Consumer Behavior 6

Consumer Response to Unethical Corporate Behavior: A Re-Examination and Extension of the Moral Decoupling Model

Kristina Haberstroh1 • Ulrich R. Orth1,3,4 • Stefan Hoffmann2 • Berit Brunk1

Received: 13 November 2014 / Accepted: 16 April 2015 / Published online: 29 April 2015 ! Springer Science+Business Media Dordrecht 2015

Abstract This research replicates Bhattacharjee et al. (J Consum Res 39(4):1167–1184, 2013) moral decoupling

model and extends the original along the dimensions of

theory, method, and context. Adopting a branding perspec- tive and focusing on the corporate domain rather than the

public figures investigated by Bhattacharjee and colleagues,

this research examines the proposition that consumers dis- sociate judgments of morality from judgments of perfor-

mance to justify purchasing from companies deemed to act

immorally. The original study is further extended by ap- plying the model in a different cultural context and em-

ploying a more realistic stimulus to establish external and

ecological validity. The results of the replication generally support the original findings, in particular, under conditions

of higher product involvement, a theoretical extension.

Keywords Firm ethics ! Germany ! Moral decoupling ! Product involvement ! Replication ! Retail

Introduction

A stream of studies investigates unethical firm behavior and

the consequences developing from it (Vitell 2003). The

majority of studies converges on the finding that information about improper firm behavior leads to undesirable conse-

quences with consumers including damages to the image

consumers hold of the company (e.g., Einwiller et al. 2006), its products (Biehal and Sheinin 2007), the brands (Schmalz

and Orth 2012), their attitudes (Ahluwalia et al. 2000), in-

tentions and behavior (Hoffmann 2013), and—ultimately— reductions in sales, profits, and stock price (Einwiller et al.

2006). There are, however, cases when apparent unethical

behavior of a firm has little or no effect on consumer re- sponse. For example, consumers keep buying electronics

despite documentaries of deplorable work conditions at the

retail site (Schmalz and Orth 2012) or continue to buy ap- parel manufactured through child labor (Joergens 2006).

The present study builds on propositions, which recently have been put forward to explain why there is no straight-

forward link between consumers judging a firm to act

unethically and their behavioral response. As such, this re- search is notably different from studies examining condi-

tions for when the impact of unethical perceptions is more or

less pronounced, hereby testing possible moderators such as consumer-brand attachment (Schmalz and Orth 2012),

availability of substitutes (Sen et al. 2001), or one’s confi-

dence that a boycott will be effective (Hoffmann 2011). A prominent theoretical foundation of the extant studies

examining why (and why not) consumer judgment of im-

morality influences subsequent behavior is the motivated

Ulrich R. Orth—Adjunct Professor and Research Fellow.

& Ulrich R. Orth [email protected]

Kristina Haberstroh [email protected]

Stefan Hoffmann [email protected]

Berit Brunk [email protected]

1 A&F Marketing – Consumer Psychology, Christian- Albrechts-Universität Kiel, Wilhelm-Seelig-Platz 6/7, 24098 Kiel, Germany

2 Institute of Business Administration, Marketing, Christian- Albrechts-Universität Kiel, Olshausenstraße 40, 24098 Kiel, Germany

3 School of Business, University of Adelaide, Adelaide, Australia

4 University of South Australia, Adelaide, Australia

123

J Bus Ethics (2017) 140:161–173

DOI 10.1007/s10551-015-2661-x

reasoning theory (Kunda 1990). A number of studies have

shown how the need to avoid cognitive dissonance (Shu et al. 2011), protect the self-concept (Mazar et al. 2008), or

to uphold one’s moral standards (Aquino and Reed 2002)

motivate consumers to reconstrue moral transgressions as less severe (for a review see Hoffmann 2013). To arrive at

the desired conclusion (i.e., that the transgression is less

relevant) consumers engage in a number of strategies, such as redefining harmful conduct, minimizing the firm’s role

in causing harm, downplaying the harm caused by the unethical act, or blaming the victim (Bandura 2010). All

moral rationalization processes, however, share one sig-

nificant shortcoming. Moral judgments are closely tied to the self (Aquino and Reed 2002) and people strive to

maintain a positive view of themselves (Baumeister 2010),

regard themselves as morally upstanding (Bandura 1999), and generally avoid behavior that could threaten their

moral standards (Bandura et al. 1996). It is conceivable that

rationalizing with oneself to keep supporting a firm that exhibits unethical behavior involves the risk to compro-

mise one’s moral standards. Compromising one’s moral

standard, however, contradicts the need to maintain a positive view of the self, hereby causing tension or cog-

nitive dissonance (Thøgersen 2004).

In this regard Bhattacharjee et al. (2013) moral decoupling model is distinct from, and perhaps superior to, moral ra-

tionalization because moral decoupling is easier to justify and

feels less wrong than moral rationalization. Moral decoupling describes the psychological separation process by which

consumers selectively dissociate judgments of morality from

judgments of performance. In a series of experiments, Bhat- tacharjee and colleagues examined the manner in which

people maintain their support for public figures (sportsmen

and politicians) who have behaved immorally. Notably, the originalstudyconfirmed the processofmoral decoupling only

outside the realm of firm and corporate behavior.

The goals of this study are twofold. First, it replicates Bhattacharjee et al. (2013) model of moral decoupling.

Second, this study extends the original work in three im-

portant ways along the dimensions of theory, method, and context (Berthon et al. 2002): Regarding context, the pre-

sent study adopts a societal perspective on branding (e.g.,

Bhattacharya and Korschun 2008; Laczniak and Murphy 2012) and examines consumer response to the unethical

behavior of a firm rather than public figures. In addition,

the cultural setting changes from the U.S. to Germany. Methodologically, the current study uses a more vivid

stimulus (video) rather than vignettes. Theoretically, this

study addresses the question of when moral decoupling is more likely to occur by testing product involvement and

credibility of information as possible boundary conditions.

By replicating and extending Bhattacharjee and coauthors’ work, this research will help determine the generalizability

of their decoupling model to show that it applies to the

wider and for the discipline particularly relevant field of business ethics.

Why Replicate?

While replications have been labeled ‘‘one of the building

blocks of the structure of knowledge’’ (Goldenberg and Muller 2014), there is some debate on the value of straight

replications (for a review see Easley et al. 2000). Some argue that a study should be replicated only after it has

been referenced extensively, viewing the number of cita-

tions as a proxy for the study’s importance (e.g., Boles et al. 2000). Others forcefully advocate that no researcher,

practitioner, or teacher should use results before the phe-

nomenon, concept, model, or theory has been replicated successfully (Evanschitzky et al. 2007).

Replications are considered particularly useful when they

are conducted early in the history of a particular stream of research (Monroe 1992), and when they additionally extend

the original research along specific dimensions (i.e., theory,

method, and context) according to Berthon et al. (2002) framework. Especially when conducted by researchers other

than the original ones, early replications can help reduce

concerns that the interaction of the researcher and study may be a reasonable explanation for the original results (Lucas

2003). The replication’s value further increases with key

characteristics being retained (Hubbard and Lindsay 2013), original authors and authors of the replication being

separated by physical distance, a lack of overlap in personal

attributes, and a lack of personal contact between ex- perimenters (Rosenthal 1990). According to these criteria,

the present study should be of high value, because it was

conducted at a different time (a couple of years after the original experiments), in a different place (in Germany rather

than the United States), and the researchers in both studies

had no communication. Implicit in this discussion is the as- sumption that the original study is worth replicating.

Clarifying exactly this point is one of the goals of the present

research, to increase the confidence in future applications relying on a robust model.

The Original Bhattacharjee et al. (2013) Study

Bhattacharjee et al. (2013) developed and tested a model that examines a process by which people selectively dis-

sociate judgments of morality from judgments of perfor-

mance. Their approach sought to confirm that such a process exists, that it is distinct from moral rationalization

and more predictive of people’s support for public figures.

162 K. Haberstroh et al.

123

The study examined the process of how the relevance of a

public figure’s transgression influences consumer support for the public figure. On the one hand, consumers could

maintain their support through moral rationalization, a

processing route implying a less strict judgment of im- morality. On the other hand, consumers could make use of

moral decoupling, which consists of a psychological

separation of moral judgment and judgment of perfor- mance. The original study examined politicians (a gover-

nor) and sportsmen (a hockey player and a baseball player) as perpetrators of ethical transgressions. Five laboratory

studies and one field study varied in conceptual focus,

sample composition (mostly students), public figures and their immoral acts, and the performance indicators used to

provide evidence for the moral decoupling model.

Findings of the original studies 1–3 represent empirical evidence for the process of moral decoupling. The par-

ticipants selectively dissociated judgments of morality

from judgments of performance. Scenarios in the ex- periments portrayed a hockey player who has led his team

to an Olympic gold medal and physically abuses his wife

and a company CEO who supports racist and sexist hiring policies and has guided his firm to become a global leader

in innovative and stylish products. Study 4 employed a 2

(occupation: baseball player vs. governor) 9 2 (transgres- sion: taking steroids vs. evading tax) to show that moral

decoupling is prevalent with students who freely chose it to

generate support for public figures. Study 6 replicated this finding in a field setting by content-analyzing comments

made by readers of online news in response to Tiger

Wood’s extramarital affair. Study 4 replicated the ex- perimental setup of study 3, and study 5 used a similar 2

(moral reasoning argument: moral decoupling vs. moral

rationalization) 9 2 (transgression: relevant vs. not rele- vant) design to provide evidence that young students find

generating support via moral decoupling easier to justify

and indicate it feels less wrong than moral rationalization. In summary, Bhattacharjee et al. (2013) findings supported

their main predictions. The question remains, if the moral

decoupling model has value beyond the realm of public figures and whether consumers employ this psychological

process to justify purchases of products made by firms

deemed to act unethically.

Conceptual Framework and Hypotheses

Seeking to replicate and extend the Bhattacharjee et al.

(2013) model, the present study re-tests the core proposi- tion of the original study, and adds two new hypotheses to

test moral decoupling in a business context: First, the ex-

pectation that moral decoupling affects purchase intention and that this influence is mediated by judgment of

performance but not by judgment of immorality, and,

second, that moral decoupling will be more pronounced as product involvement increases. Berthon et al.’s (2002)

framework serves to evaluate similarities and differences to

the original research along the dimensions of context, theory, and method.

Context

The chief research question in this replication and exten- sion was whether the moral decoupling found with public

figures holds in a business setting. If the Bhattacharjee

et al. (2013) model is a generalizable depiction of the processes consumers engage when facing ethical trans-

gressions, then the relationships between constructs should

be relatively consistent across different domains. The business environment, however, may be markedly

different from consumer evaluation of public figures.

Prominent among those differences is a change in the so- cial level of the transgression as the immoral actor is no

longer an individual (sportsperson, politician), but an ab-

stract collective entity (firm). The transgression is no longer in the personal domain, but is integral to corporate

performance. This may impact consumer judgment as

judgments of morally relevant behaviors differ for indi- viduals versus groups (Critcher and Dunning 2013). When

people consider individual behavior, they give emphasis to

individual-level factors, accounting for what a person’s moral conscience would lead one to do. When considering

groups, in contrast, people place greater weight on group-

level factors, taking into account social norms and pres- sures. These differences imply that assumptions about the

regulatory environment or competitive norms may play a

larger role in the corporate context than in the case of the individual public figures examined in the original study.

However, there is also a major similarity linking the

original with the present context, making them more comparable. Research in advertising (e.g., Keel and

Nataraajan 2012) and marketing (e.g., Thomson 2006)

advocates viewing both public figures and firms as brands. This perspective is consistent with recent advances in

stakeholder theory aimed at integrating the management

and ethics literatures (e.g., Bhattacharya and Korschun 2008; Laczniak and Murphy 2012). Unifying those studies

is the view that both celebrities (public figures) and firms

should look beyond customers as the sole target of mar- keting activities and the firm as the primary intended

beneficiary. Instead, they should adopt a societal perspec-

tive. Slowly replacing previous thought in marketing which tended to be firm centric, there appears to be growing

consensus that a firm’s constituents are actually embedded

in interconnected networks of relationships through which the actions of a firm reverberate (Bhattacharya and

Consumer Response to Unethical Corporate Behavior: A Re-Examination and Extension… 163

123

Korschun 2008). This realization is similarly reflected in

advertising where marketing celebrities is seen as funda- mentally a social force with moral responsibilities and the

potential to influence the welfare of society (Miller and

Laczniak 2011). Just like a corporate brand embodies the value system of that company (Fan 2005) human brands

have meaning beyond their name, as their image and, most

important, their personal values reflect what behaviors they endorse, condone, and condemn (Miller and Laczniak

2011). Beyond the change in ethical transgressors, there is also

a change in cultural context that sets the replication aside

from the original. Different than the original setting in the U.S., the present study was conducted in Germany. German

and U.S. consumers differ with respect to ethical beliefs

and attitudes (Auger et al. 2007). Moreover, German con- sumers strongly focus on business performance (Wit-

kowski and Wolfinbarger 2002), and have a particular

desire to obtain products at lower prices as evidenced by higher levels of price mavenism (McCarty et al. 2007).

These characteristics imply that judgment of a firm’s per-

formance (i.e., the ability to offer decent quality at a low price) should be especially relevant to study participants.

Furthermore, the original study 1A (which is the one most

closely aligned with our research goals) relied on a sample of participants recruited through the university of Pennsylva-

nia, whereas the present study collected data from con-

sumers. People’s ethical beliefs and judgments change with age (Rawwas and Singhapakdi 1998) and the moral stan-

dards of students in the U.S. and a more general consumer

sample in Germany may not fully overlap (Vitell 2003). Ultimately, potential differences in attitudes towards

business ethics and firm performance may trace back to the

culturally specific moral factors that predispose each indi- vidual to accept some behaviors more readily than others.

Moral Foundations Theory (Haidt and Joseph 2004) posits

that five moral foundations are universally present: harm/care, fairness/reciprocity, ingroup/loyalty, authority/

respect, and purity/sanctity. Depending on their cultural

background, different societies construct different moral- ities by resting their moral virtues, claims, and institutions

to varying degrees on each moral foundation (Haidt and

Kesebir 2010). Even subcultures within the same society elaborate and emphasize different moral foundations to

different degrees (Koleva et al. 2012). Accordingly, there

may be divergent results, caused by differences especially in the dimensions of harm/care (disapproval of individuals

who cause pain and suffering and approval of those who

prevent or alleviate pain) and fairness/reciprocity (sensi- tivity to issues of equality and justice and disapproval of

people who violate these principles).

A few studies have directly contrasted ethical consid- erations of consumers in Germany and the U.S. Witkowski

and Reddy (2010) found that Germans reported fewer

ethical consumption activities than Americans. Data of the world value survey indicate that—across several waves of

data collection—the share of consumers who boycott

companies is consistently higher in the U.S. than it is in Germany (World Values Survey Association 2009). Indi-

vidualism-collectivism, one of Hofstede’s cultural dimen-

sions, has the highest predictive power in explaining national differences in boycott prevalence (Hoffmann

2014). According to the GLOBE-project, individualism is significantly higher in the U.S. than it is in Germany

(Gelfand et al. 2004). Together those studies imply that the

level of ethical consumption is lower in Germany than in the U.S.; replicating Bhattacharjee et al.’s model in a

German context should therefore be a rather conservative

test. Replicating the original hypotheses we expect the following:

H1 Moral decoupling is a significant (positive) predictor of consumer behavioral intention in response to an ethical transgression.

H2 The positive effect of moral decoupling on purchase intention is mediated by (a) judgment of performance but

(b) not by judgment of immorality.

Theory

Adding a theoretical extension to the original work, our study examines the role of involvement as a possible

moderator of the moral decoupling effect. Responding to

unethical missteps of a firm (rather than a public figure) may assume considerably greater importance with con-

sumers, because firm performance likely is more directly

related to, or even the cause of, the ethical transgression due to cost-cutting efforts. Bhattacharjee et al. (2013,

p. 1173) themselves acknowledge that ‘‘people should find

it more difficult to dissociate moral judgments from judg- ments of performance when a transgression is directly

relevant to the domain of the performance in question.’’ As

a consequence, the workings of moral decoupling could be fundamentally different depending on how motivated

consumers are to evaluate a firm’s ability to offer quality

products at a reasonable price. Building on the original definition of involvement as a

motivational state or alignment with consumer values

(Mittal and Lee 1989), recent research has used constructs like motivation and identity in place of involvement (e.g.,

Klein and Ahluwalia 2005). Consistent with this perspec-

tive, we view involvement as a proxy for the role of mo- tivation, an interpretation which is not only implicit in

Bhattacharjee et al. (2013) original formulation of moral

decoupling as a theory of motivated reasoning, but is also additionally compatible with negativity research in

164 K. Haberstroh et al.

123

marketing (Schmalz and Orth 2012; Till and Shimp 1998)

and social psychology (Baumeister et al. 2001). According to the negativity bias, negative information has a stronger

impact on one’s processing than neutral or positive aspects.

Yet, the negativity effect is not universal and more recent research suggests that the effect depends on how motivated

people are. Accordingly, the strength of a person’s moti-

vation/involvement should determine whether negative information is weighted more heavily. Consistent with this

view, we assume that if we segment consumers by their involvement with a given product category, only those

motivated to process information more thoroughly (via the

central route) should exhibit a negativity effect. In contrast, those who are less motivated to process information (using

the peripheral route) should not dwell on the negative in-

formation any more than on the positives. Thus, the like- lihood of a negativity effect should increase when negative

information about the firm coincides with high involve-

ment, hereby increasing consumers’ need to decouple. We expect the following:

H3 The effect of moral decoupling will be more pro- nounced as product category involvement increases.

Method

A methodological variation in terms of stimulus employed

sets our work aside from the original. Bhattacharjee et al. (2013, except for study 6) employed an experimental

procedure in which participants first read short scenarios

that described accomplishments and missteps of public figures, before they proceeded to submit their judgments

and intention to support these public figures. While five of

the six studies comprising the original Bhattacharjee et al. (2013) work employed scenarios to present participants

with information about public figures and their ethical

(mis)behavior, this replication employed a more vivid sti- mulus consisting of a 5-min video. The documentary-style

naturalistic video showed employees of a fictitious food retailer, promoting and selling products to customers.

Employees openly discussed the company’s questionable

behavior with respect to the environment (e.g., lavish use of pesticides), employees (e.g., inadequate protective

clothing, child labor), and the biosphere (e.g., massed

livestock farming). They even stressed that the competitive prices for the products are in line with the firm’s guiding

principle ‘Accepting reality—maximizing profitability.’ In

addition, the video showed customers who inquired in depth about specific aspects of the firm’s unethical be-

havior. While some of them turned away after having re-

ceived information about the firm’s unethical practices, others continued to purchase; a few were even hoarding.

Illustrating such divergent consumer responses was in-

tended to highlight differences in descriptive norms and role models. Moral decoupling should occur, for example,

when an approach goal is salient (here: consuming the

appealing products offered at low prices), while, at the same time, an avoidance goal is salient (here: not sup-

porting unethical corporate behavior by purchase). Other

forms of motivated reasoning imply that one goal is more relevant for the consumer than the other (e.g., consuming

the unethical product is more appealing that acting in ac- cordance to moral standards). If both goals are equally

important for the consumer or if the circumstances hinder

that one goal dominates the other, moral decoupling should occur. Had the video shown only a single descriptive norm

respondents might have adjusted to this particular norm.

However, in a moral decoupling situation, there are two descriptive norms present and those norms are illustrated

and personified in the video via different consumer seg-

ments. Hence, the video offers role models for both pos- sible responses (buying and not buying). This setting has

high ecological validity as in real life there are usually also

models of both ways of behaving. This dilemma introduces uncertainty of what is the right option and moral decou-

pling offers a solution to adjust to both descriptive norms

simultaneously. In summary, the video highlighted that customers may find themselves in a quandary between

honoring performance and deprecating immoral behavior.

Table 1 summarizes similarities and differences between the original and this replication.

Empirical Study

Method

Following Brandt et al.’s (2014) replication recipe, we

present a close replication of Bhattacharjee et al.’s (2013)

research. Approved by the appropriate ethics committee, the study has been performed in accordance with the

ethical standards laid down in the 1964 Declaration of

Helsinki. All participants gave their informed consent prior to their inclusion in the study.

The original study 1A (which is the one most closely

aligned with our research goals) relied on a sample of ninety- eight participants (22 years of mean age, 61 % females)

recruited through the university of Pennsylvania. The char-

acteristics of this student-heavy sample are very similar to those of subsequent studies (with the exception of Study 6

which relied on expert judges and consumers). The present

study collected data from two hundred and sixty-one con- sumers (28 years of mean age, 75.3 % females), recruited

through online postings in several social networks.

Consumer Response to Unethical Corporate Behavior: A Re-Examination and Extension… 165

123

As described in more detail above, we used a vivid

stimulus. A naturalistic video showed employees of

‘Agraprofit,’ a fictitious food retailer, promoting and sell- ing agricultural products (e.g., bananas, eggs) and food

products (e.g., chocolate, sausages) from around the world

to visitors of a Berlin market (Agraprofit 2013).

Dependent measures

This replication adopted the original scales of BB&R’s

study, assessing transgression relevance (Ingram et al. 2005; M = 4.48, SD = .78, a = .88), moral decoupling (M = 2.16, SD = .95, a1 = .93 vs. a2 = .63), judgment of immorality (M = 4.39, SD = 1.10), and judgment of per- formance (M = 1.90, SD = 1.06). All scales were slightly

adapted to fit the firm context (see Appendix Table 4 for

scale items and summary statistics). Because two of the original items (one assessing judgment of performance and

the other judgment of immorality) were not transferable to

the corporate context, the present study used a single item for each construct to capture consumer ratings. This ap-

proach seems defensible given the concrete singular

meaning of both constructs (Bergkvist and Rossiter 2007) as indicated in the BB&R study. Due to the business

context and different to the original study, purchase in-

tention (Putrevu and Lord 1994) was the main dependent

variable (M = 1.64, SD = .86, a = .81). Product in- volvement (Mittal and Lee 1989, M = 3.37, SD = .86,

a = .73) was included as a possible moderator of the moral decoupling model. To statistically control for the perceived credibility of information, we assessed participants’ re-

sponse to the question ‘‘In your opinion, how credible is the information provided by Agraprofit?’’ on a scale from

1 = not at all credible to 5 = fully credible (M = 3.36,

SD = 1.30).

Analyses and Results

Consistent with our expectations and supporting Hy-

pothesis 1, regression analysis demonstrated that moral

decoupling increases consumers’ intention to purchase the company’s products (b = .45, t = 8.05, p = .001). We

predicted that judgment of performance (but not judgment

of immorality) would mediate this effect. Using a SPSS macro for testing multiple mediation (Preacher and Hayes

2008), we found that moral decoupling had a significant

positive effect on judgment of performance (b = .48, t = 7.54, p = .001), whereas its effect on judgment of

immorality was negative (b = -.23, t = -3.16,

p = .002). People who dissociated judgments of perfor- mance from judgments of immorality evaluated the im-

proper behavior as less immoral. As expected, we found

Table 1 Comparison of studies

Bhattacharjee et al. 2013 This study

Ethical transgressor Public figure (individual) Firm (collective entity)

Sample U.S. respondents (students in studies 1–5) Consumers in Germany

Stimulus Scenarios (Vignettes) Video

Variables Moral decoupling (IV) Moral decoupling (IV)

Judgment of immorality Judgment of immorality

Judgment of performance Judgment of performance

Support (DV) Purchase intention (DV)

Involvement (Moderator)

Transgression relevance: Relevance to job performance (study 3)

Transgression relevance: Perceived magnitude of harm

Hypotheses

H1: Moral decoupling increases H H intention H H H2a: Judgment of performance mediates X H H2b: Judgment of immorality does not mediate – H H3: Involvement moderates H H

IV independent variable, DV dependent variable

‘H’ Hypothesis was supported ‘—’ Relationship was not tested in that study

1 Cronbach’s alpha of the original Bhattacharjee et al. (2013), Study 1A. 2 Cronbach’s alpha of the present replication.

166 K. Haberstroh et al.

123

support for the positive relationship between judgment of

performance and purchase intention (b = .16, t = 3.53, p = .001). In contrast, judgment of immorality did not

influence the intention to purchase (b = -.06, t = -1.47,

p = .143). This outcome is generally consistent with the original results and indicates that consumers justify their

behavioral intention (to buy unethically produced food) by

separating judgment of performance from judgment of morality. Furthermore, as we hypothesized (H2a), there was a significant indirect effect of moral decoupling on purchase intention through judgment of performance

(b = .07, LL = .04, UL = .13). Sobel’s test confirmed the

Bootstrap results (z = 3.17, p = .002). The fact that the confidence interval of judgment of immorality contained

zero means that there was no indirect effect through

judgment of immorality (b = .01, LL = .00, UL = .05) (z = 1.28, p = .200) (see Fig. 1), supporting Hypothesis

2b.

To more directly test whether the model examined by Bhattacharjee et al. (2013) applies to the business context,

specifically, whether the effect of transgression relevance is

channeled through moral decoupling, we tested a two-step mediation using the macro supplied by Preacher and Hayes

(2008). Figure 2 shows the results for both the original

model and the replication. In line with the original findings, the total indirect effect of transgression relevance on pur-

chase intention through moral decoupling and performance

judgment was significant with a 95 % CI of [-.17, -.03] and a SE of 0.04 (original: 95 % CI of [-1.25, -0.44] and

SE of 0.20). Together with the previously reported non-

significant effect of moral decoupling through immorality judgment, this result further substantiates the model’s

generalizability.

To further test the role of involvement as a possible moderator, we followed the approach outlined by Irwin and

McClelland (2003). Perceived credibility was added as a

control variable. The first step of a moderated regression analysis revealed a significant direct effect of moral de-

coupling on consumers’ intention to purchase the compa-

ny’s products (b = .45, t = 8.05, p = .001). When moral decoupling, credibility of information, and product in-

volvement were entered jointly as independent variables in

the second step, the significant influence of moral decou- pling remained, but no other variables had a direct effect.

In the last step, all step 2 predictor variables were entered

plus their interaction terms. As can be seen on Table 2, the direct effect of moral decoupling remained, and only the

Moral decoupling x Product involvement interaction term

had an additional significant effect (b = .12, t = 2.09, p = .038), supporting Hypothesis 3.

To identify ranges of involvement scores where the

simple effect of moral decoupling on purchase intention was significant and where it was not, we conducted

floodlight analysis (Spiller et al. 2013). Unlike spotlight

analysis which tests effects of the independent on the de- pendent variable at a few select values of the moderator

(which potentially have little meaning), the floodlight

analysis introduced by Johnson and Neyman (1936) iden- tifies regions over the entire range of the moderator vari-

able where the simple effect of the focal predictor is

significant and regions where it is not. The border between these regions is known as the Johnson-Neyman point

(Hayes and Matthes 2009; Spiller et al. 2013). Conducting floodlight analysis over a range of the moderator variable

from 1 = very low involvement to 5 = very high in-

volvement yields a Johnson-Neyman point for p \ .05 (t = 1.97) at a value of 2.01. This value indicates that

when involvement was 2.01 or higher, the coefficient for

moral decoupling was significantly positive. In contrast, when involvement was below the critical value, the coef-

ficient became non-significant (see Table 3).

Figure 3 plots this interaction graphically to enable easier interpretation of the effect. As can be seen, moral

decoupling had a strong positive influence on purchase

intention at higher levels of involvement (i.e., at a value of 5), whereas the coefficients indicate a less pronounced

effect of moral decoupling on purchase intention as in-

volvement decreases (i.e., to a value of 1).

Discussion

Summary and Implications

This study replicated and extended the Bhattacharjee et al.

(2013) model on moral decoupling. Not an exact replica-

tion, the study examined the same constructs and rela- tionships but with a broader sample definition, a more

naturalistic test situation, and—most importantly—with a

firm as moral transgressor rather than public figures. The results show strong support for the decoupling

theory put forward by Bhattacharjee et al. (2013). Moral

decoupling is prevalent with consumers in Germany and a significant predictor of their intention to purchase from a

firm that exhibits unethical behavior. This process occurs

as consumers dissociate judgments of performance from judgments of morality. The moral decoupling phenomenon

occurs even when the transgression (lowering production

cost through unethical practices) is highly relevant to per- formance (offering quality products at low prices) and

thus—at least theoretically—harder to decouple. While the

positive effect of moral decoupling on judgments of per- formance perfectly aligns with the one reported in the

original study, the negative effect on judgments of morality

stands in contrast to the original where it was reported as non-significant. Additional findings from this replication

Consumer Response to Unethical Corporate Behavior: A Re-Examination and Extension… 167

123

include the following: The positive effect of moral de-

coupling on purchase intention becomes more pronounced as involvement increases, and this mechanism functions

regardless of the credibility of the information received.

Given that public figures and firms can both be viewed as brands, the findings have implications for brand man-

agement in general. Not restricted to for-profit companies,

branding can also apply to persons including politicians,

sportsmen, celebrities or artists (Bal et al. 2009; Miller and Laczniak 2011; Thomson 2006), places (Hankinson 2001;

Olins 2002), political parties and religions (Shepherd

2004), non-profit-organizations and universities (Ewing and Napoli 2005), among others. The findings obtained

here indicate that the model of moral decoupling may

Moral Decoupling Purchase Intention

Judgment of Performance

Judgment of Immorality

b=.48*** b=.16***

b= -.06nsb= -.23**

Indirect effect, b(performance)=.07, 95% CI [.04,.13]

Indirect effect, b(immorality)=.01, 95% CI [.00,.05]

Direct effect, b=.19***

Fig. 1 Model of moral decoupling as a predictor of purchase intention, mediated by judgment of performance, not judgment of immorality *p \ .05; **p \ .01; ***p \ .001

Original Results (BB&R 2013, Figure 3)

Replication Results

Consumer Support

Transgression Relevance

Moral Decoupling b2 = 0.46***

Performance Judgmenta1 = -1.16***

a3 = 0.53***

c = -1.39*** (c´= -0.57*)

*22.0=1b*16.0-=2a

Purchase Intention

Transgression Relevance

Moral Decoupling b2 = 0.15***

Performance Judgmenta1 = -0.30***

a3 = 0.48***

c = -0.20*** (c´= -0.09**)

a2 = -0.14ns b1 = 0.15**

Fig. 2 Transgression relevance influences consumer support via moral decoupling Note Multiple step mediation results (Preacher and Hayes 2008). The dashed lines denote paths that were not hypothesized. *p \ .05; **p \ .01; ***p \ .001

168 K. Haberstroh et al.

123

apply across a more diverse range of applications and

contexts. Still, further research should explore the specific characteristics in these fields.

More narrowly, moral decoupling appears to play an

important role in explaining consumer response to immoral

actions of a firm. The results fit with the observed empirical reports that consumers keep buying from firms although

they are aware of unethical behavior (Hoffmann and

Müller 2009; Joergens 2006; Schmalz and Orth 2012). It is conceivable that the moral decoupling phenomenon offers

considerable potential in explaining consumer response to

corporate missteps beyond approaches that require people to reconstruct moral transgressions (Bandura 2010). This is

not to say that we suggest moral decoupling is the one and

only appropriate model. Instead, we feel that established approaches revolving around cognitive dissonance (e.g.,

Shu et al. 2011), protection of a positive self-concept

(Mazar et al. 2008), or upholding one’s moral standards

Table 2 Testing the moderating role of product involvement

Model 1: Moral decoupling Model 2: Extensions Model 3: Full model

b t p b t p b t p

Moral decoupling (MD) .45 8.05 .001 .45 8.11 .001 .45

8.00 .001

Product involvement .07 1.23 .220 .09

1.52 .130

Credibility of information .03 .48 .632 .03

.46 .647

MD 9 product involvement .12

2.09 .038

MD 9 credibility of information -.02

-.34 .737

F 64.80 22.20 14.33

Radj 2 .20 .20

.21

Table 3 SPSS output: Johnson-Neyman point

Moderator value(s) defining Johnson-Neyman significance region(s)

Value Below (%) Above (%)

2.006 11.111 88.889

Conditional effect of X on Y at values of the moderator (M)

Involvem Effect SE t p LLCI ULCI

1.000 .120 .197 .606 .545 -.269 .508

1.200 .143 .182 .788 .431 -.215 .501

1.400 .167 .166 1.004 .316 -.160 .494

1.600 .191 .151 1.262 .208 -.107 .488

1.800 .214 .136 1.574 .117 -.054 .482

2.000 .238 .122 1.956 .052 -.002 .478

2.006 .239 .121 1.969 .050 .000 .478

2.200 .262 .108 2.426 .016 .049 .474

2.400 .285 .095 3.006 .003 .098 .472

2.600 .309 .083 3.708 .000 .145 .473

2.800 .333 .074 4.513 .000 .188 .478

3.000 .356 .067 5.326 .000 .225 .488

3.200 .380 .064 5.955 .000 .254 .506

3.400 .404 .065 6.214 .000 .276 .532

3.600 .428 .070 6.091 .000 .289 .566

3.800 .451 .079 5.739 .000 .296 .606

4.000 .475 .089 5.313 .000 .299 .651

4.200 .499 .102 4.900 .000 .298 .699

4.400 .522 .115 4.534 .000 .295 .749

4.600 .546 .129 4.220 .000 .291 .801

4.800 .570 .144 3.953 .000 .286 .853

5.000 .593 .159 3.727 .000 .280 .907

Bold values indicate p \ .05

Fig. 3 A visual depiction of the interaction between moral decou- pling and product involvement

Consumer Response to Unethical Corporate Behavior: A Re-Examination and Extension… 169

123

(Aquino and Reed 2002) may benefit from adding a moral

decoupling perspective. Such explanations may prove su- perior to competing theories based solely on the avail-

ability of substitutes (Sen et al. 2001) or

convenience/transaction cost (John and Klein 2003) which provide less insight into the within-person processes.

Limitations and Future Research

This replication adds to the knowledge about the psycho- logical processes involved in shaping consumer response to

unethical firm behavior. At the same time, the findings as

well as contextual and methodological specifics of the study suggest at least four avenues for future research.

First, scholars should explore in more detail Why moral

decoupling should be a better predictor of consumer re- sponses than moral rationalizations. In line with their

theorizing, the psychometric measures and analyses em-

ployed in BB&R’s studies 4 and 5 indicate that consumers find moral decoupling feels less wrong and easier to justify

than moral rationalization. Extending the original and this

replication both conceptually and methodologically, future studies could assess the constructs through which effects are

channeled more directly, for example, by employing an

implicit association test (Greenwald et al. 2009). Conducting straightforward analyses of mediation (Preacher and Hayes

2004) with the new variables should then be beneficial for

better understanding the predicted motivational mechanism. Second, there is a need to further clarify When and How

the judgments impacted by moral decoupling influence

consumer behavioral response. For example, research on the ‘bad is stronger than good’ principle (for a review see

Grant and Schwartz 2011) and negativity bias (Rozin and

Royzman 2001) suggests that people often exhibit a ten- dency to focus more on negative information. Extending

this perspective to the context of moral decoupling sug-

gests that the deficiencies of markedly low levels of one type of judgment (either performance or immorality) may

overshadow the merits of higher levels of the other one.

Below a certain extent of performance, its deficiencies may outweigh ethical merits so that the firm will primarily be

seen as incompetent. In contrast, below a certain extent of

moral judgment, ethical deficiencies may overshadow performance merits, so that a firm will be primarily seen as

unethical. Integrating the negativity bias literature with the

moral decoupling perspective suggests that more research is needed to clarify what levels of judgments constitute

boundary conditions for the effect to take hold.

Third and related to the issue of When consumers are more (and less) likely to engage the moral decoupling

process is the question what situational and individual

variables may impact moral decoupling effects. The

current study is the first to analyze possible moderators

(i.e., involvement). Additional information is needed on characteristics of the individual (e.g., moral foundations,

need for cognition, ethical beliefs) or the situation (e.g.,

cognitive capacity available for processing) possibly af- fecting the relationships between information on unethical

firm behavior and moral decoupling and between moral

decoupling and behavioral outcomes. While the results suggest that high-involvement con-

sumers decouple more, it can be speculated that domains exist in which this relationship might reverse. For example,

high-involvement coffee drinkers are more likely to care

about how beans are harvested and may thus care more about fair trade certification (de Pelsmacker et al. 2005),

possibly resulting in less moral decoupling. More generally,

it seems important to distinguish the desire for hedonic (product) enjoyment from both the desire to support the

brand (or firm) and the identity-relevance of the product

domain. Future studies could tackle this question by dis- entangling motivational drivers in more detail, hereby

clarifying the individual and situational conditions for when

moral decoupling effects are more or less pronounced. Finally, while we believe that our video stimulus is

certainly more vivid and engaging than the newspaper

excerpts used by Bhattacharjee et al. (2013), the perceived realism of a firm video in which current employees de-

scribe bad firm behavior may be questioned. Given that our

aggregate measure of credibility did not permit detailing whether participants saw the employee descriptions as re-

liable within the context of the video, or whether they

thought the firm and situation themselves were real, no specific predictions can be made.

The video also included a variety of consumer responses

to information about unethical behavior which exemplified different descriptive norms and different role models. In-

cluding these scenes should serve to highlight the uncer-

tainty of choosing the right option; in turn promoting moral decoupling as a solution to adjust to both descriptive norms

simultaneously. Future study might develop experimental

designs which systematically vary the descriptive norms and the role models offered. We expect that the likelihood

that consumers use the mechanism of moral decoupling to

justify their consumption decision is higher in situations were no clear descriptive norm is presented. Moreover,

future studies might test the level of uncertainty induced by

different ways of presenting the conflicting moral situation. In conclusion, while the replication shows that the moral

decoupling model is useful for better explaining consumer

response to unethical firm behavior (and possibly more generalizable branding contexts), much can be done to

further enhance our understanding of the psychological

processes involved.

170 K. Haberstroh et al.

123

Appendix

See Table 4.

References

Agraprofit. (2013). Agraprofit—fair prices—full transparency [Agraprofit—Faire Preise—volle Transparenz]. Retrieved September 18, 2013 from http://agraprofit.de.

Ahluwalia, R., Burnkrant, R. E., & Unnava, H. R. (2000). Consumer response to negative publicity: The moderating role of commit- ment. Journal of Marketing Research, 37(2), 203–214.

Aquino, K., & Reed, A, I. I. (2002). The self-importance of moral identity. Journal of Personality and Social Psychology, 83(6), 1423–1440.

Auger, P., Devinney, T. M., & Louviere, J. J. (2007). Using best– worst scaling methodology to investigate consumer ethical beliefs across countries. Journal of Business Ethics, 70(3), 299–326.

Bal, Anjali S., Pitt, L., Berthon, P., & Des Autels, P. (2009). Caricatures, cartoons, spoofs and satires: Political brands as butts. Journal of Public Affairs, 9(4), 229–237.

Bandura, A. (1999). Moral disengagement in the perpetration of inhumanities. Personality and Social Psychology Review, 3(3), 193–209.

Bandura, A. (2010). Selective moral disengagement in the exercise of moral agency. Journal of Moral Education, 31(2), 101–119.

Bandura, A., Barbaranelli, C., Caprara, G. V., & Pastorelli, C. (1996). Mechanisms of moral disengagement in the exercise of moral agency. Journal of Personality and Social Psychology, 71(2), 364–374.

Baumeister, R. F. (2010). The self. In R. F. Baumeister & E. J. Finkel (Eds.), Advanced social psychology: The state of science (pp. 139–175). Oxford: Oxford University Press.

Table 4 Scale items and summary statistics for construct measures

Construct IFC M SD VE a

Transgression relevance (Ingram et al. 2005)

The firm’s method of production does not cause any harmr .86

There is a very small likelihood that the firm’s action will cause any harmr .86

4.48 .78 74 % .88

Judgment of performance (Bhattacharjee et al. 2013)

The firm’s low cost production is a stellar achievement – 1.90 1.06 – –

Judgment of immorality (Bhattacharjee et al. 2013)

It is morally wrong to manufacture products in this way – 4.39 1.10 – –

Moral decoupling (Bhattacharjee et al. 2013)

The method of production does not change my assessment of the firm’s performance (offering quality products at low prices)

.73

Judgments of performance should remain separate from judgments of morality .82

Reports of wrongdoing should not affect our view of the firm’s performance .72

2.16 .95 57 % .63

Product involvement (Mittal and Lee 1989)

These food products are very important to me .93

I have a strong interest in these food products .92

There are differences between the food products they offered on the market .54

3.37 .86 68 % .73

Credibility of information

In your opinion, how credible is the information on the firm’s behavior provided by Agraprofit?

– 3.36 1.30 – –

Purchase intention (Putrevu and Lord 1994)

It is very likely that I will buy the firms’ products .82

I will purchase the firms’ products the next time I go to the market .89

My willingness to buy the firms’ products is high .88

1.64 .86 75 % .81

IFC item-factor-correlation, M mean, SD standard deviation, VE variance explained, a Cronbach’s alpha, r reverse-coded

Consumer Response to Unethical Corporate Behavior: A Re-Examination and Extension… 171

123

Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D. (2001). Bad is stronger than good. Review of General Psy- chology, 5(4), 323.

Bergkvist, L., & Rossiter, J. R. (2007). The predictive validity of multiple-item versus single-item measures of the same con- structs. Journal of Marketing Research, 44(2), 175–184.

Berthon, P., Pitt, L., Ewing, M., & Carr, C. L. (2002). Potential research space in MIS: A framework for envisioning and evaluating research replication, extension, and generation. Information Systems Research, 13(4), 416–427.

Bhattacharjee, A., Berman, J. Z., & Reed, A. (2013). Tip of the hat, wag of the finger: How moral decoupling enables consumers to admire and admonish. Journal of Consumer Research, 39(4), 1167–1184.

Bhattacharya, C. B., & Korschun, D. (2008). Stakeholder marketing: Beyond the four Ps and the customer. Journal of Public Policy & Marketing, 27(1), 113–116.

Biehal, G. J., & Sheinin, D. A. (2007). The influence of corporate messages on the product portfolio. Journal of Marketing, 71(2), 12–25.

Boles, J. S., Johnson, J. T., & Barksdale, H. C. (2000). How salespeople build quality relationships: A replication and exten- sion. Journal of Business Research, 48(1), 75–81.

Brandt, M. J., IJzerman, H., Dijksterhuis, A., Dijksterhuis, A., Farach, F. J., Farach, F. J., et al. (2014). The replication recipe: What makes for a convincing replication? Journal of Experimental Social Psychology, 50(January), 217–224.

Critcher, C. R., & Dunning, D. (2013). Predicting persons’ versus a person’s goodness: Behavioral forecasts diverge for individuals versus populations. Journal of Personality and Social Psy- chology, 104(1), 28–44.

De Pelsmacker, P., Driesen, L., & Rayp, G. (2005). Do consumers care about ethics? Willingness to pay for fair-trade coffee. The Journal of Consumer Affairs, 39(2), 363–385.

Easley, R. W., Madden, C. S., & Dunn, M. G. (2000). Conducting marketing science: The role of replication in the research process. Journal of Business Research, 48(1), 83–92.

Einwiller, S. A., Fedorikhin, A., Johnson, A. R., & Kamins, M. A. (2006). Enough is enough! When identification no longer prevents negative corporate associations. Journal of the Acade- my of Marketing Science, 34, 185–194.

Evanschitzky, H., Baumgarth, C., Hubbard, R., & Armstrong, J. S. (2007). Replication research’s disturbing trend. Journal of Business Research, 60(4), 411–415.

Ewing, M. T., & Napoli, J. (2005). Developing and validating a multidimensional nonprofit brand orientation scale. Journal of Business Research, 58(6), 841–853.

Fan, Y. (2005). Ethical branding and corporate reputation. Corporate Communications: An International Journal, 10(4), 341–350.

Gelfand, M. J., Bhawuk, D. P. S., Nishii, L. H., & Bechtold, D. J. (2004). Individualism and collectivism’. In R. J. House, P. J. Hanges, M. Javidan, P. W. Dorfman, & V. Gupta (Eds.), Culture, leadership, and organizations: The GLOBE study of 62 societies (pp. 437–512). Thousand Oaks, CA: Sage.

Goldenberg, J., & Muller, E. (2014). IJRM replication corner— structure and process. Retrieved September 20, 2014 from http:// www.journals.elsevier.com/international-journal-of-research-in- marketing/news/ijrm-replication-corner-structure-and-process/.

Grant, A. M., & Schwartz, B. (2011). Too much of a good thing: The challenge and opportunity of the inverted U. Perspectives on Psychological Science, 6, 61–76.

Greenwald, A. G., Poehlman, T. A., Uhlmann, E. L., & Banaji, M. R. (2009). Understanding and using the implicit association test: III. Meta-analysis of predictive validity. Journal of Personality and Social Psychology, 97(1), 17–41.

Haidt, J., & Joseph, C. (2004). Intuitive ethics: How innately prepared intuitions generate culturally variable virtues. Daedalus, 133(4), 55–66.

Haidt, J., & Kesebir, S. (2010). Morality. In S. Fiske, D. Gilbert, & G. Lindzey (Eds.), Handbook of social psychology (5th ed., pp. 797–832). Hoboken, NJ: Wiley.

Hankinson, G. (2001). Location branding: A study of the branding practices of 12 English cities. The Journal of Brand Manage- ment, 9(2), 127–142.

Hayes, A. F., & Matthes, J. (2009). Computational procedures for probing interactions in OLS and logistic regression: SPSS and SAS implementations. Behavior Research Methods, 41(3), 924–936.

Hoffmann, S. (2011). Anti-consumption as a means to save jobs. European Journal of Marketing, 45(11/12), 1702–1714.

Hoffmann, S. (2013). Are boycott motives rationalizations? Journal of Consumer Behaviour, 12(3), 214–222.

Hoffmann, S. (2014). Does national culture impact consumer boycott prevalence? A multi-country study. European Journal of Inter- national Management, 8(2), 141–159.

Hoffmann, S., & Müller, M. (2009). Consumer boycotts due to factory relocation. Journal of Business Research, 62(2), 239–247.

Hubbard, R., & Lindsay, R. M. (2013). The significant difference paradigm promotes bad science. Journal of Business Research, 66(9), 1393–1397.

Ingram, R., Skinner, S. J., & Taylor, V. A. (2005). Consumers’ evaluation of unethical marketing behaviors: The role of customer commitment. Journal of Business Ethics, 62(3), 237–252.

Irwin, J. R., & McClelland, G. H. (2003). Negative consequences of dichotomizing continuous predictor variables. Journal of Mar- keting Research, 46(August), 366–371.

Joergens, C. (2006). Ethical fashion: Myth or future trend? Journal of Fashion Marketing and Management, 10(3), 360–371.

John, A., & Klein, J. G. (2003). The boycott puzzle: Consumer motivations for purchase sacrifice. Management Science, 49(2), 1196–1209.

Johnson, P. O., & Neyman, J. (1936). Tests of certain linear hypotheses and their application to some educational problems. Statistical Research Memoirs, 1, 57–93.

Keel, A., & Nataraajan, R. (2012). Celebrity endorsements and beyond: New avenues for celebrity branding. Psychology & Marketing, 29(9), 690–703.

Klein, J. G., & Ahluwalia, R. (2005). Negativity in the evaluation of political candidates. Journal of Marketing, 69(1), 131–142.

Koleva, S. P., Graham, J., Iyer, R., Ditto, P. H., & Haidt, J. (2012). Tracing the threads: How five moral concerns (especially Purity) help explain culture war attitudes. Journal of Research in Personality, 46(2), 184–194.

Kunda, Z. (1990). The case for motivated reasoning. Psychological Bulletin, 108(3), 480–498.

Laczniak, G. R., & Murphy, P. E. (2012). Stakeholder theory and marketing: Moving from a firm-centric to a societal perspective. Journal of Public Policy & Marketing, 31(2), 284–292.

Lucas, J. W. (2003). Theory-testing, generalization, and the problem of external validity. Sociological Theory, 21, 236–253.

Mazar, N., Amir, O., & Ariely, D. (2008). The dishonesty of honest people: A theory of self-concept maintenance. Journal of Marketing Research, 45(6), 633–644.

McCarty, J. A., Horn, M. I., Szenasy, M. K., & Feintuch, J. (2007). An exploratory study of consumer style: Country differences and international segments. Journal of Consumer Behaviour, 6(1), 48–59.

172 K. Haberstroh et al.

123

Miller, F., & Laczniak, G. (2011). The ethics of celebrity-athlete endorsement: What happens when a star steps out of bounds? Journal of Advertising Research, 51(3), 499–510.

Mittal, B., & Lee, M.-S. (1989). A causal model of consumer involvement. Journal of Economic Psychology, 10(3), 363–389.

Monroe, K. (1992). On replications in consumer research: Part 1. Journal of Consumer Research, 19, i–iii.

Olins, W. (2002). Branding the nation. The historical context. Journal of Brand Management, 9(4), 241–248.

Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36, 717–731.

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(3), 879–891.

Putrevu, S., & Lord, K. R. (1994). Comparative and noncomparative advertising: Attitudinal effects under cognitive and affective involvement conditions. Journal of Advertising, 23(2), 77–91.

Rawwas, M. Y. A., & Singhapakdi, A. (1998). Do consumers’ ethical beliefs vary with age? A substantiation of Kohlberg’s typology in marketing. Journal of Marketing Theory and Practice, 6(2), 26–38.

Rosenthal, R. (1990). Replication in behavioral research. Journal of Social Behavior & Personality, 5(4), 1–30.

Rozin, P., & Royzman, E. B. (2001). Negativity bias, negativity dominance, and contagion. Personality and Social Psychology Review, 5, 296–320.

Schmalz, S., & Orth, U. R. (2012). Brand attachment and consumer emotional response to unethical firm behavior. Psychology and Marketing, 29, 869–884.

Sen, S., Gurhan-Canli, Z., & Morwitz, V. (2001). Withholding consumption: A social dilemma perspective on consumer boycotts. Journal of Consumer Research, 28(3), 399–417.

Shepherd, I. (2004). Religious marketing. Reflections from the other side of politics. Journal of Public Affairs, 4(3), 317–341.

Shu, L. L., Gino, F., & Bazerman, M. H. (2011). Dishonest deed, clear conscience: When cheating leads to moral disengagement and motivated forgetting. Personality and Social Psychology Bullet- in, 37, 330–349.

Spiller, S. A., Fitzsimons, G. J., Lynch, J. G, Jr, & McClelland, G. H. (2013). Spotlights, floodlights, and the magic number zero: Simple effects tests in moderated regression. Journal of Marketing Research, 7(April), 277–288.

Thøgersen, J. (2004). A cognitive dissonance interpretation of consistencies and inconsistencies in environmentally responsible behavior. Journal of Environmental Psychology, 24(1), 93–103.

Thomson, M. (2006). Human brands: Investigating antecedents to consumers’ strong attachments to celebrities. Journal of Mar- keting, 70(3), 104–119.

Till, B. D., & Shimp, T. A. (1998). Endorsers in advertising: The case of negative celebrity information. Journal of Advertising, 27, 67–82.

Vitell, S. J. (2003). Consumer ethics research: Review, synthesis and suggestions for the future. Journal of Business Ethics, 43(1–2), 33–47.

Witkowski, T. H., & Reddy, S. (2010). Antecedents of ethical consumption activities in Germany and the United States. Australasian Marketing Journal, 18(1), 8–14.

Witkowski, T. H., & Wolfinbarger, M. F. (2002). Comparative service quality: German and American ratings across service settings. Journal of Business Research, 55(11), 875–881.

World Values Survey Association (2009) Official aggregate data file v.20090901. Retrieved 15 May 2010, from http://www.worldva luessurvey.org.

Consumer Response to Unethical Corporate Behavior: A Re-Examination and Extension… 173

123

���������� �� ���������� �� � � ������� � � ���� ���� �� ������������ ��� ������ �� ���

�����������