found in diagnosticity of information. From this perspective, negativity bias occurs because negative information is generally more diagnostic than positive (or moderate) information (Skowronski & Carlston, 1987). This is translated into behavioral actions and personality trait categories as such that “immoral actions are more diagnostic of negative traits than moral actions are of positive traits” (Folks & Kamins, 1999, p. 245). The underlying reason is the perception that “negative behaviors are thought to be characteristic primarily of actors who belong in negative categories; while positive behaviors may be characteristic of actors in both positive and negative categories” (Skowronski & Carlston, 1987, p. 690).
Folks and Kamins (1999) applied the negativity bias in the context of evaluations of organizations; organizations’ negative behaviors weight more in consumers’ evaluation of the organization compared to positive behaviors.
The distinction between self-disclosure and third-party disclosure of CSR information influences consumers’ response to negative CSR information (Fennis & Stroebe, 2014). This results in self-disclosure of negative information having a less negative effect than a third-party disclosure. Depending on the valence of the company’s reputation especially companies with poor reputation profit from the strategy of self-disclosure; for companies with an established positive reputation the type of disclosure (self versus third-party) does not have a significant differential effect (Fennis & Stroebe, 2014).
2.4 Consumers’ responses to inconsistent CSR
As previously mentioned, nowadays consumers place importance on CSR in their purchase decision, which pressures companies to behave in socially responsible ways (Mohr et al., 2001). More and more companies engage in CSR activities and communicate this to their stakeholders. However, at the same time due to increasing public information, social irresponsible behaviors are disclosed more easily. More often these social irresponsible behaviors are contrary to their stated standards of social responsibility, which results in inconsistent messages (Wagner et al., 2009). The issue of inconsistent CSR messages has not been extensively addressed in the literature yet. Most literature focuses on the impact of CSR information that is either positive or negative. Wagner et al. (2009) were the first to explore phenomenon of corporate hypocrisy caused by inconsistent CSR information. He tested contradictory (i.e. positive and negative) CSR messages in the same CSR domain about a company. More specifically, he tested inconsistent CSR messages that were both concerned with either environmental, employee or national economy issues. It is of interest whether the same findings hold for testing inconsistent messages that are unrelated (i.e. within different CSR domains). The current research will address this matter.
For the purpose of this research, inconsistent CSR messages are conceptualized as contradictory (i.e. positive (ethical) and negative (unethical)) information about a company’s CSR activities. Why and how inconsistent information with regard to a company’s social responsibility activities affects consumers remains unclear and will be the focus of this research.
Helm and Tolsdorf (2013) investigated how a firm’s previous reputation affects customer loyalty in a corporate crisis. In a crisis, consumers experience cognitive dissonance (Helm & Tolsdorf, 2013). Cognitive dissonance is the
psychological discomfort a person experiences when holding contradictory beliefs, ideas, values, or behaviors at the same time (Festinger, 1962). According to the dissonance theory, dissonance occurs when “two cognitions are relevant but the opposite of the one cognition follows from the other” (Helm & Tolsdorf, 2013, p. 146). In a crisis, the cognition ‘Company X has a good reputation’ is in dissonance with ‘Company X is involved in a crisis’ (Helm & Tolsdorf, 2013). People will strive to reduce the psychological discomfort and engage in one of the following dissonance reduction strategies: changing behavior, changing cognition, adding cognition, or ignoring information (Festinger, 1962). Depending on the strategy the person chooses, either a ‘backfire’ or a ‘buffering’ effect can be noticed. In the literature, researchers advocate for one of these two effects, and support their findings using different theories.
Vanhamme, Swaen, Berens and Janssen (2014) recognize a decrease in attitude toward the company when the consumer is exposed to CSI information. However, they highlight the influential effect of ex ante CSR communication. They researched how initial CSR communications unrelated to the CSI information can aggravate or provide some protection when allegations surface. Vanhamme et al. (2014) cite Wagner’s et al. (2009) measures, but use inconsistent messages from different CSR domains in contrast to Wagner et al. (2009) who used inconsistent messages from the same CSR domain. They state that the source revealing the initial CSR communication (before allegations of CSI) can influence, through activating persuasion knowledge, whether a backfire or buffering effects appears. When the initial CSR information is communicated through company controlled sources, a buffering effect (decrease drop in attitude) is triggered. More specifically, the initial CSR communication activates persuasion knowledge in the sense that the consumer
perceives the firm as making an appropriate persuasion attempt, which causes him not to entirely discount the earlier positive CSR information in their final evaluation. Conversely, when the initial CSR communication is transmitted through third-party sources, a backfire effect (increased drop in attitude) arises. In this case, the persuasion knowledge becomes activated after the CSI information, and consumers believe that the firm is making an inappropriate persuasion attempt through the third-source which causes the initial positive CSR communication to backlash (Vanhamme et al., 2014).
2.4.1 Backfire effect
When information about a company’s social responsibility activities is inconsistent, a backfire effect might occur. When customers become suspicious about the CSR activity, the backfire effect implies that the company is evaluated more negatively than without any CSR (Yoon et al., 2006). Concentrating on environmental issues in information inconsistency, green washing is defined as “intentionally misleading or deceiving consumers with false claims about a firm’s environmental practices and impact” (Nyilasy, Gangadharbatla & Paladino, 2014, p. 693). More specifically, firms claim to protect the environment but their actions and behavior fail to comply. Public confusion is one of the major consequences of green washing (Dahl, 2010). Consumer confusion is defined as “consumer failure to develop a correct interpretation of various facets of a product/service during the information processing procedure”
(Turnbull, Leek & Ying, 2000, p. 145). Consumer confusion arises from; inter alia, ambiguous stimuli/information (Turnbull et al., 2000).
et al., 2009). Whether a person will be perceived as hypocrite occurs when there is a “distance between assertions and performance” (Shklar, 1984, p. 62). Wagner et al. (2009) strengthen the concept of hypocrisy to organizations, which is supported by social psychological research, and define corporate hypocrisy as “the belief that a firm claims to be something that it is not” (p. 79). Perceived corporate hypocrisy is found to negatively affect consumers’ attitudes toward a firm, both directly and indirectly as a mediator through CSR beliefs (Wagner et al., 2009). However, there are three communication strategies that mitigate the negative impact of the CSR inconsistencies on corporate hypocrisy, CSR beliefs and finally attitude toward the firm; (1) the use of an proactive/reactive strategy, (2) the abstractness of the firm’s CSR statements, and (3) the use of an inoculation strategy (Wagner et al., 2009). A proactive communication encompasses that a company’s CSR statements go ahead of the inconsistent/conflicting) behavior (Murray & Vogel, 1997). A reactive strategy involves the CSR statements to pursue the conflicting behavior (Murray & Vogel, 1997). The findings indicate that a proactive communication strategy results in higher levels of perceived corporate hypocrisy than a reactive strategy. By making the CSR statement more abstract, the risk of perceived corporate hypocrisy is reduced in the case of a proactive communication strategy. Contrary, a more abstract CSR statement with a reactive strategy is found to increase perceived corporate hypocrisy (Wagner et al., 2009). The use of an inoculation strategy will reduce perceived corporate hypocrisy for both types of communication strategies (reactive and proactive). The inoculation theory originates from the opinion literature and states that when an initial communication strategy is two-sided (i.e. arguments that support the speaker’s conclusion and arguments for the opposite that counter the speaker’s conclusion) a person will right away already have considered both positive and negative arguments
while still maintaining a positive position. When the same person is subsequently exposed to contrary arguments, he will refute the opposing arguments and maintain the positive position (Lumsdaine & Janis, 1953). Basically, through such a two-sided strategy the audience is ‘inoculated’ against subsequent contra information (Lumsdaine & Janis, 1953). Wagner et al. (2009) applied a two-sided message strategy to the context of CSR communication. A positive CSR statement causes primary positive CSR beliefs about a company’s social responsible activities. Subsequently, the company publishes a socially related irresponsible action statement, accompanied by a rejection. Releasing such a statement decreases the impact of subsequent negative irresponsible behavior. In summary, an inoculation communication strategy will enable consumers to discount inconsistent CSR information and therefore reduce perceived corporate hypocrisy (Wagner et al., 2009). As mentioned above, Helm and Tolsdorf (2013) studied the relation between a crisis and customers’ loyalty. In accordance with dissonance theory, Helm and Tolsdorf (2013) found that the when a crisis occurs, customers’ loyalty is decreased. The valence of a firm’s reputation plays a role; a pre-existing favorable reputation presents a certain liability because consumers have higher expectations of firms with favorable reputations versus unfavorable reputations. After all, firms with pre-existing unfavorable reputations have less to lose in the event of a crisis, and therefore suffer less reduction in customer loyalty and firm profitability (Helm & Tolsdorf, 2013). During a crisis situation, as suggested by expectancy violation, loyalty was decreased more strongly for firms with a pre-existing favorable reputation because the expectations from such firms are violated. These findings are supported by the expectancy disconfirmation theory developed by Oliver (1980). The theory describes that a consumer has pre-purchase expectations about an outcome when he purchases a
good or a service. When the consumer has purchased the product/service, the actual outcome is compared against the expected outcome. When both outcomes match expectancy confirmation occurs. On the contrary, when the outcomes mismatch expectancy disconfirmation occurs. More specifically, positive (negative) disconfirmation occurs when the actual outcome is superior (inferior) to the expected outcome (Pizam & Milman, 1993). According to research, confirmation and positive disconfirmation cause satisfaction and negative disconfirmation causes dissatisfaction. In the event of CSR inconsistency, negative disconfirmation might arise resulting in dissatisfaction, which might negatively affect consumers’ attitude.
Another theory underlying the backfire effect is the negativity effect. As mentioned above, from the consumers’ perspective negative information is more diagnostic than positive information (Skowronski & Carlston, 1987). In the formation of evaluative judgments, negative information compared to equally positive information has a greater weight, reinforcing the backfire effect.
2.4.2 Buffering effect
Other researchers suggest the contrary of the backfire effect and expect a buffering effect instead. Negative publicity does not always inherently lead to negative corporate associations (Einwiller, Fedorikhin, Johnson & Kamins, 2006). Consumer-company identification moderates the effect of negative corporate social messages on corporate associations. More specifically, consumers experiencing a strong identification with the company obtain less negative corporate associations than consumers having weak identification when being exposed to negative publicity. This corresponds to the findings of Ahluwalia, Burnkrant, and Unnava (2000), who reveal that how the consumer processes negative information is moderated by how
committed he feels toward the brand. Yet, this buffering effect only holds up to moderate levels of negative publicity; the strong identification does not mitigate the effects of extremely negative publicity (Einwiller et al., 2006).
In order to reduce cognitive dissonance a person experiences when he is exposed to inconsistent information, he can also choose to disregard information (as mentioned earlier). The negative crisis-related information is ignored by the person who has a positive attitude towards the firm (Helm & Tolsdorf, 2013). This is in accordance with the findings from Coombs and Holladay (2001) who found that in times of a crisis, a pre-existing favorable reputation operates as a shield for negative information and prevents the reputation to decline. In times of a crisis, a favorable pre-crisis reputation is believed to protect the organization’s reputation (Coombs & Holladay, 2006). A favorable pre-crisis reputation is built through an organization-stakeholder relationship derived from positive interactions and allows organization-stakeholders to discount or ignore negative information about the organization. The halo effect is an explanatory frame to understand how a company’s reputation affects the crisis situation. A halo effect is the “bias due to a measure that spills over to another measure” (Klein & Dawar, 2004, p. 204). The halo effect claims that “the previous reputation affects the acceptance and interpretation of new information” (Coombs & Holladay, 2001, p. 335). The halo effect can be explained by two dynamics; as a shield or as benefit of the doubt, although the former is more supported (Coombs & Holladay, 2006). In the case of the halo as benefit of the doubt, the crisis responsibility that the stakeholder attributes to the firm is influenced by its favorable reputation. The stakeholder will assign less crisis responsibility to the firm (compared to unknown or unfavorable reputation), which results in less reputational damage (Coombs & Holladay, 2006). Hence, under this dynamic, stakeholders do take the
negative information into consideration for firm’s evaluation, but just simply ‘give it the benefit of doubt’. Klein and Dawar (2004) expanded the model with ‘blame’ that is negatively related to brand evaluation. They found that prior CSR reputation affects consumers’ attributions of blame when there is negative information. For firms that have a favorable prior CSR reputation, the trigger event of the negative information will be judged as more external, less stable, and less controllable. Consumers will therefore attribute less blame to the company, which will diminish the damaging effect of negative CSR messages on brand evaluation. However, this halo effect will only hold for consumers for whom CSR is an important decision criteria (Klein & Dawar, 2004). The implications for such a CSR halo include that even if positive CSR associations do not increase immediate profitability, the prior favorable CSR associations may assist in reducing the risk of damage to brand evaluations in times of crises. As cited from Klein and Dawar (2004): “CSR is like an insurance policy that is there when you need it” (p. 215).
Contrariwise, sometimes stakeholders do not even consider the negative information and just simply ignore the negative information. This is the explanation for halo as a shield. This explanation is part of a larger psychological phenomenon of expectancy conformation (Coombs & Holladay, 2006). Consumers are reluctant to adjust their initial opinion/expectation of a firm or event, even when they are exposed to new information conflicting their initial expectation (Traut‐Mattausch, Schulz‐ Hardt, Greitemeyer & Frey, 2004). Consumers’ prior expectations can be interpreted as a key moderator for the effect of a product-harm crisis on brand equity (Dawar & Pillutla, 2000). The consumers’ prior favorable expectations about the firm can reduce the devastating effect of a product-harm crisis on brand equity. Consumers hold strong prior confirmatory expectations and when these expectations are inconsistent
with the evidence, a dilemma arises. Consumers solve this dilemma by discounting the disconfirmatory evidence rather that updating expectations (Olsen & Dover, 1979).
In addition, CSR is found to have a bigger effect than just enhancing companies’ reputation. There is a positive relationship between a company’s CSR actions and consumers’ attitudes toward both the company and the product (Brown & Dacin, 1997). Engaging in social responsibility can also alter consumer perceptions of product performance (Chernev & Blair, 2015). More specifically, the products of companies that are committing social goodwill by including CSR, are perceived as performing better than products originating from companies not engaging in CSR. This effect even holds when the CSR messages are unrelated to the company’s core business, which is the case in charitable giving (Chernev & Blair, 2015). The above represents “the benevolent Halo of social goodwill” (Chernev & Blair, 2015, p. 1414). Displaying social goodwill behavior can provide rise of a halo effect that will influence evaluations of unrelated properties in a manner that is consistent with the evaluation of the prior property (in this case the CSR behavior).
The benevolent halo of social goodwill is also addressed by Klein and Dawar (2004). They highlight that the CSR halo is most obvious when the “affected measures are unrelated to the source of the halo, because when the affected measures are related, it is not possible to separate the halo effect from a nomological effect” (p. 204). Interestingly CSR might affect unrelated product evaluations, but not in times of a product-harm crisis. Eisingerich, Rubera, Seifert and Bhardwaj (2010) believe that prior CSR beliefs prevent firms only from negative information about CSR practices (which is in this case obviously related to the source of the halo), but not to information related to the firms’ core service offerings. Remarkably, this is in contrast
to the statement of Chernev and Blair (2015) mentioned earlier. Therefore, Eisingerich et al. (2010) conclude that CSR might offer less of an insurance policy than assumed in previous research. They focus on the importance of the balance in focusing on customer and/or service quality orientation and CSR. Customer orientation refers to the company’s caring behavior towards the customer and the extent to which the firm is attentive to the customer’s needs (Eisingerich et al., 2010). Service quality orientation refers to the extent the company is viewed as focusing on outcome-related features of the service (Eisingerich et al., 2010). Additionally, the composition of the firm’s consumer base needs to be taken into consideration; only for a consumer’s base consisting of mainly novices, firms will benefit from greater levels of consumer resistance to negative information because these types of consumers are more likely to use extrinsic cues in service provider evaluations (Chernev & Blair, 2015; Eisingerich et al., 2010). Consumers with higher levels of expertise will rely more on functional aspects of a product and will therefore strengthen the positive effects of service quality orientation on consumers’ resistance to negative information (Eisingerich et al., 2010).
In conclusion, according to prior CSR literature, engaging in CSR brings along positive consumers’ attitudes. However, when consumers are exposed to information concerning the company’s irresponsible behavior, consumers’ attitude will drop. The literature regarding contradicting information suggests that sometimes a pre-existing favorable reputation can protect the organization due to a halo effect from the prior positive reputation. However, sometimes this positive reputation can actually backfire as a consequence of expectancy violation. The next part covers spill over effects
originating from the branding literature in order to understand how CSR associations can spill over to different brand levels.
Part two: Branding
Social Responsibility and Corporate Ability are two types of corporate associations. Corporate associations can arise at the corporate level and can be transferred to the product level (Biehal & Sheinin, 2007). This phenomenon is referred to as a spillover effect.
2.5 Spillover effects between corporate and product brands
Transfer refers to “the influence of corporate messages on product beliefs and attitudes” (Biehal & Sheinin, 2007, p. 12). The degree of transfer from a corporate message to the company’s portfolio is related to the firm’s branding strategy. The branding strategy is linked to brand architecture, which is “an organizing structure of the brand portfolio that specifies brand roles and the nature of relationships between brands” (Aaker & Joachimsthaler, 2000, p. 8). The brand relationship spectrum is constructed across two extremes: a house of brands vs. a branded house. A house of brands is characterized by a set of independent stand-alone brands. An example would be Unilever that owns different independent brands such as Dove, Rexona and Lipton. On the opposite side, a branded-house (also called a monolithic brand) involves a master brand that provides an umbrella under which many of the company’s products brands operate, such as the British brand Virgin (Aaker & Joachimsthaler, 2000). This suggests that a company using a branded-house strategy can make use of corporate
retrieval cues (such as the corporate name or logo) in their product messages (Biehal & Sheinin, 2007).
Furthermore, when employing a branded-house strategy, associations from the corporate message will have overlap with associations from the product’s message and thus contain beliefs that are more diagnostic for evaluating the product (Biehal & Sheinin, 2007). Indeed, corporate brand associations can enhance credibility (Aaker, 2004). Other benefits of the branded house include maximizing clarity, synergy and leverage (Aaker & Joachimsthaler, 2000). However, the downside of a branded-house strategy involves the risk of visible negatives; when leveraging the corporate brand, a negative message of the sub-brand will affect the entire corporate brand (Aaker, 2004).
Corporate associations are defined as “all the information about a company that a person holds” (Brown & Dacin, 1997, p. 69). According to Brown and Dacin (1997) corporate associations are divided into corporate ability (CA) and corporate social responsibility (CSR). The former refers to the associations “related to the company’s expertise in producing and delivering its outputs” while the latter covers “the organization’s status and activities with respect to its perceived societal obligations” (Brown & Dacin, 1997, p. 68). Both CSR - and CA associations are found to transfer and influence consumers’ reactions to new products introduced by the company. CA associations influence the overall company evaluation and specific product attribute perceptions more strongly compared to CSR associations through a dual process mechanism, since CA messages are positioned on product-relevant attributes (Brown & Dacin, 1997; Biehal & Sheinin, 2007). To explain this matter in more detail, both CA and CSR associations affect corporate evaluation, which in turn impacts product evaluation due to ‘transfer’. However, in addition CA associations
also affect product sophistication (specific product attribute), and therefore influence product evaluation. In this way, CA associations influence new product evaluation through its effect on both product sophistication and corporate evaluation (dual), while CSR associations influence product evaluation only through its effect on corporate evaluation (Biehal & Sheinin). Other literature concerning consumers’ evaluations of product brand extensions suggest the moderating effect of perceived fit between the corporate company’s image and the new product (Madrigal, 2000). Fit is defined as “the consumer perception of how logically and sensibly the new product is related to a parent brand. Perceived fit between a parent brand and the brand extension thus acts as a heuristic for consumers in their brand extension evaluations” (Dwivedi & Merrilees, 2013, p. 451). When consumers evaluate a new product, they are found to put more weight on corporate associations (both CA and CSR) where there is high vs. low perceived fit (Madrigal, 2000).
Biehal and Sheinin (2007) extend the findings of Brown and Dacin (1997) by measuring the impact of different corporate information/associations on product evaluations across an entire portfolio instead of just a single product. They support their findings with the accessibility-diagnosticity framework from Feldman & Lynch, 1988). The degree to which a message or information is transferred to an evaluation depends on its (1) accessibility, which refers to “the ease with which it can be retrieved from memory”, (2) diagnositicity, or the level to which this information is relevant for the evaluation and (3) accessibility and diagnosticity “relative to other information” (Biehal & Sheinin, 2007, p. 13). Hence, in the case of the influence of corporate messages on product evaluations: in order for the corporate message to be transferred to product evaluation it must be accessible when consumers think of the product information, diagnostic in evaluating the product, and more accessible and
diagnostic relative to other product information sources (Biehal & Sheinin, 2007). Their findings indicate that a corporate message (vs. a product message) was more diagnostic for evaluations of other products in the company’s portfolio, meaning that a corporate message is transferred more onto other products in the portfolio compared to a product message. However, a product message (vs. a corporate message) is found to be more diagnostic for evaluations of that same product (Biehal & Sheinin, 2007). In the formation of overall evaluations, the greater weight is expected from the more diagnostic piece of information (Feldman & Lynch, 1988). This is an important finding for the current research. Finally, when distinguishing between CA and CSR messages, CA messages are more diagnostic than CSR messages in influencing product evaluations across the company portfolio. This correlates with the research by Brown and Dacin (1997) because CA messages are positioned on product-related attributes.
Interestingly, when scrutinizing the different types of corporate associations, the transfer of CSR and CA associations differ as such that a branded house strategy strengthens CA association transfer, while an endorsed corporate branding strategy (i.e. the brand is still independent but is also endorsed by an organizational brand) increases the transfer of CSR associations (Grohmann & Bodur, 2014; Berens et al., 2005). The dominance of the corporate brand enhances the accessibility of CA associations, while reducing accessibility of CSR associations. Therefore, when the corporate brand is dominant in product communications, such as in a monolithic branding strategy, it is perceived as a driver and induces people to focus on the brand’s CA. CSR will not have any effect on product evaluations. Making the corporate brand less dominant changes the corporate brand’s role from driver to an endorser, and allows for the transfer of CSR associations due to higher accessibility.
Noticeably, in this case either fit must be high or involvement must be low, because CSR only has limited diagnostic value (Berens et al., 2005). Visibility, the branding strategy and the corporate associations transfer moderate the transfer of CSR associations to the brand level (Grohmann & Bodur, 2014). When the corporate brand is not visible at all, as is the case in a house of brands strategy, consumers may not be aware of common ownership of brands and consequently social responsibility perceptions are likely to ascend at the product level, resulting in differences in social responsibility perceptions across the corporate and the product brand. In this case, perceived CSR and BSR will differentiate substantially from each other (Grohmann & Bodur, 2014).
When multiple information sources are available to consumers to make an evaluation, they only use enough information cues to satisfy a “diagnositicity threshold” – “that is a minimum level of certainty about the evaluation” (Berens et al. 2005, p. 36). Fiske and Talyor (1991) argue that consumers use just enough information to be able to make an evaluation and do not consider all accessible and diagnostic information sources because “information search is costly” (as cited in Biehal & Sheinin, 2007, p. 13). Consumers are therefore also called to be ‘cognitive misers’.
The transfer of corporate messages is also referred to as a ‘spillover effect’ and is connected to the associative network theory where spreading activation activates related notes and strengthens their associations in the network through pre-existing links (Anderson, 1983). In the context of brand portfolios, a brand node is activated by external information (e.g. a product harm crisis) and this activation spreads or spills over to other related brand nodes through pre-existing linkages. The strength of associations between brands is conceptualized as ‘brand relatedness’ and
encompasses the “connection between the association and the brand node” (Till, Baack & Waterman, 2011, p. 93). The associative network theory indicates that the linkages between two concepts can point in both directions. Spillover from the parent to the product brand is referred to as ‘forward spillover effect’ while spillover from the product to the parent brand is defined as ‘reciprocal spillover effect’ (Balachander & Ghose, 2003). This implies that the spillover effect is a function of not only the strength- but also the directionality of brand associations (Lei, Dawar & Lemmink, 2008). The strength and directionality in turn depend on the number and the salience of associations linked to each brand. The implication suggested by Lei, Dawar and Lemmink (2008) shows that accurate predictions of spillover cannot be made without indicating the direction of the linkage between two brands; especially in the case of asymmetric spillover.
The reciprocal spillover effect (feedback effect) is also studied by Ahluwalia and Gürhan-Canli (2000). While most previous research was mainly focused on high accessibility, they highlighted the importance of the moderating effect of accessibility of extension information in the relationship of valence of extension information and extension category on brand evaluations. With high information accessibility, negative information led to dilution and positive information led to enhancement, irrespective the extension category (far vs. close) (Ahluwalia & Gürhan-Canli, 2000). In contrast, when there is low information accessibility, the effects where observed in response to more diagnostic information; positive information leads to enhancement for a far extension while negative information leads to dilution for a close extension. To clarify, to extend in far product categories is perceived as more difficult and therefore firms are not expected to excel in far extensions. Being successful in a difficult ability-related task is acknowledged to be more diagnostic of a person’s
ability than failure (Skowronski & Carlston, 1987). Indeed, any firm, irrespective of the brand’s quality, could fail in manufacturing products far from its core capabilities. Meanwhile, all brands, irrespective of the brand’s quality, are expected to succeed in extensions that are close to their core capabilities. Therefore, positive information about a close extension is not very diagnostic and not signaling the family’s brand quality. In contrast, negative information about a close extension is indicative of a low quality family brand and subsequently found to be rated as more diagnostic (Ahluwalia & Gürhan-Canli, 2000).
In conclusion, according to the branding literature corporate CSR associations can spill over to the product brand and be used for product brand evaluations. However, hitherto this is only researched when the product brand CSR perceptions were neutral. In addition, product CSR perceptions better predict product brand outcomes including attitude. Besides, product message perceptions are more diagnostic for forming product beliefs and evaluations of that same product. The current research will therefore test for spill over effects when the CSR information on the brands levels is inconsistent.
2.6 Hypotheses development
The following section combines the two different literature streams in order to develop the hypotheses. The question at the heart of the current research is: ‘To what extent do inconsistent CSR messages between the corporate brand (on the corporate level) and product brand (on the product level) influence consumers’ attitude towards the product brand?’
According to the CSR literature, involving in CSR might evoke positive outcomes such as enhancing the company’s reputation, customers’ loyalty, positive word of mouth and resilience to negative information (Sen & Bhattacharya, 2004). In contrast, negative CSR messages result in negative consumers’ responses such as negative word of mouth, complaining, and boycotting. Negative information harms consumers’ attitude more than positive information enhances it, due to negativity effect (Skowronski & Carlston, 1989). But how are consumers’ responses affected when there is contradicting CSR information about a company? Consumers will then experience cognitive dissonance, which they will subsequently try to reduce. According to the CSR literature, depending on the reducing strategy they use and the context, either a buffering or a backfire effect might occur.
A backfire effect occurs when expectancy violation is experienced. When the CSR activity backfires, the company is even more negatively evaluated than without the CSR activity (Yoon et al., 2006). A firm with a positive pre-existing favorable reputation is expected to behave according to its reputation (Helm & Tolsdorf, 2013). Negative disconfirmation causes dissatisfaction that is expected to negatively affect attitude. The company is perceived as hypocrite and believed to claim to be something that it is not (Wagner et al., 2009). Furthermore, as mentioned above, negative information is more diagnostic than positive information, due to negativity bias. The negative information message will therefore weight more heavily compared to the positive information message in the event of inconsistent information.
Conversely, other researchers suggest that negative information does not inherently lead to negative attitudes and expect a buffering effect to occur (Einwiller et al., 2006). The person experiencing cognitive dissonance is expected to ignore the negative information to reduce the discomfort. According to Coombs and Holladay
(2001) a pre-existing favorable reputation operates as a shield for negative information or as a benefit of the doubt in times of a crisis, indicating a halo effect. The favorable pre-crisis reputation is believed to protect the organization and allows stakeholders to discount or ignore negative information. When the firm obtains the benefit of the doubt, the stakeholder assigns less crisis responsibility to the firm (Coombs & Holladay, 2001). Interestingly this is in contradiction to what is reasoned for the backfire effect where the favorable pre-existing firm reputation is seen as a burden since firms are expected to live up to their favorable reputation status. Importantly, the halo benefit of the doubt effect only holds for consumers for whom CSR is an important decision criteria (Klein & Dawar, 2004). Expectancy confirmation is one of the theories underlying this effect. Consumers are reluctant to adjust their initial opinion, even when they are exposed to new information conflicting their initial expectation (Traut-Mattausch et al., 2004).
According to the branding literature, the associative network theory suggests that linkages between two concepts can point in both directions. Spill over is therefore a function of not only strength but also directionality of brand associations (Lei et al., 2008). Spill over from the corporate to the product brand is referred to as forward spill over effect and is relevant for the current research. Corporate associations, both CA and CSR, can spill over and influence consumers’ responses to new products introduced by the company (Brown & Dacin, 1997). However, CA associations are found to be more diagnostic and influence overall company evaluation more strongly compared to CSR associations. More specifically, corporate CSR messages are less diagnostic for forming product beliefs across the portfolio than are CA messages. Expanding on the different brand levels, CSR perceptions on the product level are found to be better predictors for product brand outcomes (product brand attitude,
product brand purchase intent and brand equity) while corporate CSR perceptions are found to be more influential for corporate outcomes such as employment intent etc. (Grohmann & Bodur, 2014). Consequently, the product CSR message will be more important in consumers’ evaluation of a product brand. Accordingly, the branding literature arguments that the product message perceptions are more diagnostic for forming product beliefs and evaluations of that same product (Biehal & Sheinin, 2007). In the formation of overall evaluations, the greater weight is expected from the more diagnostic piece of information (Feldman & Lynch, 1988). To conclude, there are spill over effects within a business portfolio from the corporate brand to the product brand, but meanwhile the product message will be more influential in consumers’ responses to the product brand when there are contradictory CSR messages at different brand levels. In addition, in a house of brands branding strategy, CSR perceptions are likely to ascend at the product level and will consequently differ from corporate CSR perceptions. For the purpose of this study a corporate brand employing a house of brands strategy will be used, to initially separate CSR perceptions between the corporate and the product brand. Finally, consumers are cognitive misers and do not use all available accessible and diagnostic information. Taking all these arguments together it is therefore expected that whether a buffering or backfire effect occurs in an inconsistent CSR situation depends on the valence of the product message; after all, the product message is most influential, compared to a corporate message, in establishing consumers’ responses toward the product brand.
In addition to testing the hypotheses concerning inconsistent CSR messages on different brand levels, it must be checked whether solely the product messages have an effect on consumers’ attitude towards the product brand. It must therefore be investigated whether the attitude of respondents exposed to a negative (positive)
product CSR message without a corporate message is lower (higher) than when no CSR information at all is shown to respondents (the ultimate baseline). These two conditions (PosProd/NoCorp and NegProd/NoCorp) function as sort of baselines in order to test for a backfire of buffering effect. These effect of the product CSR messages will be tested as additional analyses, see section 4.4.
Firstly, the possibility that the product message predominates in the inconsistent situations, because the corporate message in general would not have an effect on consumers’ attitude towards the product brand, must be ruled out. After scrutinizing the branding literature, it is certainly predicted that the corporate message spills over to the product level and affects product brand’s attitude. Therefore the following is expected:
H1a: Negative corporate messages negatively affect product attitude as such that
when the consumer is exposed to a negative corporate CSR message, the consumer’s
attitude towards the product brand is expected to be more negative compared to when
the consumer is not exposed to CSR messages at all.
However, when the negative corporate CSR message is combined with a contradicting positive product message the product message is expected to prevail due to higher diagnosticity and consequently the negative corporate CSR information is expected to be valued less. The negative product brand attitude evoked when the consumer was exposed to only a negative corporate CSR message (H1a) will be, at least partly, canceled out.
