Position Paper, Values, ethics and sustainability shape much of the more interesting news events. Scan the press, magazines, on-line news sources, idea handout sheets, and TV, and choose a topic from

The Effectiveness of CSR History in Countering Negative Publicity Joe¨ lle Vanhamme Bas Grobben ABSTRACT. Corporate crises call for effective communication to shelter or restore a company’s reputa- tion. The use of corporate social responsibility (CSR) claims may provide an effective tool to counter the negative impact of a crisis, but knowledge about its effectiveness is scarce and lacking in studies that consider CSR commu- nication during crises. To help fill this gap, this study investigates whether the length of company’s involvement in CSR matters when it uses CSR claims in its crisis communication as a means to counter negative publicity.

The use of CSR claims in crisis communication is more effective for companies with a long CSR history than for those with a short CSR history, and consumer skepticism about claims lies at the heart of this phenomenon.

KEY WORDS: crisis communication, CSR history, negative publicity, skepticism, reputation Many real-life examples (e.g., IKEA, The Body Shop, Texaco, Nike) serve to show how fragile company image can be when a corporation is accused publicly of misbehavior. Whether the accusations are true or not, the resulting negative publicity causes particular damage to how consumers perceive the company and its products (e.g., Dean, 2004; De Raaf,2000; Riezebos,1996; Rossiter and Percy,1998). A damaged reputation can impact on the company’s sales and pro ts, especially if it fails to respond adequately to the crisis. For example, Texaco lost up to $1.1 billion after being subject to allegations of discrimination (Bollen,2004); Nutricia, a Dutch food company that recalled morethan 10 million contaminated jars of its baby food, suffered $20 million in losses, lost substantial market share in the Dutch market, and experienced a 7.5% stock price decrease after announcing the contami- nation (De Raaf,2000).

In crises such as these, companies must engage in effective communication to shelter or restore their reputation, which has been jeopardized by the negative publicity (e.g., Benoit,1997; De Raaf, 2000; Sturges,1994). However, knowledge about effective crisis communication strategies and how consumers react to these efforts remain rather limited in practice (Mitroff,2001).

A decade ago, crisis scholars argued that corporate social responsibility (CSR) communication might be an effective tool to counter the negative impact of a crisis (Coombs,1995). Similar to Maignan et al.

(1999), we de ne CSR as the extent to which organizations meet the legal, economic, ethical, and discretionary responsibilities placed on them by various stakeholders. Since the mid-1990s, business experts increasingly have stressed the importance of social initiatives to build strong company reputa- tions. Several polls also highlight that consumers consider such corporate activities very important (e.g., Cone Corporate Citizenship Study,2004; MORI CSR Research,2002). And, academic lit- erature has established that CSR in uences con- sumer evaluations of a company and its products positively (e.g., Brown and Dacin,1997).

Consumer support for CSR suggests that corporate communication about social efforts could help a company build a reputation that might protect its image against negative publicity or help restore it.

Business practices seem to con rm this belief.

Varadarajan and Menon (1988), in their survey of case histories, nd that companies use cause-related Joe¨ lle Vanhamme is assistant professor at Erasmus University Rotterdam and member of ERIM Research Institute, the Netherlands.

Bas Grobben is Trade Marketeer at Henkel Benelux B.V, the Netherlands.

Journal of Business Ethics (2009) 85:273–283 Springer 2008 DOI 10.1007/s10551-008-9731-2 marketing, a speci c aspect of CSR, to realize various objectives, including thwarting negative publicity.

However, studies that examine the effectiveness of CSR communication during corporate crises have been scarce and investigate only its in uence in the context of a product-harm crisis. For example, CSR reduces the risk of brand damage in a product-harm crisis, and consumers tend to hold a company less responsible for a crisis when it possesses a strong CSR reputation (Klein and Dawar,2004). With just limited studies of the effectiveness of CSR crisis communi- cation, we lack clear information about the effective implementation of proper CSR communication during crises. This study attempts to ll that gap.

Speci cally, we determine whether the length of company involvement in CSR matters when using CSR claims as a means of countering negative pub- licity. Using CSR communication during a crisis represents a type of defense of company legitimacy, but in defending themselves, companies may protest too much, which makes their attempts seem suspi- cious to consumers and therefore produces the opposite effects rather than those desired (Ashforth and Gibbs,1990). Moreover, the existence of a vested interest and the lack of a strong reputation are known to undermine the credibility of legitimation attempts (McGuire,1985). CSR communication is ultimately a persuasion attempt of the company to create positive consumer perceptions. Thus, we suggest that com- panies involved for a long time in CSR activities when the crisis breaks out should be seen as less opportu- nistic and should enjoy a stronger reputation in terms of CSR than companies with shorter-term CSR involvement. As a result, the CSR crisis communi- cation should be more effective for companies with a long CSR history than for those with a short CSR history. We also propose that consumer skepticism about CSR claims determines this phenomenon.

Therefore, we build our theoretical framework and hypotheses on the basis of the concepts of organiza- tional legitimation and consumer skepticism.

Hypotheses CSR and company legitimacy Using Suchman’s (1995, p. 574) framework, which states that ‘‘Legitimacy is a generalized perception orassumption that the actions of an entity are desirable, proper, or appropriate within some socially con- structed system of norms, values, beliefs, and de - nitions,’’ Handelman and Arnold (1999) distinguish two types of legitimation relevant for our present study: pragmatic and social. A rm gains pragmatic legitimacy through performative actions, which demonstrate the ‘‘company’s congruence with the norms of the task environment’’ (Handelman and Arnold,1999, p. 35), and include decisions about functional attributes, such as price, location, and assortment (e.g., offer products at the right price and quality). However, pragmatic legitimacy is not suf- cient for a company’s long-run survival. In the short term, shifts in consumer tastes and direct competitive threats can change the norms of the competitive environment, so that legitimacy must be earned constantly (Suchman,1995). To enjoy con- sumer support in the long run, companies must achieve social legitimacy, which ensues if the com- pany’s institutional actions are compatible with the broader social norms of the community, as when it engages in CSR initiatives to donate to charities (Handelman and Arnold,1999). Social legitimacy is stronger than pragmatic legitimacy in protecting a company against short-run changes in the competi- tive environment (i.e., because of its CSR associa- tions, the company maintains a positive image even if consumers change their attitudes about the func- tional bene ts of its products). Furthermore, insti- tutional actions strengthen the effectiveness of the performative initiatives, because CSR associations can cause consumers to make positive inferences about unknown functional company bene ts.

When a crisis about a company’s responsibility arises, the company must defend its social legitimacy.

Case evidence shows that companies communicate about their CSR initiatives as a means to defend their reputation (Varadarajan and Menon,1988), so a company’s past performance provides an important piece of information in a crisis. Consumers retain more positive attitudes toward companies with good reputations than toward those with bad reputations (Coombs and Holladay,1996; Dawar and Pillutla, 2000; Dean,2004) and the general public seems more willing to forgive a company with a positive performance history than one with a history of problems (e.g., Barton,1993). That is, a positive performance history results in a halo effect by 274Joe¨ lle Vanhamme and Bas Grobben projecting a positive company image and thereby creating credibility and trustworthiness among the general public. Referring to a positive CSR history therefore should be an effective means to defend company legitimacy in a crisis situation, but its overall effectiveness likely depends on the length of the CSR history. As Suchman (1995, p. 574) states, ‘‘an organization may occasionally depart from societal norms yet retain legitimacy because the departures are dismissed as unique.’’ Companies with a long history of CSR involve- ment have earned the trust and goodwill of their stakeholders. This goodwill becomes corporate credit, such that the rm can deviate occasionally from social norms without seriously damaging its reputation (Ashforth and Gibbs,1990; Suchman, 1995). In other words, a long history of ‘‘good’’ actions should provide leverage for defending the company against challenges to its legitimacy.

Companies with only recent involvement in CSR initiatives, in contrast, do not enjoy this buffer of goodwill earned over the years. Hence:

H1: Consumer perceptions of a company and its products during a crisis are lower for compa- nies with a short CSR history than for com- panies with a long CSR history.

Consumer skepticism and company legitimacy Using CSR history as a buffer against the effects of a crisis can be perceived as an attempt to compensate for illegitimacy in one area by displaying legitimate behavior in another. In trying to defend their threatened legitimacy, however, companies can ‘‘protest too much,’’ which may trigger a vicious dynamic (Ashforth and Gibbs,1990). Legitimation involves a process of social construction, negotiation, and labeling (Ashforth and Gibbs,1990); it is a ‘‘collective making of meaning’’ (Neilsen and Rao 1987, p. 524). Therefore, consumers probably are aware of the vested interest the company has in attempting to defend its legitimacy (Ashford and Gibbs,1990), which undermines the company’s credibility in claiming legitimacy (McGuire,1985).

Ashford and Gibbs (1990) suggest that the greater the need for legitimation, the more skeptical con- sumers are of legitimation attempts, a phenomenonalso known as the ‘‘self-promotor’s paradox’’ (Jones and Pittman,1982), which represents a crucial caveat to legitimacy management (Suchman,1995). That is, people tend to discount attempts to defend the self if they perceive the defender as manipulative (e.g., the defense is a ploy to gain acceptance) or self- serving (Baumeister and Scher,1988).

Marketing literature further substantiates con- sumer skepticism about companies’ use of CSR activities as a tool to attain legitimacy. Skepticism about companies’ true motivation to engage in CSR emerges in several consumer surveys; for example, Webb and Mohr (1998) nd that consumers are more skeptical about the cause-related marketing efforts of for-pro t companies than of the same efforts by non-governmental organizations because they perceive the for-pro t companies’ motivations as mainly self-serving. Overall, consumers fear that CSR is just a ‘‘gimmick’’ that rms use to manip- ulate them (Webb and Mohr,1998, p. 234), which results in negative attitudes toward such rms (e.g., Ellen et al.,2002; Mohr et al.,2001). Corroborating work by Webb and colleagues, Forehand and Grier (2003) show that consumers react negatively to CSR communications when they feel they are being de- ceived. The belief that CSR is mostly for show or motivated by pro t interests rather than altruism also appears in the business press (The Economist,2005).

In connection with the legitimacy approach, consumer skepticism can be explained according to the persuasion knowledge model (PKM) developed by Friestad and Wright (1994). This model inte- grates theories about attitude change (e.g., Petty and Ciacoppo,1986) to explain how consumers react to marketing persuasion attempts. It posits that con- sumers hold and develop knowledge about market- ers’ persuasion attempts (e.g., advertisements), which in uences their responses to these attempts. In addition, the PKM suggests consumers have opin- ions about the appropriateness and effectiveness of marketing tactics, and their skepticism relates to the amount of persuasion knowledge they use to develop judgments about the persuasive marketing communication (Friestad and Wright,1994). In- creased skepticism regarding the ‘‘true’’ underlying motives of a persuasive message inhibits its effec- tiveness (Campbell and Kirmani,2000; Holt,2002; Lafferty and Goldsmith,1999; Vonk,1998,1999). The Effectiveness of CSR History in Countering Negative Publicity275 On the basis of the legitimacy approach and PKM, we expect the length of CSR history to serve as a cue that consumers use to derive their percep- tions of the motivations companies have to engage in CSR initiatives. Because consumers probably per- ceive companies with short CSR histories as more accountable for the crisis (i.e., they lack the kind of goodwill that companies with longer CSR histories have), they also likely infer more self-serving and less altruistic motivations for these companies. That is, when companies with a short history use CSR claims in crisis communication, consumers probably view it as a ‘‘quick x’’ or trick to restore their image. Consumers thus might be more skeptical about the altruistic intent of these companies’ CSR, which creates more negative perceptions about the company and its products. In this sense, we expect the effect of CSR history length on consumers’ perceptions to be mediated by the level of skepticism that consumers develop. H2: Communicating about CSR involvement to defend the company against negative publicity leads to more consumer skepticism when the company has a short CSR history than when the company has a long CSR history. H3: In a crisis context, skepticism mediates the impact of the length of CSR history on con- sumers’ perceptions of the company and its products.

We present our research model in Figure1, which includes the length of CSR history, skepticism, and consumers’ perceptions, as well as consumer support of CSR as a covariate. This for the reason that consumers vary in their stance toward CSR actionsby rms, which in uences their responses to these initiatives (Webb and Mohr,1998). As Sen and Bhattacharya (2001) nd, consumers who are more supportive of CSR efforts embrace more positive perceptions of companies involved in CSR than those who do not support CSR efforts, because the former perceive a greater identity overlap with CSR companies than the latter. Methodology Procedure and stimuli To test the hypotheses, we used a between-subject experimental design with two conditions (long versus short CSR history). In both conditions, respondents rst read two texts. The rst text con- sists of an of cial press release from a ctive foreign cosmetic company planning to open a point of sales in the Netherlands, describing the company, its products, and its CSR activities (i.e., responsible human resources, environmental protection, and charitable contributions to cancer-related organiza- tions). We adapted the content of the text from Swaen and Vanhamme (2004) to the purposes of our study by including a manipulation of CSR history length. These authors showed that the company description is realistic and believable. We used this ctive foreign company to ensure the absence of any a priori knowledge of the company among respon- dents, which could in uence the results.

The second text consists of a newspaper article discussing the issue of animal testing and accusing the ctive foreign company of using animal testing to develop its products. We borrowed this text from Swaen and Vanhamme (2004), who demonstrate that the newspaper article signi cantly and nega- tively in uences consumer perceptions.

We counterbalanced the order of the two texts across conditions and manipulated the length of the company’s CSR history (long versus short) by including three statements in the press release (heading, text itself, and text box in the middle of the press release with a quote from the company’s CEO). These statements depict the company’s CSR involvement as either long or short. In the long history condition, the three statements indicate the company has been involved in social responsibility Consumer SkepticismCovariate: Consumer Support of CSR Consumer Perceptions:

- Company - Product - Company integrity Length of CSR History Figure 1. Research model.Notes: Although we use text order counterbalancing as an additional covariate in the model, we do not depict it, because its purpose is statis- tical rather than theoretical.

276Joe¨ lle Vanhamme and Bas Grobben activities for 10 years, whereas in the short history condition, they note that the company started to be active in CSR this year.

After reading the two texts, respondents lled in a questionnaire that measures the dependent and independent variables (company and product per- ceptions, perceived company integrity, skepticism, and support for CSR causes, including protection of human, consumer, animal, employee rights, and the environment) and includes questions related to demographics, the perceived purpose of the study (no respondent guessed the actual purpose), and room for any additional comments.

Measures To measure perceived integrity, we used Gurviez and Korchia’s (2002) items – the company is sincere toward its customers, honest toward its customers, and clearly shows interest toward its customers – measured on 10-point scales (1 = totally disagree, 10 = totally agree). Similar to Sen and Bhattacharya (2001), we measured consumers’ perception of the company as their overall impression rating (1 = very unfavorable; 10 = very favorable). The product perception measure is similar but measures the overall impression of the products on a 10-point scale. To measure consumers’ support for CSR, we asked respondents to state their degree of support for various CSR activities (e.g., environmental protec- tion, animal rights), similar to Sen and Bhattacharya’s (2001) approach. Finally, to assess the degree of skepticism consumers hold with respect to claims made by companies, we included Obermiller and Spangenberg’s (1998) scale of ad skepticism (9 items,10-point scales, 1 = totally disagree, 10 = totally agree) in the questionnaire.

Respondents and data collection A convenience sample of 125 Dutch consumers participated in the study and were randomly assigned to the different conditions. Respondents include owners, volunteers, and employees working for various companies and stores located in major cities in the Netherlands, as well as acquaintances of the researchers who were not aware of the purpose of the study. We approached potential respondents at home or on company/store premises and asked them to take the questionnaire home and return it within a week. The composition of respondents in this study represents a fairly representative spectrum of the Dutch population in terms of age, gender, education level, and household status (see TableI).

Pretests Before conducting the experiment, we used a small- scale pretest with 10 non-student respondents surveyed on a train to check our manipulation of the length of the company’s CSR history. The respondents viewed either the short or long history version of the press release. All respondents accurately answered a question about the length of the CSR involvement of the company after having read the text.

In addition, we back-translated the stimuli and questionnaires from English to Dutch or French to Dutch and then pretested each for clarity and the absence of ambiguities. TABLE I Respondent demographics Variable Civil status Married: 47%; divorced: 4%; widowed: 2%; single: 29%; living together: 18% Education Elementary school: 4%; professional education: 9%; secondary school: 50%; higher education: 37% Gender Male: 48%; female: 52% Age Mean: 39 years old; standard deviation: 13 years; range: 19–83 years Children Mean: 1; standard deviation: 1; range: 0–3 The Effectiveness of CSR History in Countering Negative Publicity277 Results We conducted the usual checks for unidimension- ality and reliability for the multi-items scales. Two of the skepticism and one of the integrity items were discarded because they reduced the reliability of their respective measure. We provide descriptive statistics of the study variables in TableII. Cronbach alphas are .92 for skepticism and .85 for integrity (the Pearson correlation between the two integrity items is .74).

As we expected and in support of our rst hypothesis, perceived integrity is signi cantly greater for companies with a long CSR history than for those with a short CSR history (t(123) = 2.079, p 1)tail = .020). Company and product perceptions also are signi cantly higher but ata= 0.10 (com- pany:t(123) = 1.427,p 1)tail = .078; product:

t(123) = 1.350,p 1)tail = .090). In Figure2,we provide the magnitudes of the respective means.

In addition, the level of consumer skepticism is higher for companies with a short history than for companies with a long history (t(120) =)2.332, p 1)tail = .011), in support of H2; we again provide the magnitude of the mean in Figure2.

Finally, the data provide full support to H3.

Speci cally, to test for full mediation, we conducted three regressions and checked four conditions (Baron and Kenny,1986, Chumpitaz and Vanhamme,2003).

First, the in uence of the independent variable (mediated variable) on the dependant variable must be signi cant when the mediator is excluded from the regression (see Regression 1, TableIII). The data show a signi cant impact of the length of CSR historyon company perceptions, product perceptions, and perceived integrity, so the rst condition is satis ed.

Second, the independent variable must signi - cantly in uence the potential mediator. This con- dition is also met; the length of CSR history has a signi cant effect on skepticism (see Regression 2, TableIV). Third, for full mediation, the in uence of the independent variable on the dependant variable must become not signi cant when the mediator is included in the regression, and fourth, the mediator must have a signi cant in uence on the dependant variable. As we show in TableIII(Regression 3), the in uence of the length of CSR history on TABLE II Descriptive statistics for model variables NMinimum Maximum Mean Standard deviation Company perception 125 1.000 9.000 5.760 1.873 Product perception 125 1.000 9.000 6.430 1.889 Perceived integrity 125 1.000 10.000 4.896 1.826 Skepticism 122 2.860 10.000 7.032 1.443 Environment 123 1.000 10.000 6.690 1.955 Human rights 125 2.000 10.000 7.310 1.820 Animal rights 125 1.000 10.000 7.290 1.913 Consumer rights 125 1.000 10.000 6.460 2.157 Employee rights 125 1.000 10.000 6.940 1.872 0 1 2 3 4 5 6 7 8 Company perceptionProduct perceptionPerceived integritySkepticism Short CSR history Long CSR history Figure 2. Means for short and long CSR histories. 278Joe¨ lle Vanhamme and Bas Grobben TABLE III Results of regressions 1 and 3 (H3) Company perception Product perception Perceived integrity BSE BetatSig.BSE BetatSig.BSE BetatSig.

Regression 1:

(Constant) 9.041 0.768 11.768 0.000 9.035 0.821 11.004 0.000 5.874 0.854 6.878 0.000 Length CSR history a )0.253 0.147)0.137)1.7260.044)0.264 0.157)0.143)1.6860.047)0.324 0.163)0.179)1.9910.025 Covariates:

Environment 0.011 0.090 0.011 0.118 0.906 0.116 0.097 0.122 1.198 0.233)0.173 0.100)0.186)1.718 0.088 Human rights 0.062 0.115 0.057 0.537 0.593 0.000 0.123 0.000 0.003 0.997 0.025 0.128 0.024 0.197 0.844 Animal rights)0.497 0.105)0.511)4.736 0.000)0.453 0.112)0.466)4.040 0.000)0.066 0.117)0.069)0.565 0.573 Consumer rights 0.181 0.120 0.205 1.508 0.134 0.086 0.129 0.098 0.672 0.503 0.279 0.134 0.322 2.085 0.039 Employee rights)0.188 0.120)0.189)1.565 0.120)0.082 0.128)0.082)0.636 0.526)0.187 0.133)0.193)1.402 0.164 Counterbalancing 0.403 0.148 0.218 2.720 0.008 0.244 0.159 0.132 1.536 0.127)0.096 0.165)0.053)0.583 0.561 Regression 3 (Constant) 11.958 1.058 11.297 0.000 12.624 1.107 11.406 0.000 11.085 1.029 10.769 0.000 Length CSR history a )0.124 0.145)0.066)0.8510.199)0.112 0.152)0.061)0.7390.231)0.114 0.141)0.062)0.8030.214 Skepticism)0.388 0.102)0.296)3.8130.000)0.453 0.106)0.349)4.2640.000)0.720 0.099)0.562)7.2850.000 Covariates:

Environment 0.009 0.089 0.009 0.097 0.923 0.087 0.093 0.089 0.939 0.350)0.164 0.086)0.171)1.901 0.060 Human rights 0.094 0.113 0.084 0.835 0.405 0.028 0.118 0.025 0.238 0.812 0.108 0.110 0.098 0.982 0.328 Animal rights)0.557 0.102)0.564)5.439 0.000)0.507 0.107)0.517)4.739 0.000)0.182 0.100)0.188)1.826 0.071 Consumer rights 0.171 0.115 0.191 1.484 0.141 0.076 0.121 0.085 0.629 0.531 0.261 0.112 0.298 2.328 0.022 Employee rights)0.175 0.115)0.175)1.524 0.130)0.076 0.120)0.076)0.629 0.530)0.169 0.112)0.172)1.512 0.133 Counterbalancing 0.363 0.146 0.195 2.491 0.014 0.176 0.153 0.095 1.152 0.252)0.142 0.142)0.078)1.004 0.317 aDummy variable coded)1 for long CSR history and 1 for short CSR history.

R2are 0.383, 0.317, and 0.393 for company perception, product perception, and perceived integrity, respectively.

Notes:p-values in bold are one-tailed; all hypotheses are directional hypotheses (all ps<.05). The Effectiveness of CSR History in Countering Negative Publicity279 company perceptions, product perceptions, and perceived integrity is not signi cant, whereas that of skepticism is signi cant. Thus, all four conditions pertaining to full mediation are satis ed. Further corroborating this result, we nd that the Sobel test for mediation (Baron and Kenny,1986) is signi cant for product perceptions and perceived integrity at a= .05 (Sobel test value = 1.968,p= .049) and for company perceptions ata= .06 (Sobel test value = 1.908;p= .056).

Discussion From a marketing communication perspective, crisis accusations are critical because they can have a del- eterious impact on consumers’ perceptions of the incriminated company and its products. Years of work invested in building a strong company repu- tation can be annihilated virtually overnight by an article published or broadcast in the media because of the vast attention crises usually attract. The media tend to emphasize crises, especially when their im- pact makes them newsworthy events for journalists (Heath,1998). Therefore, companies absolutely must react properly to crises. As is evident from the literature, CSR involvement might offer an adequate tool to counter the negative impact induced by crises, because consumer awareness of a company’s CSR involvement usually leads to posi- tive evaluations (e.g., Brown and Dacin,1997; Senand Bhattacharya,2001). However, the in uence of CSR in crisis situations has been studied only spo- radically, so the literature lacks clear guidelines that companies can use to determine their CSR claims in communication strategies and thereby limit the deleterious impact of a crisis. We provide a rst step in lling this gap.

Because of the advantages companies can gain from CSR involvement, they largely have embraced CSR initiatives through donations or community programs. Nevertheless, we clearly show that com- panies facing a crisis should be cautious when using CSR as a tool to defend their reputations. Compa- nies with a long history of involvement in CSR activities have earned the right to mention their good deeds without raising the suspicion of the consumers. However, companies that only recently have started a CSR program have not earned that right. Should these companies give in to temptation and communicate publicly about their newly initiated CSR activities, they will face consumer skepticism, disbelief, and, in turn, more negative company and product perceptions, as well as doubts about the company’s integrity.

Short-term company CSR involvement seems to act as a cue that triggers skepticism about its underlying motivation, which leads to more nega- tive perceptions of the company, its products, and its integrity. Thus, using the company’s brief CSR history as a means to defend its maligned reputation seems to be a gimmick or a quick x. In contrast, TABLE IV Results of regression 2 (H3) BSE BetatSig.

(Constant) 7.063 0.723 9.770 0.000 Length CSR history a 0.301 0.132 0.211 2.2780.013 Covariates:

Environment 0.012 0.082 0.017 0.150 0.881 Human rights 0.128 0.104 0.148 1.224 0.223 Animal rights)0.150 0.094)0.199)1.599 0.113 Consumer rights)0.035 0.107)0.052)0.329 0.743 Employee rights 0.035 0.107 0.045 0.326 0.745 Counterbalancing)0.039 0.136)0.027)0.288 0.774 aDummy variable coded)1 for long CSR history and 1 for short CSR history.

Notes:p-values in bold are one-tailed; all hypotheses are directional hypotheses; dependent variable: skepticism.

280Joe¨ lle Vanhamme and Bas Grobben companies with a long CSR track record can dismiss a crisis as a one-time incident, and consumers will tend to believe their more credible CSR claims.

Our study further highlights that consumer skepticism helps clarify the effect of CSR history in crisis communication and thus extends existing knowledge about the in uence of consumer skep- ticism in marketing communication, a topic that deserves further research attention (Forehand and Grier,2003). Academic research indicates that con- sumers tend to be increasingly critical and skeptical about in uences by marketers (Eisend,2004), pos- sibly as a result of the changes that have accompanied the societal transition from modernism to post- modernism. In a postmodern period, society is characterized by pluralism and fragmentation, which creates liberalism in choices and increased insecurity in consumption contexts (Firat and Venkatesh, 1995). In this setting, consumers adhere to a variety of values, adopt various lifestyles, and reject the authorities of modernity, making them more expe- rienced, more critical, and more skeptical about marketing in uences (Eisend,2004; Elliot et al., 1993). This general consumer skepticism appears in various CSR surveys, such as the 2004 Cone Corporate Citizenship Study, which reported that a majority of Americans surveyed considered infor- mation about CSR activities that they gain from third-party sources (i.e., friends, relatives, govern- ment, news organizations, Internet) far more credi- ble than that communicated by sources controlled by the company itself.

If the trend of increased skepticism continues in the future, it might become increasingly dif cult and precarious to use CSR involvement in communi- cation campaigns to counter the deleterious effects of a crisis. Companies therefore should avoid falling into the pitfall of being perceived as protesting too much. Practitioners should be aware of the powerful impact of cues that decrease the credibility of their claims; bragging about a short CSR track record is one of them. Companies should also keep in mind that some demographic variables in uence the level of consumer skepticism. For example, consumers are usually more knowledgeable about marketing tactics as they grow older or gain more education (Friestad and Wright,1994; Obermiller and Spangenberg, 1998), which makes older, more educated con- sumers dif cult targets to persuade with marketingtactics. A thorough understanding of the factors that increase consumers’ skepticism and thorough knowledge of the characteristics of the target market therefore seem essential in crisis situations, because with them, marketers can adapt and ne-tune their crisis communication for speci c target markets.

This study investigates only one type of crisis (animal testing accusations) and uses a ctive cos- metic company. Although we have no reason to believe that the phenomenon we discover will not hold in different settings, to increase the external validity of our ndings, further research should replicate our study with different types of crises, CSR activities, and companies.

Moreover, additional research might de ne the minimum length of CSR involvement that will not raise consumers’ suspicions, because instead of studying the in uence of CSR history as a contin- uum, we simply contrasted the in uence of a short versus long history.

In addition, our recommendations are based on the reactions of consumers; we do not investigate other stakeholders’ points of view (e.g., employees, share- holders). The implications of our study are thus con- ned to external, as opposed to internal, marketing communication practices. It would be interesting to test whether our model also applies to other stake- holders, which would extend its generalizability as well.

Overall, we do not claim that companies should not invest in social activities. Rather, we suggest it is better not to mention such good deeds in a crisis communication strategy when it is likely that the claims will generate disbelief among consumers, such as when the company’s involvement in CSR activ- ities has been relatively short. This knowledge should prove helpful for marketing practitioners when they must craft CSR communication strate- gies for consumers. References Ashforth, B. E. and B. W. Gibbs: 1990, ‘The Double- Edge of Organizational Legitimation’,Organization Science1(2), 177–194.

Baron, R. M. and D. A. Kenny: 1986, ‘The Moderator– Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic and Statistical Con- siderations’,Journal of Personality and Social Psychology 51(6), 1173–1182. The Effectiveness of CSR History in Countering Negative Publicity281 Barton, L.: 1993,Crisis in Organizations: Managing and Communicating in the Heat of Chaos(College Divisions South-Western Publishing, Cincinnati, OH).

Baumeister, R. F. and S. J. Scher: 1988, ‘Self-Defeating Behavior Patterns Among Normal Individuals:

Review and Analysis of Common Self-Destructive Tendencies’,Psychological Bulletin104(1), 3–22.

Benoit, W. L.: 1997, ‘Image Repair Discourse and Crisis Communication’,Public Relations Review23(2), 177–186.

Bollen, A.: 2004, ‘The Rise and Rise of Non-Financial Reporting: How to Use Research to Measure Your Reputation’,MORI White Paper, January, available at http://www.mori.com.

Brown, T. J. and P. A. Dacin: 1997, ‘The Company and the Product: Corporate Associations and Consumer Product Responses’,Journal of Marketing61(1), 68–84.

Campbell, M. C. and A. Kirmani: 2000, ‘Consumers’ Use of Persuasion Knowledge: The Effects of Accessibility and Cognitive Capacity on Perceptions of an In uence Agent’,Journal of Consumer Research27(1), 69–83.

Chumpitaz, R. and J. Vanhamme: 2003, ‘Les Processus Mode´ rateurs et Me´ diateurs: Distinction Conceptuelle, Aspects Analytiques et Illustrations’,Recherches et Applications en Marketing18(2), 67–100.

Cone Corporate Citizenship Study: 2004, available at http://www.mybizwarehouse.com/2ndbusiness/2004 ConeCorporateCitizenshipStudy.pdf.

Coombs, W. T.: 1995, ‘Choosing the Right Words: The Development of Guidelines for the Selection of ‘Appropriate’ Crisis-Response Strategies’,Management Communication Quarterly8(4), 447–476.

Coombs, W. T. and S. J. Holladay: 1996, ‘Communi- cation and Attributions in a Crisis: An Experimental Study of Crisis Communication’,Journal of Public Relations Research8(2), 279–295.

Dawar, N. and M. M. Pillutla: 2000, ‘Impact of Product- Harm Crises on Brand Equity: The Moderating Role of Consumer Expectations’,Journal of Marketing Research37(2), 215–226.

Dean, D. H.: 2004, ‘Consumer Reaction to Negative Publicity: Effects of Corporate Reputation, Response, and Responsibility for a Crisis Event’,Journal of Busi- ness Communication41(2), 192–211.

De Raaf, D. I.: 2000,Protecting Brands: How to Respond on Adverse Brand Publicity(Van Spaenendonck Drukkerij B.V, Tilburg).

The Economist: 2005, ‘The Good Company: A Survey of Corporate Social Responsibility’, January 22, 3–8.

Eisend, M.: 2004, ‘Is It Still Worth to be Credible? A Meta-Analysis of Temporal Patterns of Source Cred- ibility Effects in Marketing’, in B. E. Kahn and M. F.

Luce (eds.),Advances in Consumer Research, Vol. 31(Association for Consumer Research, Valdosta, GA), pp. 352–357.

Ellen, P. S., Mohr, L. A. and Webb, D. J.: 2002. ‘Pure or Mixed Motives: Consumer Attributions for Corporate Pro-Social Marketing Programs’, in Punam Anand Keller and Dennis W. Rook (eds.),Advances in Consumer Research(Association for Consumer Research, Valdosta, GA), pp. 322–324. Full paper obtained from the authors.

Elliot, R., S. Eccles and M. Hodgson: 1993, ‘Re-Coding Gender Representations: Women, Cleaning Products, and Advertising’s ‘‘New Man’’’,International Journal of Research in Marketing10(3), 311–324.

Firat, A. F. and A. Venkatesh: 1995, ‘Liberatory Post- modernism and the Reenchantment of Consumption’, Journal of Consumer Research22 (3), 239–267.

Forehand, M. R. and S. Grier: 2003, ‘When is Honesty the Best Policy? The Effect of Stated Company Intent on Consumer Skepticism’,Journal of Consumer Psychology13(3), 349–356.

Friestad, M. and P. Wright: 1994, ‘The Persuasion Knowledge Model: How People Cope with Persua- sion Attempts’,Journal of Consumer Research21(1), 1–31.

Gurviez, P. and M. Korchia: 2002, ‘Proposition d’une Echelle de Mesure Multidimensionnelle de la Con - ance dans la Marque’,Recherche et Applications en Mar- keting17(3), 41–62.

Handelman, J. M. and S. T. Arnold: 1999, ‘The Role of Marketing Actions with a Social Dimension: Appeals to the Institutional Environment’,Journal of Marketing 63(3), 33–48.

Heath, R.: 1998,Crisis Management for Managers and Executives: Business Crises – the De nitive Handbook to Reduction, Readiness, Responses and Recovery(Financial Times Professional Limited, London).

Holt, D. B.: 2002, ‘Why do Brands Cause Trouble? A Dialectical Theory of Consumer Culture and Brand- ing’,Journal of Consumer Research29(1), 70–91.

Jones, E. E. and T. S. Pitman: 1982, ‘Toward a General Theory of Strategic Self-Presentation’, in J. Suls (ed.), Psychological Perspectives on the Self, Vol. 1 (Erlbaum, Hillsdale, NJ), pp. 231–262.

Klein, J. and N. Dawar: 2004, ‘Corporate Social Responsibility and Consumers’ Attributions and Brand Evaluations in a Product-Harm Crisis’,International Journal of Research in Marketing21, 203–217.

Lafferty, B. A. and R. E. Goldsmith: 1999, ‘Corporate Credibility’s Role in Consumers Attitudes and Purchase Intentions When a High Versus Low Cred- ibility Endorser is Used in the Ad’,Journal of Business Research44(2), 109–116. 282Joe¨ lle Vanhamme and Bas Grobben Maignan, I., O. C. Ferrell and G. T. Hult: 1999, ‘Corporate Citizenship: Cultural Antecedents and Business Bene ts’,Journal of Academy of Marketing Science27(4), 455–469.

McGuire, W. J: 1985, ‘Attitudes and Attitude Change’, in G. Lindsey and E. Aronson (eds.),Handbook of Social Psychology, 3rd Edition, Vol. 2 (Random House, New York), pp. 233–246.

Mitroff, I. I.: 2001,Managing Crises Before They Happen:

What Every Executive and Manager Needs to Know About Crisis Management(Amacom, New York).

Mohr, L. A., D. J. Webb and K. E. Harris: 2001, ‘Do Consumers Expect Companies to be Socially Responsible? The Impact of Corporate Social Responsibility on Buying Behavior’,Journal of Con- sumer Affairs35(1), 45–72.

MORI CSR Research: 2002, available athttp:// www.mori.com/csr/.

Neilsen, E. H. and M. V. H. Rao: 1987, ‘The Strategy- Legitimacy Nexus: A Thick Description’,The Academy of Management Review12(3), 523–533.

Petty, R. E. and J. T. Ciacoppo: 1986, ‘The Elaboration Likelihood Model of Persuasion’, in H. A. M. Wilke (ed.),Orie¨ ntatie in de Sociale Psychologie, Vol. 6 (Bohn, Sta eu, van Loghum, Houten), pp. 135–140.

Obermiller, C. and E. R. Spangenberg: 1998, ‘Devel- opment of a Scale to Measure Consumer Skepticism Toward Advertising’,Journal of Consumer Psychology 72(2), 159–186.

Riezebos, H. J.: 1996,Merkenmanagement(Wolters- Noordhof, Groningen).

Rossiter, J. R. and L. Percy: 1998,Advertising Commu- nications and Promotion Management, 2nd Edition (McGraw-Hill, Singapore).

Sen, S. and C. B. Bhattacharya: 2001, ‘Does Doing Good Always Lead to Doing Better? Consumer Reactions toCorporate Social Responsibility’,Journal of Marketing Research38(2), 225–243.

Sturges, D. L.: 1994, ‘Communicating Through Crisis: A Strategy for Organizational Survival’,Management Communication Quarterly7(3), 297–316.

Suchman, M. C.: 1995, ‘Managing Legitimacy: Strategic and Institutional Approaches’,Academy of Management Review20(July), 571–610.

Swaen, V. and Vanhamme, J.: 2004, ‘See How Good We are: The Dangers of Using Corporate Social Activities in Communication Campaigns’, in B. Kahn and M. F.

Luce (eds.),Advances in Consumer Research, Vol. 31 (Association for Consumer Research, Valdosta, GA), pp. 302–303.

Varadarajan, P. R. and A. Menon: 1988, ‘Cause-Related Marketing: A Coalignment of Marketing Strategy and Corporate Philanthropy’,Journal of Marketing52(3), 58–74.

Vonk, R.: 1998, ‘Impression Formation and Impression Management: Motives, Traits, and Likeability Inferred from Self-Promoting and Self-Depreciating Behavior’, Social Cognition17(4), 390–412.

Vonk, R.: 1999, ‘The Slime Effect: Suspicion and Dislike of Likeable Behavior Towards Superiors’,Journal of Personality and Social Psychology74(4), 849–864.

Webb, D. J. and L. A. Mohr: 1998, ‘A Typology of Consumer Responses to Cause-Related Marketing:

From Skeptics to Socially Concerned’,Journal of Public Policy & Marketing17(2), 226–238. Department of Marketing, Rotterdam School of Management, Erasmus University Rotterdam, PO Box 1738, 3000 DR, Rotterdam, The Netherlands E-mail: [email protected] The Effectiveness of CSR History in Countering Negative Publicity283