• No results found

Corporate social responsibility in the fashion industry : do we really care about what we wear?

N/A
N/A
Protected

Academic year: 2021

Share "Corporate social responsibility in the fashion industry : do we really care about what we wear?"

Copied!
54
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Master’s Thesis

Corporate social responsibility in the fashion industry:

Do we really care about what we wear?

University of Amsterdam Graduate School of Communication Master’s program Corporate Communication

Supervisor dr. W.J.L. Elving Student Lisa Johanna Hobus (6056652)

(2)

Corporate social responsibility in the fashion industry:

Do we really care about what we wear?

Abstract

Recently, labor conditions of workers in the fashion industry receive a lot of attention. Media and activist groups report on incidents in factories in Bangladesh, while fashion retailers increasingly focus on communicating their socially and environmentally responsible business practices. The triple bottom line of the economic, social, and environmental performance is generally referred to as Corporate Social Responsibility (CSR). However, research on CSR in the fashion industry is limited. Additionally, consumers’ perceptions of publicity regarding corporate social behavior and ethical fashion brands are currently under-addressed. Therefore, the present study seeks to investigate how consumers respond to incidents of unethical corporate behavior in the media and how they evaluate fashion brands’ CSR communication. The results of an experiment indicate that there is no significant effect of fashion brands’ CSR positioning on consumers’ evaluations and behavioral intentions. However, negative publicity on CSR practices of brands significantly impacts consumer outcomes in a negative way. Furthermore, the effect of publicity on consumers is mediated by skeptical responses. Finally, a significant but weak interaction effect is found of CSR information of brands and the media on brand attitude. These findings show that fashion retailers should provide CSR information in more transparent and proactive ways in order to maintain a positive image. If fashion brands fail to do so, consumers will predominantly be informed by media and activist group, which can have negative consequences for brands’ reputation by provoking skeptical responses.

(3)

Introduction

April 24th, 2013, the eight-story Rana Plaza building in Bangladesh, housing three textile factories that manufacture for fashion labels such as Benneton, Mango and Primark, collapsed. More than a thousand people were killed and over two thousand workers, mainly women, were injured (Burke, 2013; Yardley, 2014). Five months earlier, a fire in the Tarzeen factory in the same district of Bangladesh caused the death of 114 factory workers who produced clothing for, among others, Walmart and C&A (Burke, 2013; Labour Behind the Label, 2014). These incidents drew large attention to the working conditions in factories in low-wage countries that manufacture for Western fashion brands. Pressured by activist campaigns such as the Clean Clothes Campaign (2012), the fashion industry responded to the Rana Plaza disaster by grouping themselves in a 40 million dollars compensation fund for the victims (Yardley, 2014). In addition, seventy fashion brands signed a legally binding safety agreement that is to ensure independent factory inspections and that holds the brands financially responsible for the upkeep of safety standards (Neville, 2013).

The Irish international fashion retailer Primark was among the brands that admitted responsibility and donated to the fund (Yardley, 2014). While Primark focuses on communicating corporate social responsibility (CSR) on its corporate website (Primark, 2013), the fashion retailer has received a lot of negative publicity in which it is accused of unethical business practices (e.g. Dhariwal, 2009; Hickman, 2009; Jones, Temperley & Lima, 2009; McDougall, 2009). Surprisingly, the Guardian reported in July 2013 that, despite the Bangladesh factory disaster, Primark’s sales had increased with twenty percent (Neville, 2013). The chief financial officer of Associated British Food, the parent business of Primark, explained: "The biggest effect on sales has been the weather. We have not seen any real change from the tragedy" (Neville, 2013). This raises questions on how consumers perceive these media and corporate messages about labor conditions in the fashion industry and how it impacts their attitude and behavior towards brands.

Over the last decade, there has been a rapid expansion of small-scale ethical fashion brands and sustainable sub-labels of larger fashion brands (Islam & Deegan, 2010; Ethical

(4)

Fashion Forum, 2011; Goworek, Fisher, Cooper, Woodward & Hiller, 2012; Pedersen & Gwozdz, 2014). Furthermore, companies have increasingly reported on their social and environmental contributions as a response to stakeholder pressures (Campbell, 2007; Matten & Moon, 2008; Bhattacharya, Korschun & Sen, 2009; Capriotti, 2009; Bitzer & Glasbergen, 2010). These efforts are meant to show that companies are taking on their corporate social responsibility (CSR). CSR can be defined as “context-specific organizational actions and policies that take into account stakeholders’ expectations and the triple bottom line of economic, social, and environmental performance” (Aguinis, 2011, p. 855).

Consumers are one of the most important stakeholder groups, as they are vital for companies’ long-term survival (Clarkson, 1995; Campbell, 2007; Du, Bhattacharya & Sen, 2010). Both companies and media are responsible for a greater involvement of consumers in CSR (Sahlin-Andersson, 2006). Fashion brands provide in-store information about the ethical standards and origin of their products (Dickson, 2001; Goworek et al., 2012) and carefully provide CSR information (DeTienne & Lewis, 2005; Islam & Deegan, 2010). The press and various activist groups are important watchdogs that inform the public and governments about the business conduct of companies, focusing especially on supply chain and labor issues (e.g. De Tienne & Lewis, 2005; McMullen & Maher, 2009; Islam & Deegan, 2010; Alam, Blanch & Smith, 2011; Goworek et al., 2012; Pedersen & Gwozdz, 2014).

It is important to determine consumers’ perception of CSR and their motivations to purchase ethically manufactured products (Bhardway & Fairhurst, 2010; Goworek et al., 2012). However, research on consumer perceptions has only emerged fairly recently (Öberseder, Schlegelmilch & Murphy, 2013) and studies about CSR in the fashion industry remain scarce (Jones et al., 2009; Islam & Deegan, 2010; Arrigo, 2013; Johnson, Lee, Choi,

Mun & Yoo, 2013; Pedersen & Gwozdz, 2014). Much of the published research indicates an

ambiguous relationship between CSR and consumer reactions. On the one hand, there is a market of ethically minded consumers that forces brands to be more socially responsible (Shen, Wang, Lo & Shum, 2012), while on the other hand studies indicate that style, quality and price are still of primary consideration (Dickson, 2001; Maignan, 2001; Joergens, 2006;

(5)

Ethical Fashion Forum, 2011; Goworek et al., 2012). Consumers genuinely seem to care about ethical labor standards, but they do not actually purchase ethical fashion labels. This is referred to as the attitude-behavior gap (Boulstridge & Carrigan, 2000).

No conclusive evidence has been found on what causes the attitude-behavior gap. It might be caused by the lack of brand information on CSR that consumers currently perceive, even when they want to purchase ethical fashion (Joergens, 2006; Ethical Fashion Forum, 2011). Furthermore, research on the impact of CSR information that is communicated by other, non-corporate sources remains scarce (Groza, Pronschinske & Walker, 2011). Overall, mass media focus on economic instead of social aspects of businesses (Capriotti, 2009). Additionally, current studies investigating the impact of media coverage of CSR were mainly focused on negative publicity (DeTienne & Lewis, 2005; Islam & Deegan, 2010; Goworek et al., 2012; Pedersen & Gwozdz, 2014).

In order to address this gap in the literature, this study focuses on testing the impact of CSR information about fashion brands on consumers. Both corporate communication and publicity on the labor conditions of fashion brands are included here as a source of CSR information. The following research question is central in this study:

What is the effect of publicity on fashion brands’ labor conditions on consumers and to what extent does CSR communication of fashion brands influence consumers?

Theoretical framework

CSR is a global trend that involves several actors, including businesses, states, international organizations and civil society organizations (Sahlin-Andersson, 2006). Primarily, it refers to the policies and actions of organizations (Aguinis & Glavas, 2012). Companies deal with economic and legal responsibilities to be profitable within the legal framework, while they are expected to take established social norms and activities into account (Carroll, 1999; Aguinis, 2011). However, the concept expands organizational boundaries as it is influenced by a diverse set of other actors as well, such as shareholders, suppliers, employees, consumers, governments and activist groups (Aguinis & Glavas, 2012;

(6)

Öberseder et al., 2013). From a stakeholder perspective, organizations are obliged to satisfy the often-conflicting needs of their stakeholder groups (Freeman, 1984, 1999). Therefore, it is one of the main motives of companies to engage in CSR (Sahlin-Andersson, 2006; Campbell, 2007; Aguinis & Glavas, 2012). Other motives are intrinsic motivations to contribute to society, governmental regulations or competitive advantages (Öberseder et al., 2013). In practice, companies often stay close to their core business (Aguinis & Glavas, 2012) and focus on one or a few activities in a specific area (Elving, 2013). Philanthropy, environmental responsibilities, humane employee treatment and cause-related marketing are a few examples (Ellen, Webb & Mohr, 2006).

From the 1990s onwards, CSR became more extensively covered in the fashion literature as well (Dickson & Eckman, 2006). These studies were mainly focused on consumer behavior, the production/supply chain and labor conditions (Johnson et al., 2013). Furthermore, case studies can be found on CSR management of fashion brands (e.g. Graafland, 2002; Arrigo, 2013) and social disclosure practices as a response to stakeholder and media pressures (DeTienne & Lewis, 2005; Islam & Deegan, 2010). Reports point out the voluntary base of the fashion industry’s engagement in CSR (Bhardway & Fairhurst, 2010) that is mainly the result of conformance to requirements of the institutional environment (Pedersen & Gwozdz, 2014). Due to increased media attention, stakeholders are generally more aware of CSR and hold companies accountable for the societal and environmental impact of their business practices (Wagner, Lutz & Weitz, 2009; Goworek et al., 2012; Zyglidopoulos, Georgiadis, Carroll & Siegel, 2012; Pedersen & Gwozdz, 2014). As it is assumed that businesses are more likely to act socially responsible when they know they are monitored (Campbell, 2007), companies incorporate CSR as an important element of their strategy (Bielak, Bonini & Oppenheimer, 2007) and corporate communications (Carroll & McCombs, 2003). This is also noticeable in the fashion industry, as there has been a significant increase of small-scale ethical fashion brands and sustainable sub-labels of larger fashion brands (Islam & Deegan, 2010; Goworek et al., 2012). The next sections will elaborate on CSR communication and publicity.

(7)

CSR communication

Companies provide consumers with information about their CSR activities on for example their corporate websites, through advertising and on point of purchase (Du et al., 2010). A large body of research shows that CSR communication is generally associated with positive effects on outcomes such as product attitude and a company’s reputation (Fombrun & Shanley, 1990; Dawkins, 2004; Sen, Bhattacharya & Korschun, 2006; Pomering & Dolnicar, 2009; Goworek et al., 2012). Unsurprisingly, CSR has become a strategic marketing tool for stakeholder communication (Sen et al., 2006; Nan & Heo, 2007; Lii & Lee, 2012; Walker & Kent, 2013). In order to reap the benefits of CSR, companies must take their competitive context into account (Bhattacharya & Sen, 2004). One key element is the extent to which companies position themselves as a ‘CSR brand’ relative to competitors. In this case, the positioning of a company is based on its socially responsible business practices and policies (Du, Bhattacharya & Sen, 2007). The upcoming of ethical fashion brands and sustainable sub-labels of large fashion retailers are arguably examples of a CSR brand positioning. More conventional marketing strategies focus on brand positioning based on traditional product-related values, such as excellent quality (Brown & Dacin, 1997).

However, in practice, companies are careful when communicating CSR because of possible countervailing effects. Consumer skepticism that might arise could have negative consequences for a company’s reputation (Forehand & Grier, 2003; Yoon, Gürhan-Canli & Schwarz, 2006; Becker-Olsen, Cudmore & Hill, 2006; Ellen et al., 2006). In addition, companies cannot control the outcomes of CSR communication and high expectations can unintentionally be created (Schultz & Wehmeier, 2010). Therefore, because of the companies’ hesitance in communicating CSR, recent findings indicate that consumer awareness of CSR practices of fashion brands is low; consumers perceive a lack of brand information and media coverage in order to make ethical choices (Joergens, 2006; Goworek et al., 2012). Therefore, we also examine the role of media to inform consumers about CSR practices.

(8)

CSR in the media

Besides corporate communications, consumers base their evaluations of companies on what is reported in the media (Fombrun & Shanley, 1990). Technological developments have increased the public’s access to a virtually unlimited amount of information about CSR behavior (Wagner et al., 2009). Due to globalization of amongst others fashion retailers, the complexity of the supply chain causes an increase of incidents of socially irresponsible behavior (Amaeshi, Osuji & Nnodim, 2008; Wagner et al., 2009). As news media prefer to report about bad news (Dennis & Merrill, 1996; Harcup & O'neill, 2001) this is quickly passed on to the general public. The media agenda setting theory describes media’s abilities to influence the salience of topics on the public agenda (McCombs & Shaw, 1972). Current research utilizes the theory to explain how media attention for fashion brands’ labor conditions has shaped consumers’ concern of these issues (Islam & Deegan, 2010; Pedersen & Gwozdz, 2014).

Hence, there is a large and increasing amount of CSR communicators that differ in their level of independency (company-controlled vs. third party-controlled) and credibility. The less controllable the source is, the more credible it is perceived (Du et al., 2010). Consequently, the effectiveness of CSR communication in changing recipients’ evaluations and behavior is dependent on the source of information (Yoon et al., 2006; Du et al., 2010; Groza et al., 2011; Eberle, Berens & Li, 2013; Skard & Thorbjørnsen, 2013). It generally is suggested that companies should communicate their practices through third party, non-corporate sources, even though these messages are less controllable (Simmons & Becker-Olsen, 2006; Yoon et al., 2006; Du et al., 2010). However, in the case of incidents or scandals, negative publicity can seriously damage the company’s image (Dean, 2004; Wagner et al., 2009). As current studies have mainly focused on the impact of either positive of negative publicity, we investigate the impact of both types of publicity on consumers.

(9)

Consumer outcomes

In addition to recent findings about consumer perceptions of CSR and their motivations to purchase ethically manufactured products, this study focuses on the impact of CSR information of fashion brands and media on consumers. All information that an individual holds about a company is defined as one’s corporate associations (Brown & Dacin, 1997). In response to CSR positioning, consumers construct CSR beliefs about the extent to which a brand is socially responsible (Sen et al., 2006; Du et al., 2007). They hold certain beliefs about the corporate ability (CA) that is defined as “the company’s expertise in producing and delivering high quality products and services” (Brown & Dacin, 1997; Du et al., 2007). Both types of corporate associations have an important influence on outcomes such as identification and purchase intention (Brown & Dacin, 1997; Mohr & Webb, 2005). For five selected consumer outcomes (brand attitude, brand reputation, identification, advocacy and purchase intention) the impact of knowledge of a fashion brand’s labor conditions is further elaborated on, based on existing frameworks of CSR effects (Sen et al., 2006; Du et al., 2010; Lii & Lee, 2012).

Brand attitude. Brands that integrate their social initiatives into their competitive

positioning, i.e. CSR positioning, are associated with more positive consumer beliefs (Du et al., 2007). When stakeholders become aware of a company’s CSR practice, this results in several positive outcomes (Bhattacharya & Sen, 2004), such as a positive attitude towards the company (Sen et al., 2006; Du et al., 2007; Du et al., 2010; Lii & Lee, 2012). In addition, consumers hold more positive attitudes towards companies that support a social cause (Nan & Heo, 2007).

Conversely, other studies did not find an effect of CSR (Dean, 2004) or a more positive effect of corporate ability messages on product attitude (Biehal & Sheinin, 2007). ‘Loud’ CSR communication can backfire as it creates higher stakeholder expectations of the organization as well (Morsing, Schultz & Nielsen, 2008; Schultz & Wehmeier, 2010). Here, consumers react more positively to CSR information from an independent, neutral source than from a corporate source (Yoon et al., 2006; Simmons & Becker-Olsen, 2006; Du et al.,

(10)

2010). However, one study indicated that the positive effect of publicity was only generated for high-reputation brands (Skard & Thorbjørnsen, 2013). In the case of negative publicity, consumers’ evaluations of companies and their products could be damaged (Dean, 2004; Biehal & Sheinin, 2007; Marin & Ruiz, 2007). Furthermore, negative CSR associations appear to have a stronger impact than positive ones (Maheswaran & Meyers-Levy, 1990; Brown & Dacin, 1997; Sen & Bhattacharya, 2001; Biehal & Sheinin, 2007; Du et al., 2007; Eberle et al., 2013). This is due to an increased level of message processing in the case of negative information and publicity, which is more critically scrutinized for faults than corporate advertising (Eisend & Küster, 2011), especially when brand reputation is low (Skard & Thorbjørnsen, 2013).

Brand reputation. CSR is one of the pillars of the reputation temple, as companies

with extended CSR programs generally have better reputations (Fombrun & Shanley, 1990). Corporate reputation can be defined as the collective representation of past images of an organization, which are induced through communication or past experiences and established over time (Cornelissen, 2008). Additionally, it influences how consumers perceive CSR: in the case of a good reputation, charity is generally perceived as a mutually beneficial action, whereas consumers attribute self-interested motives to companies with a bad reputation (Bae & Cameron, 2006; Elving, 2013).

The overall evaluation of a company that is presented in the media is defined as a company’s media reputation (Deephouse, 2000). Based on the media agenda setting theory, it is assumed that media coverage influences the public’s knowledge and opinion and thus the reputation of companies. Hence, when media visibility is extensive and positive, this positively influences the company’s reputation (Fombrun & Shanley, 1990). Additionally, a good CSR reputation is found to act as a buffer in the case of negative publicity (Klein & Dawar, 2004).

Identification. Another possible competitive advantage of CSR is the positive effect

on consumers’ level of identification, which can be defined as “a cognitive link between the definitions of the organization and the self” (Dutton, Dukerich & Harquail, 1994, p. 242). More

(11)

specifically, company-consumer identification can be defined as consumers’ identification with a company when it satisfies at least partly their self-definitional needs (Scott & Lane, 2000). Several studies have linked corporate engagement in CSR with higher levels of company identification (Balmer, 2001; Sen et al., 2006; Currás-Pérez, Bigné-Alcañiz & Alvarado-Herrera, 2009; Du et al., 2010; Lii & Lee, 2012) and this effect is stronger for consumers that perceive congruence between a company’s and their own character and personal values (Bhattacharya & Sen, 2003).

Additionally, one study indicated that CSR information provided by non-corporate sources on online interactive media is perceived as more credible and is related to higher levels of company identification than company-controlled messages (Eberle et al., 2013).

Advocacy. Several studies report the long-term benefits of CSR positioning on brand

advocacy behavior (Du et al., 2007; Bhattacharya et al., 2009; Du et al., 2010). Based on the organizational citizenship behavior theory, it is assumed that consumers can exhibit citizen behavior that is not directly or explicitly related to a reward system (Organ, 1988). Furthermore, the social exchange theory (Blau, 1964) describes how social exchanges between consumers and companies might result in positive gains (e.g. identity experience) when the company is positively evaluated. Hence, an existing relationship between CSR activities and in- and extra-role behavior can result in beneficial outcomes for companies (Lii & Lee, 2012). In the crisis communication literature, negative publicity is related to higher levels of advocacy behavior such as sharing information and willingness to boycott (Schultz, Utz & Göritz, 2011).

Purchase intention. Finally, the benefits of CSR programs that are related to

positive attitudes and stronger levels of identification ultimately result in higher levels of purchase intention as well (Ellen et al., 2006; Sen et al., 2006; Du et al., 2010). When CSR initiatives are in line with consumers’ beliefs about the CSR domain and they evaluate it as reinforcing rather than detracting from CA development, this positively influences their purchase intention (Mohr & Webb, 2005; Sen & Bhattacharya, 2001). However, CSR is not the most important predictor of consumers’ purchase behavior; CSR and CA attributions

(12)

have a trade-off effect on consumers’ willingness to pay for a product and this effect decreases when the price of a product increases (Dickson, 2001; Feldman & Vasquez-Parraga, 2013). Additionally, focus groups have revealed that, although consumers generally care about labor issues, these concerns are not necessarily presented in their actual purchase behavior of ethically manufactured products (Joergens, 2006; Eckhart, Belk & Devinney, 2010; Carrington, Neville & Whitwell, 2010; Goworek et al., 2012). Discrepancy between ethical beliefs and purchase behavior is referred to as the attitude-behavior gap (Boulstridge & Carrigan, 2000). In sum, although CSR generally has a positive impact, the current findings about the effect on purchase intention are inconsistent.

The mediating role of skepticism

Much of the current literature on CSR communication pays particular attention to the role of skepticism. Skepticism may arise if consumers evaluate CSR as a reputational motive and it can mediate the effect of CSR communication on consumer outcomes (Forehand & Grier, 2003; Ellen et al., 2006; Sen et al., 2006; Bae & Cameron, 2006; Becker-Olsen et al., 2006; Yoon et al., 2006; Lii & Lee, 2012; Elving, 2013; Goworek et al., 2012). Attribution theory describes the process in which individuals evaluate others’ motives and it explains how these perceived motives influence subsequent behavior (Jones & Davis, 1965; Kelley, 1973). Consumers can distinguish between motives to engage in CSR that are extrinsic and self-centered (‘increasing profits’) or intrinsic and others-centered (‘a genuine concern’) (Becker-Olsen et al., 2006; Vlachos, Tsamakos, Vrechopoulus & Avramidis, 2009; Du et al., 2010) and both types of attributions can be made at the same time (Ellen et al., 2006; Sen et al., 2006). In case consumers attribute intrinsic motivations over extrinsic motivations, this has a positive impact on the attitude towards the company (Forehand & Grier, 2003; Bae & Cameron, 2006; Becker-Olsen et al., 2006; Yoon et al., 2006; Wagner et al., 2009). On the other hand, extrinsic attributions negatively impact stakeholders’ attitude, the company’s reputation, advocacy behavior, and purchase intention (Ashfort & Gibbs, 1990; Mohr, Eroǧlu & Ellen, 1998; Yoon et al., 2006; Elving, 2013; Walker & Kent, 2013). Additionally,

(13)

self-serving motives are attributed more often when compared to CSR messages from non-corporate sources (Yoon et al., 2006; Groza et al., 2011; Eberle et al., 2013).

Based on the current literature about the impact of CSR brand positioning and publicity, the following hypotheses will be tested in this study (see Figure 1):

H1: A CSR brand positioning has a positive effect on consumer outcomes.

H2: Negative publicity has a negative effect on consumer outcomes.

H3: Intrinsic and extrinsic attributions of skepticism mediate the effect of brand

positioning on consumer outcomes.

H4: Intrinsic and extrinsic attributions of skepticism mediate the effect of publicity on

consumer outcomes.

Figure 1. Conceptual model of the first four hypotheses about the direct effects of CSR

information, operationalized as brand positioning and publicity, and the mediating effect of skepticism on five types of consumer outcomes.

There currently exists little research on the interaction effect of CSR communication and publicity on consumers. Even though it is commonly assumed that CSR positioning generates positive brand outcomes, inconsistency between what a company communicates about its CSR activities and what is reported in the media evokes perceptions of corporate hypocrisy (Wagner et al., 2009). Additionally, discrepancy between stated and perceived motives can seriously harm a company (Forehand & Grier, 2003). Although consumers are less influenced by negative publicity when a brand has a good CSR reputation (Du et al.,

(14)

2007), it also creates higher expectations of the company’s ethical standards (Dean, 2004; Morsing et al., 2008; Schultz & Wehmeier, 2010). Based on these findings and the common belief that honesty and transparency is fundamental in CSR communication (Öberseder et al., 2013), we formulate the final, fifth hypothesis:

H5: The difference in consumer outcomes that results from negative compared to

positive publicity will be larger for a CSR positioning than for a CA positioning.

Control variables

A selection of control variables will be included in this study, as previous research reported that these variables could influence the investigated effects. Brand familiarity and brand choice are taken into account, as brand familiarity influences someone’s attitude towards a company (Alexandris, Douka, Bakaloumi & Tsasousi, 2008). Furthermore, the impact of CSR information is dependent on the level of involvement in the CSR domain that reflects

consumers’ personal values and opinions (Grau & Folse, 2007; Pickett-Baker & Ozaki, 2008; Öberseder et al., 2013). Involvement can be defined as “a personal connection or bridging experience for an individual” (Grau & Folse, 2007, p. 20). It also influences how positive and negative information is processed (Chaiken, 1980; Petty, Cacioppo & Schumann, 1983). Consumers with low levels of involvement elaborate in simple inferences and associate more with positive than negative cues (Grau & Folse, 2007), whereas highly involved consumers elaborate more on negative information, which they process in a more detailed way

(Maheswaran & Meyers-Levy, 1990). Hence, the direct effects of brand positioning and publicity will be controlled for the moderating effect of involvement.

Method

Sample

Participants were recruited with an online advertisement or personal invitation to participate in the research using convenience sampling. To make sure that all information was thoroughly read and watched, only participants who took longer than three minutes for

(15)

the experiment were included. Also, data from participants younger than eighteen years old were deleted, because of the lack of parental consent. Of the 330 participants that started the experiment, the final sample consisted of 164 participants of which 39.4% were male with an average age of 26.1 years (SD = 8.0). Most of the subjects completed a university bachelor or master degree (80.2%) and had a yearly gross income of less than € 20,000 (64.2%).

Experimental design

A 2 x 2 between-subjects experiment was conducted in which the effects of brand positioning (CSR versus CA) and publicity (negative versus positive) on consumer outcomes were tested. The manipulations shown consisted of brand information about a fashion brand followed by a compilation of a news item about the labor conditions that were associated with the brand. The online survey tool that was used randomly assigned the participants to one of four conditions (see Table 1).

Table 1. Experimental study design with four conditions (N = 164).

Brand positioning Publicity Negative Positive CSR positioning n = 36 n = 44 CA positioning n = 38 n = 46 Procedure

Participants were recruited on social media by asking them to participate by clicking on an online link. First, they were provided with some general information about the research and an informed consent form that they had to read carefully before continuing the research. Then, participants were randomly directed to a page that introduced a fashion brand. This brand information was focused either on the social responsibility initiatives or the style and quality of the brand. This was followed by a short commercial, after which a pre-test was conducted to check for preconceptions in participants’ familiarity with the fashion brand. Then, participants continued to the next page that shortly introduced a news item about the

(16)

labor conditions the fashion brand was recently associated with, which was either positive or negative publicity. This was followed by questions about participants’ brand attitude, attributions of skepticism, identification, involvement, advocacy, brand reputation and their purchase intention. Participants first evaluated the fashion brand before they were asked to make attributions about the brands motives, based on the literature about causal attribution processes (Forehand & Grier, 2003). The last pages included a manipulation check of the credibility of the information and questions about participants’ demographics. Finally, participants were thanked for their participation and informed that the brand information and news items were fictitious. They could leave their e-mail addresses when they were interested in a summary of the research findings.

Ethics

Several ethical standards were taken into account in the experiment. Firstly, participants had to read an informed consent, in which anonymity and truly voluntary participation was stressed. A realistic account of the average time the survey would take was indicated as well. In case of questions or complaints, participants were provided with contact information of the ethical committee and research supervisor. After completion of the survey, participants were made aware that the manipulations were only constructed for research purposes.

Independent variables

Brand positioning. The brand positioning was either focus on social responsibility

initiatives (i.e. CSR positioning) or characteristics such as style, quality and price (i.e. CA positioning) (Du et al., 2007). Two fashion brands were included, which were COS and American Apparel. COS, part of the H&M Group, a fashion retailer that is associated with poor labor conditions in the past (DeTienne & Lewis, 2005), is a brand that focuses its communication on style and quality. On contrary, American Apparel strongly communicates its CSR practices on its corporate website and overall has a good socially responsible

(17)

reputation (Joergens, 2006). Based on the ‘About’ section of the corporate websites of the brands, two types of brand positioning (CSR vs. CA) were constructed for each fashion brand by shortening and slightly adjusting this information. This brand information was followed by a compilation of an existing commercial video of the brand.

Publicity. The second manipulation consisted of either negative or positive publicity

on the labor conditions of the fashion brand. Two compilations were constructed of ABC news items about a) the poor labor conditions in the Rana Plaza factory in Bangladesh that were associated with COS and b) the ethical labor conditions in the American factory of American Apparel. COS was associated with negative publicity in both conditions (CSR and CA positioning), whereas American Apparel received positive publicity in both conditions. For an example of the manipulated material please consult Appendix 2.

Dependent variables

Brand attitude. Brand attitude was measured with “How do you evaluate the brand?” by three semantic differentials like “extremely unlikable – extremely likable” on a 7-point scale (Nan & Heo, 2007; Forehand & Grier, 2003). A principal component analysis (PCA) yielded one factor (Cronbach’s α = .95). A new scale was constructed by averaging all items. A low score indicates a negative attitude, a high score a positive attitude.

Brand reputation. Brand reputation was measured by four items on a 7-point Likert

scale (strongly disagree – strongly agree) of Till and Busler (2000). An example of an item is “I have good feelings about this brand”. A PCA confirmed one factor (Cronbach’s α = .93). A low score indicates a negative reputation and a high score a positive reputation.

Identification. Identification refers to the extent to which participants identify with the fashion brand based on Du et al. (2007). It was measured on a 7-point Likert scale (strongly disagree – strongly agree) by the item “My sense of this brand matches my sense of who I am”. A low score indicates a low level of identification and a high score a high level of identification.

(18)

Advocacy. Advocacy refers to positive behavioral intentions towards the fashion

brand. It was measured by three items of Du et al. (2007) on a 7-point Likert scale (strongly disagree – strongly agree). An example of an item is “I would like to try new products introduced under this brand name”. Based on a PCA a reliable scale was constructed (Cronbach’s α = .85). A low score indicates a low level of advocacy and a high score a high level of advocacy.

Purchase intention. Purchase intention refers to the likelihood of participants to

purchase clothing from the fashion brand based on Ponzi, Fombrun and Gardberg (2011). It was measured by three semantic differentials like “extremely unlikely – extremely likely” on a 7-point scale. A PCA yielded one underlying factor (Cronbach’s α = .97). A low score indicates a low purchase intention, whereas a high score indicates a high purchase intention. Skepticism. Participants’ attributions about the fashion brand’s motives to provide ethical labor conditions for factory workers are referred to as skepticism. Based on Du et al., (2007) two types of attributions (intrinsic and extrinsic) measured skepticism. Both types were measured by two items on a 7-point Likert scale. These items were reformulated in order to be applicable on the CSR topic of this study. Examples of items are “This brand has genuine concern for the labor conditions of workers” and “This brand tries to make a good image of the company by providing good labor conditions”. As the two types of attributions are theoretically different in nature, two separate scales were constructed for intrinsic (Cronbach’s α = .96) and extrinsic attributions (Cronbach’s α = .82). A high score on intrinsic attributions indicates a low level of skepticism, whereas a high score on extrinsic attributions indicates a high level of skepticism.

Control variables

Brand familiarity. Brand familiarity refers to how familiar participants are with the

fashion brand based on Sen et al. (2006). It was measured with “How familiar are you with the brand?” on a 7-point scale (not at all familiar - very familiar). A low score indicates a low level of familiarity and a high score a high level of familiarity.

(19)

Brand choice. Brand choice refers to any recent purchases of participants from the

brand (Du et al., 2007). It was measured with “Did you purchase the brand recently?” on a dichotomous scale (yes, no).

Involvement. Involvement refers to the degree to which the participants perceive

labor conditions as personally relevant based on Maheswaran and Meyers-Levy (1990). It was measured with the statement “Good labor conditions for workers are…” followed by five semantic differentials like “extremely unimportant - extremely important”. All items were measured on a 7-point scale and combined in a reliable new scale (Cronbach’s α = .91).

Manipulation check

A manipulation check was included in order to test the perceived credibility of the manipulations. Participants indicated to what degree they found the CSR information about the brand “extremely unbelievable - extremely believable” and “extremely incredible - extremely credible” on a 7-point scale based on Yoon et al. (2006). A low score indicates low credibility, whereas a high score indicates high credibility of the manipulations. When combining the two items in one scale, this yielded a reliable measurement (Cronbach’s α =.88).

Data analysis

The following statistical tests were used to analyze the data. In order to test the credibility of the manipulations, one-way analyses of variance (ANOVA) were carried out with the independent variables brand positioning and publicity and the dependent variable credibility. Furthermore, one-way ANOVA’s were conducted to assess the effect of brand positioning (hypothesis 1) and publicity (hypothesis 2) on consumer outcomes. In order to investigate the mediating role of skepticism (hypotheses 3 and 4), mediation analyses were carried out. As proposed by Baron & Kenny (1986), prerequisites for mediation analysis are significant direct effects of the independent variable on both the mediator and the dependent variable (Step 1). When these are significant, multiple regression analyses can be conducted

(20)

with the independent variable and mediator as predictors of the dependent variable (Step 2). When the regression coefficient of the independent variable changes after controlling for the mediator, there is a mediation effect. Finally, a Sobel’s Z test was conducted to indicate if the mediation was significant. In order to conduct linear regression, continuous dummy variables were constructed of brand positioning and publicity, as these originally were categorical variables. In order to test hypothesis 5 concerning the interaction effect of brand positioning and publicity, two-way ANOVA’s were carried out. Finally, the control variables are investigated with hierarchical multiple regression models that included brand positioning, publicity, brand familiarity and brand choice. Moderation analyses were carried out with the independent variables brand positioning and publicity, the standardized variable involvement and the standardized interaction term of brand positioning or publicity and involvement.

Results

Manipulation check

No significant differences are found for participants’ evaluations of the credibility of brand positioning (F (1, 160) = .28, p = ns). Participants do not evaluate a CSR positioning as more credible (M = 4.04, SD = .88) than participants who are exposed to a CA positioning (M = 3.97, SD = .90). Furthermore, a non-significant difference of credibility is indicated for the two conditions of publicity (F (1, 160) = .81, p = ns). Participants who are exposed to negative publicity do not experience this information as more credible (M = 4.08, SD = .88) than participants who are exposed to positive publicity (M = 3.95, SD = .90). Hence, it can be concluded that there are no significant differences in the experienced credibility of the CSR information between the four conditions, so the manipulations can be used to test the hypotheses.

Testing the hypotheses

Firstly, hypothesis 1 concerning the positive effect of CSR positioning on consumer outcomes was tested. A non-significant effect is found of brand positioning on brand attitude,

(21)

brand reputation, identification, advocacy, and purchase intention (see Figure 2 and Appendix 1, Table 1). Hence, there is no significant difference between participants that are exposed to a CSR or a CA positioning. Hypothesis 1 is therefore rejected. In hypothesis 2 it was assumed that negative publicity has a negative effect on consumer outcomes. The analyses yield significant but weak effects of publicity on brand attitude, brand reputation, identification, advocacy, and purchase intention (see Figure 3 and Appendix 1, Table 1). Hence, participants who are exposed to negative publicity score significantly lower on all consumer outcomes than participants that are exposed to positive publicity. Therefore, hypothesis 2 is accepted.

Figure 2. Mean scores of the effect of brand positioning (CSR vs. CA) on brand attitude,

(22)

Note. *p < .05, ** p < .01, *** p < .001

Figure 3. Mean scores of the effect of publicity (negative vs. positive) on brand attitude,

brand reputation, identification, advocacy and purchase intention.

According to hypothesis 3, intrinsic and extrinsic attributions of skepticism mediate the effect of brand positioning on consumer outcomes. However, as the results of hypothesis 1 already indicated, brand positioning does not yield significant direct effects on all consumer outcomes (see Appendix 1, Table 1 and 3). In addition, non-significant effects are found of brand positioning on intrinsic and extrinsic attributions of skepticism (see Appendix 1, Table 3, Model 1 and 3). Hence, the effect of brand positioning on consumer outcomes is not mediated by skepticism. Therefore, hypothesis 3 is rejected.

According to hypothesis 4, intrinsic and extrinsic attributions of skepticism mediate the effect of publicity on consumer outcomes. As the results of hypothesis 2 already indicated, publicity has a significant effect on all consumer outcomes (see Appendix 1, Table 1 and 3). Furthermore, publicity has a significant effect on intrinsic (F (1, 161) = 100.49, p < .001) and extrinsic attributions of skepticism (F (1, 161) = 82.47, p < .001). However, the

(23)

strength of this prediction is moderate as 38 percent of the variation of intrinsic (R2 = .38) and 34 percent of the variation of extrinsic attributions (R2 = .34) is predicted by publicity. In case of negative publicity, intrinsic attributions decrease by 1.82 (b = 1.82, p < .001) and extrinsic attributions by 1.62 (b = 1.62, p < .001) (see Appendix 1, Table 3, Model 2 and 4).

Next, mediation analyses are carried out, which yield significant effects of publicity and intrinsic attributions of skepticism on brand attitude (F (2, 160) = 62.36, p < .001), brand reputation (F (2, 160) = 86.93, p < .001), identification (F (2, 160) = 31.85, p < .001), advocacy (F (2, 160) = 30.61, p < .001), and purchase intention (F (2, 160) = 21.54, p < .001). The strength of this prediction is moderate for brand attitude and brand reputation, as 44 percent of the variation of brand attitude (R2 = .44) and 52 percent of brand reputation (R2 = .52) is predicted by publicity and intrinsic attributions. The prediction is weak for identification (R2 = .29), advocacy (R2 = .28) and purchase intention (R2 = .21) (see Appendix 1, Table 4, Model 1-5). Ultimately, when controlling for intrinsic attributions, the effect of publicity on brand attitude remains significant, indicating partial mediation. The effect of publicity on brand reputation, identification, advocacy and purchase intention is no longer significant, indicating full mediation (see Table 2).

Furthermore, the effects of publicity and extrinsic attributions of skepticism are significant for brand attitude (F (2, 160) = 51.84, p < .001), brand reputation (F (2, 160) = 53.91, p < .001), identification (F (2, 160) = 25.49, p < .001), advocacy (F (2, 160) = 20.11, p < .001), and purchase intention (F (2, 160) = 17.34, p < .001). However, the strength of this prediction is moderate, as 39 percent of the variation in brand attitude (R2 = .39) and 40 percent of brand reputation (R2 = .40) is predicted by publicity and extrinsic attributions. The strength of the prediction is weak for identification (R2 = .24), advocacy (R2 = .20) and purchase intention (R2 = .18) (see Appendix 1, Table 4, Model 6-10). When controlling for extrinsic attributions, the effect of publicity on brand attitude remains significant, indicating partial mediation. The effect of publicity on brand reputation, identification, advocacy and purchase intention is no longer significant, indicating full mediation (see Table 2). Hence,

(24)

intrinsic and extrinsic attributions of skepticism are found to mediate the effect of publicity on consumer outcomes. Therefore, hypothesis 4 is accepted.

Table 2. Mediation analysis: unstandardized regression coefficients of publicity, controlled for

intrinsic and extrinsic attributions of skepticism and Sobel’s Z value of the significance of the mediation effect. Brand attitude N = 163 Brand reputation N = 163 Identification N = 163 Advocacy N = 163 Purchase intention N = 163 Publicity 1.41*** .99*** 93*** .56** .97*** Publicity

Controlled for intrinsic attributions

.62** -.21 .00 -.40 .44

Sobel’s Z 5.40*** 7.41*** 5.32*** 5.82*** 3.36***

Publicity

Controlled for extrinsic attributions

.82*** .10 .20 -.15 . 44

Sobel’s Z 4.55*** 6.04*** 4.55*** 4.72*** 3.31***

*p < .05, ** p < .01, *** p < .001

Regarding hypothesis 5 concerning the interaction effect of brand positioning (CSR vs. CA) and publicity (negative vs. positive), a significant but very weak two-way interaction effect is found on brand attitude (F (1, 160) = 5.28, p < .05, η2 = .02). Consumers’ brand attitude are significantly more negative for a CSR positioning than for a CA positioning in the case of negative publicity, as opposed to a significantly more positive brand attitude for a CSR positioning than for a CA positioning in the case of positive publicity (see Figure 4 and Appendix 1, Table 1).

Two-way interaction effects on other consumer outcomes are non-significant (see Appendix 1, Table 1). In case of a CSR positioning, consumers’ brand reputation, identification, advocacy and purchase intention are not significantly more negative after exposure to negative publicity and not significantly more positive after exposure to positive publicity, when compared to a CA positioning. In conclusion, hypothesis 5 is accepted for brand attitude, but rejected for brand reputation, identification, advocacy and purchase intention.

(25)

Figure 4. Mean score of brand attitude for brand positioning (CSR vs. CA) and publicity

(negative vs. positive).

Control variables

Several control variables appear to influence the consumer outcomes, of which only the significant effects are presented here. Hierarchical multiple regression models of brand positioning, brand familiarity and brand choice are significant for brand attitude (F (3, 157) = 4.98, p < .01), brand reputation (F (3, 156) = 5.63, p < .001), advocacy (F (3, 159) = 7.24, p < .001) and purchase intention (F (3, 156) = 9.02, p < .001). However, these models (very) weakly predict the variance in consumer outcomes (see Appendix 1, Table 2). The associations between brand familiarity and consumer outcomes are significant for brand attitude (b = .16, p < .01), brand reputation (b = .15, p < .01), advocacy (b = .13, p < .01), and purchase intention (b = .17, p < .01). Brand choice is significantly associated with advocacy (b = .66, p < .05) and purchase intention (b = .75, p < .05). Hence, participants who are familiar with the brand score slightly higher on brand attitude, brand reputation, advocacy and purchase intention. Participants that recently purchased products of the brand have slightly higher advocacy and purchase intentions.

Hierarchical multiple regression models of publicity, brand familiarity and brand choice are significant for brand attitude (F (3, 157) = 23.85, p < .001), brand reputation (F (3,

(26)

156) = 13.23, p < .001), identification (F (3, 156) = 7.32, p < .001), advocacy (F (3, 156) = 8.58, p < .001) and purchase intention (F (3, 156) = 13.92, p < .001). The regression models moderately predict the variance in brand attitude and only weakly predict the other consumer outcomes (see Appendix 1, Table 2). After publicity is included in the model the strength of the prediction changes for brand attitude (23%), brand reputation (11%), identification (11%), advocacy (2%), and purchase intention (7%). Now, only brand choice is significantly associated with advocacy (b = .71, p < .05) and purchase intention (b = .85, p < .01). Hence, participants that recently purchased products of the brand score significantly higher on advocacy and purchase intention.

Finally, the impact of involvement in labor conditions of fashion brands on consumer outcomes was tested. This yields significant effects on brand attitude (F (3, 159) = 25.53, p < .001), brand reputation (F (3, 159) = 10.22, p < .001), identification (F (3, 159) = 6.51, p < .001), advocacy (F (3, 159) = 4.33, p < .01), and purchase intention (F (3, 159) = 11.01, p < .001). However, significant associations are only found between the interaction term and brand attitude, advocacy and purchase intention and the strength of this prediction is moderate for brand attitude and (very) weak for advocacy and purchase intention (see Table 3). Hence, participants with high levels of involvement score slightly lower on brand attitude, advocacy and purchase intention after exposure to publicity than low-involved participants. For more details on the findings of the control variables please consult Appendix 1, Table 2.

Table 3. Moderation analyses: regression models of publicity, involvement and the

interaction effect of publicity and involvement on consumer outcomes.

Brand attitude

Brand reputation

Identification Advocacy Purchase

intention R2 = .33 N = 163 R2 = .16 N = 163 R2 = .11 N = 163 R2 = .08 N = 163 R2 = .17 N = 163 Constant 6.44*** 5.75*** 5.17*** 5.05*** 5.48*** Publicity -1.42*** -.99*** -.93*** -.56** -.97*** ZInvolvement .73** .30 -.06 .62* 1.00** PublicityxZInvolvement -.44* -.19 .04 -.42* -.55** *p < .05, ** p < .01, *** p < .001

(27)

Discussion

The first aim of this study was to determine the influence of CSR communication of fashion brands on consumers. Contrary to our expectations, this study did not find a significant positive effect of a brand positioning focused on ethical working conditions of factory workers on consumer outcomes. This is not in line with the large body of literature that generally stresses the benefits of a CSR message strategy on brand attitude and reputation (e.g. Fombrun & Shanley, 1990; Sen et al., 2006; Pomering & Dolnicar, 2009; Du et al., 2010). However, this finding does not necessarily mean that consumers do not care about labor conditions. On the contrary, we suggest that they already expect fashion brands to manufacture in an ethical and proper way, which might explain why an emphasis on CSR in corporate communications did not positively impact consumer outcomes. CSR is then perceived as a baseline condition and not as a unique selling point. In sum, although both ethical and mainstream fashion brands have recently been focusing on communicating their CSR practices (Islam & Deegan, 2010; Goworek et al., 2012; Pedersen & Gwozdz, 2014), the results of this study illustrate that the socially responsible image that brands wish to portray is not necessarily perceived as a value proposition by consumers.

The second aim of this study was to reveal how publicity on CSR influences consumers. In line with previous studies (Dean, 2004; Capriotti, 2009; Wagner et al., 2009; Zyglidopoulos et al., 2012), publicity on companies is found to shape consumer evaluations. The current study indicated a significant but weak negative effect of negative publicity on consumer outcomes. This finding supports the impact of publicity specifically for labor conditions in the fashion industry. Whereas consumers demand transparency, fashion brands currently provide limited information about CSR due to potential image and reputational risks (Morsing et al., 2008; Schultz & Wehmeier, 2010). Hence, when CSR messages are communicated, only positive aspects will be mentioned (DeTienne & Lewis, 2005). On the other hand, news media primarily seize on incidents to report about companies’ CSR practices in order to create news value. Consumers depend on both corporate and non-corporate sources regarding CSR information for their evaluations and

(28)

behavioral intentions. When fashion brands fail to provide clear statements about their ethical standards, non-corporate sources are of importance to provide CSR information, as they are often perceived as unbiased and trustworthy (Dawkins, 2004; Simmons & Becker-Olsen, 2006; Du et al., 2010; Skard & Thorbjørnsen, 2013). However, we noticed that highly involved consumers were less influenced by publicity, as they more critically process this information (Maheswaran & Meyers-Levy, 1990). In conclusion, the significant effect of publicity that was found in this study supports the current literature that stresses the important role of non-corporate sources in shaping consumer evaluations.

In the third and fourth hypothesis, a mediating role of skepticism was assumed. Contrary to earlier findings, the current study did not find evidence of skepticism as a mediator of the effect of brand positioning on consumer outcomes. Understandably, companies try to avoid skeptical consumer responses as they carefully present CSR as a value proposition. Although the CSR positioning used in this study was unsuccessful in emphasizing the added CSR value (hypothesis 1), skeptical responses were not elicited either. Fashion brands are expected to meet ethical labor conditions, which cause consumers not to think skeptically. Furthermore, consumers are not likely to make any negative attributions when they recently purchased a brand’s product, as CSR is closely related to the self-identity (Bhattacharya et al., 2009). This was also indicated by the control check that revealed that brand familiarity and brand choice had a small but significant influence on the investigated consumer outcomes.

However, the results did confirm a mediating role of skepticism on consumer outcomes when news media report about the CSR practices of brands. Generally, consumers perceive these non-corporate sources as more credible and trustworthy (Yoon et al., 2006; Du et al., 2010) and when news items do not correspond with companies’ propositions, consumers will question their motives for CSR. Although this inconsistency can be perceived as a disadvantage for companies, the findings of this study indicate that skeptical responses have a limited impact on behavioral intentions. Intrinsic and extrinsic

(29)

attributions of skepticism mainly predicted consumers’ brand attitude and reputation, which is in line with the attitude-behavior gap (Boulstridge & Carrigan, 2000).

Finally, the fifth hypothesis about the interaction effect of brand positioning and publicity was only confirmed for the effect on brand attitude. Consumers’ brand attitude is significantly more positive when CSR information provided by fashion brands and the media is consistent, but this effect was not very strong. Inconsistency lowered brand attitude outcomes, as opposed to fashion brands that receive negative publicity and do not externally communicate any CSR propositions. As the control test indicated, familiarity with the brand may explain how consistent messages positively reinforce existing brand associations. However, in line with other studies (Morsing et al., 2008; Schultz & Wehmeier, 2010; Elving, 2013), this study confirms the idea that companies should be careful with creating expectations with CSR communication, as it can have disastrous effects when different sources reveal that the actual CSR behavior is inconsistent with what is presented by the company. Thus, it can be suggested that the relationship between companies and the media is two-folded: media have the power to reinforce but also demolish the CSR image of fashion brands. However, we only found a limited effect, as no significant effects of (in)consistency on other consumer outcomes were yielded. More media exposure and encounters with the brand might be needed to influence consumers’ behavioral intentions.

Limitations and future research

The current study provides valuable new insights in consumer reactions to ethical labor conditions of fashion brands. Future research should determine if the findings are applicable on other CSR-related domains, such as the environmental impact of the manufacturing process of the fashion industry. The non-significant effects of brand positioning might be due to the use of descriptive manipulations. It is important that future research include more interactive brand information with sound and moving images. In addition, future studies should perform a control check on participants’ perceptions of brand positioning and publicity. The assumption that there is a gap between consumers’

(30)

expectations and fashion brands’ value proposition of CSR in their external communication programs should be tested.

Additional research needs to be undertaken to more clearly understand the association between CSR publicity and consumer outcomes. Other non-corporate sources such as activist groups or customers should be included who can inform stakeholders about CSR practices of brands in online environments, as the Internet allows for an increased information flow by a wider variety of sources (Jones et al., 2009; Stokes & Turri, 2013). In addition, following the example of Eberle et al. (2013), a mix of positive and negative publicity should be tested in future research as this resembles the real-life setting more.

It is possible that the inclusion of existing fashion brands in the experiment elicited any preconceptions and prior associations. However, the manipulation check did not indicate significant differences between fictional and non-fictional CSR information and existing brands provided realistic brand information.

Three practical limitations of this study remain. Firstly, due to convenience sampling, there is limited variation in participants’ demographics. Secondly, a lot of participants did not finish the experiment, which was most likely due to the time it took to watch the video materials. In future research, respondents might be awarded with financial compensation to motivate them to finish the experiment. Lastly, only short-term effects are measured. In the current research design, the effect of different CSR messages is measured right after exposure, while concepts such as brand attitude and reputation are based on more than one encounter with a brand (Cornelissen, 2008). Future research should include a follow-up study or a longitudinal approach in which several interactions with CSR information are taken into account. Actual behavior instead of behavioral intentions should be tested in real-life settings such as in warehouses or shops.

Managerial implications

The findings of this study have some implications for fashion brands’ CSR communication. Generally, companies do not provide conclusive information about their

(31)

actual CSR activities, especially when they are not forced to provide inside information. However, companies should realize that it could have disastrous consequences when media starts reporting negatively about their CSR practices. This publicity can undermine the current CSR reputation of a brand and activate stakeholder pressures. Therefore, it is important that brands are more transparent about their business conduct and proactively provide stakeholders with CSR information.

Furthermore, fashion brands should invest more in consumer relations, as they are one of the most important stakeholders. It is suggested that brands’ CSR activities and their consumers’ needs and expectations of CSR should be congruent so that brands can benefit from engagement in CSR. Social media are suitable channels for interaction and dialogue that brands can embed in their CSR communication strategies. Another opportunity is the provision of in-store information about the origin of products. More importantly, brands should engage their employees so they can act as (online) CSR ambassadors of the organization (Du et al., 2010; Verhoeven, 2012). This might reduce skeptical reactions, since several studies have shown the potential of CSR engaged employees (e.g. Morsing et al., 2008; Ter Hoeven & Verhoeven, 2013). In this case publicity will rather reinforce than undermine the positive effect of brands’ CSR engagement.

(32)

References

Aguinis, H. (2011). Organizational responsibility: Doing good and doing well. In S. Zedeck (Ed.), APA handbook of industrial and organizational psychology Vol. 3 (pp. 855-879). Washington, DC: American Psychological Association.

Aguinis, H., & Glavas, A. (2012). What we know and don’t know about corporate social responsibility a review and research agenda. Journal of Management, 38(4), 932-968.

Alam, K., Blanch, L., & Smith, A. (2011). Stitched up: Women workers in the Bangladeshi

garment sector [Report]. London: War on Want.

Alexandris, K., Douka, S., Bakaloumi, S., & Tsasousi, E. (2008). The influence of spectators' attitudes on sponsorship awareness: A study in three different leisure events. Managing

Leisure, 13(1), 1-12.

Amaeshi, K. M., Osuji, O. K., & Nnodim, P. (2008). Corporate social responsibility in supply chains of global brands: A boundaryless responsibility? Clarifications, exceptions and implications. Journal of Business Ethics, 81(1), 223-234.

Arrigo, E. (2013). Corporate responsibility management in fast fashion companies: The Gap

Inc. case. Journal of Fashion Marketing and Management, 17(2), 175-189.

Ashforth, B. E., & Gibbs, B. W. (1990). The double-edge of organizational legitimation.

Organization Science, 1(2), 177-194.

Bae, J., & Cameron, G. T. (2006). Conditioning effect of prior reputation on perception of corporate giving. Public Relations Review, 32(2), 144-150.

(33)

Balmer, J. M. (2001). Corporate identity, corporate branding and corporate marketing-seeing through the fog. European Journal of Marketing, 35(3/4), 248-291.

Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of

Personality and Social Psychology, 51(6), 1173.

Becker-Olsen, K. L., Cudmore, B. A., & Hill, R. P. (2006). The impact of perceived corporate social responsibility on consumer behavior. Journal of Business Research, 59(1), 46-53.

Bhardwaj, V., & Fairhurst, A. (2010). Fast fashion: Response to changes in the fashion industry. The International Review of Retail, Distribution and Consumer Research, 20(1), 165-173.

Bhattacharya, C. B., Korschun, D., & Sen, S. (2009). Strengthening stakeholder–company relationships through mutually beneficial corporate social responsibility initiatives. Journal of

Business Ethics, 85(2), 257-272.

Bhattacharya, C. B., & Sen, S. (2003). Consumer-company identification: A framework for understanding consumers’ relationships with companies. Journal of Marketing, 67(2), 76-88.

Bhattacharya, C.B., & Sen, S. (2004). Doing better at doing good: When, why, and how consumers respond to corporate social initiatives. California Management Review, 47(1), 9-24.

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

(34)

Bielak, D., Bonini, S. M., & Oppenheim, J. M. (2007). CEOs on strategy and social issues.

McKinsey Quarterly, 10(1).

Bitzer, V., & Glasbergen, P. (2010). Partnerships for sustainable change in cotton: An

institutional analysis of African cases. Journal of Business Ethics, 93(2), 223-240.

Blau, P. M. (1964). Exchange and power in social life. New Jersey, NJ: Transaction Publishers.

Boulstridge, E., & Carrigan, M. (2000). Do consumers really care about corporate responsibility? Highlighting the attitude—behaviour gap. Journal of Communication

Management, 4(4), 355-368.

Brown, T. J., & Dacin, P. A. (1997). The company and the product: Corporate associations

and consumer product responses. The Journal of Marketing, 68-84.

Burke, J. (2013, December 8). Bangladesh factory fires: Fashion industry's latest crisis. The

Guardian. Retrieved from http://www.theguardian.com

Campbell, J. L. (2007). Why would corporations behave in socially responsible ways? An institutional theory of corporate social responsibility. Academy of Management Review,

32(3), 946–967.

Capriotti, P. (2009). Economic and social roles of companies in the mass media: The impact

media visibility has on businesses' being recognized as economic and social actors.

(35)

Carrington, M. J., Neville, B. A., & Whitwell, G. J. (2010). Why ethical consumers don’t walk their talk: Towards a framework for understanding the gap between the ethical purchase

intentions and actual buying behavior of ethically minded consumers. Journal of Business

Ethics, 97(1), 139-158.

Carroll, A. B. (1999). Corporate social responsibility evolution of a definitional construct.

Business & Society, 38(3), 268-295.

Carroll, C. E., & McCombs, M. (2003). Agenda-setting effects of business news on the

public's images and opinions about major corporations. Corporate Reputation Review, 6(1),

36-46.

Chaiken, S. (1980). Heuristic versus systematic information processing and the use of

source versus message cues in persuasion. Journal of Personality and Social Psychology,

39(5), 752.

Clarkson, M. E. (1995). A stakeholder framework for analyzing and evaluating corporate social performance. Academy of Management Review, 20(1), 92–117.

Clean Clothes Campaign. (2012). Hazardous workplaces: Making the Bangladesh garment industry safe. Retrieved from http://www.cleanclothes.org/resources/publications/2012-11-hazardousworkplaces.pdf

Cornelissen, J. (2008). Corporate communication: A guide to theory and practice. London: Sage Publications.

Referenties

GERELATEERDE DOCUMENTEN

In this paper, our main contribution is that we present combinations of measurements for error modeling that can be used to estimate the quality of arbitrary GNSS receivers

Now we are ready to define attributed graph transformation: while the rule graphs L and R are attributed with elements from the term algebra, the graphs to be rewritten are

As literature on CSR practices in the global fast fashion industry is still at an early stage, this research will contribute by investigating if consumers that are aware of

For all hypotheses, highly significant positive relationships between the independent variable corporate environmental performance and the dependent variables

Further research will be conducted on the roll of promotion focus as such, but specially the impact of this promotion focus towards the level of proactive behavior and the possible

More precisely, this paper studies the relation between environmental policy and environmental patenting activity in the area of four renewable energy technologies (i.e. wind,

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright

Andere factoren die in dit onderzoek niet onderzocht zijn kunnen ook invloed hebben op het corporate brand image van business-to-business organisaties... Journal