The Effect of Corporate Social Responsibility Performance
on Non-Governmental Organization Activism in the Fashion
Industry
Master Thesis MSc Business Administration
Entrepreneurship and Management in the Creative Industries University of Amsterdam
By
Daisy Altelaar (10632905) Under supervision of K. Quintelier
Statement of originality
This document is written by Daisy Altelaar who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
ABSTRACT
Companies in the fashion industry are increasingly pressured by NGOs to behave in a more socially responsible manner. Consequently, fashion companies proactively adjust their policies and invest in CSR under the assumption that their likelihood to become the target of NGO activism is reduced. In this study we propose that companies where the level of CSR performance is low are more likely to become the target of NGO activism. We tested this with data from 41 company reports of Sustainalytics. Contrary to our expectations, results of this study reveal that companies with a good CSR performance are actually experiencing more NGO activism than companies with a poor CSR performance. Similar results are found in the relation between CSR performance and NGO activism for the social performance of fashion companies in the contractor & supply chain (C&S) and no relationship is found for environmental performance in the C&S. An explanation of the findings can reside in stakeholder scepticism, the perceived motive from the company’s communicated CSR message and in a potential two-way causality of the NGO activism-CSR performance relationship. Nevertheless, managers of fashion companies should be aware that good CSR performance by itself does but not take away the likelihood to become targeted by NGO activists. Therefore, companies should carefully manage their multiple identities and work together to address issues related to the society and environment.
TABLE OF CONTENTS 1. Introduction ---P. 4 2. Literature review --- P. 5 2.1. Stakeholder activism --- P. 6 2.2. NGO activism --- P. 8 2.2.1. Private politics --- P. 8 2.2.2. Target selection --- P. 9 2.3. Corporate social responsibility --- P. 10
2.3.1. Consequences of CSR ---P. 11
3. Theoretical framework --- P. 12
3.1. Fashion industry ---P. 12 3.2. Contractor & supply chain --- P. 13
4. Method --- P. 15 4.1. Database research --- P. 15 4.1.1. Sample --- P. 15 4.2. Measures --- P. 16 4.2.1. Independent variable --- P. 16 4.2.2. Dependent variable --- P. 16 4.2.3. Control variables --- P. 15
5. Analysis & results ---P. 17
5.1. Descriptives --- P. 17 5.2. Correlations --- P. 18 5.3. Normality --- P. 21 5.4. Logistic regression analyses --- P. 22
6. Discussion --- P. 27
6.1. The effect of CSR performance on NGO activism --- P. 27 6.2. Contractor & supply chain --- P. 28 6.3. Managerial implications --- P. 30 6.4 Scientific implications --- P. 31 6.5. Limitations and future directions --- P. 31
1. INTRODUCTION
Actions taken by non-governmental organizations (NGOs) which aim for more ethical and social responsible business practices have caused significant changes in corporate management (Doh & Guay, 2006). Especially in the fashion industry, companies are progressively pushed to modify their strategies and to behave in more socially responsible ways (Hilton, Choi, & Chen, 2004; Kozlowski, Bardecki, & Searcy, 2012). NGOs push these companies by means of boycotting campaigns and spreading sensitive information about the (unethical) behaviour of brands (Degli Antoni & Sacconi, 2013). Only recently, the NGO People for the Ethical Treatment of Animals (PETA) successfully forced more than a dozen fashion companies to ban angora wool from their production. PETA did this by placing well-known retailers, such as Esprit, H&M and Tommy Hilfiger, on the court of popular opinion. PETA released video footage showing how workers violently rip out the wool of living angora rabbits for its fur to be used in the clothing sold by these retailers. The footage dominated the news and angry consumers joined forces at NGOs to take action against the unethical behaviour of the fashion brands (Martens, 2002). This enabled PETA to successfully pressure fashion companies to terminate the production of angora products. Such actions taken by NGOs is named NGO activism (Doh & Guay, 2006).
Companies have a hard time dealing with NGO activism as the negative public exposure damages their reputation and results in distrusting consumers, ultimately leading to negative company performances (Maistruau & Bonardi, 2014). In recent years, more than a few companies underestimated the impact of NGO activism and had to pay the price with their reputation and economic performance (Doh & Guay, 2006). Consequently, companies respond to activism with an (increased) investment in corporate social responsibility (CSR) (Baron, 2001; Maistruau & Bonardi, 2014; Mcwilliams, Siegel, & Wright, 2006). By spending money, for instance on environment protection plans and educational programs for society at large, companies can improve and restore their reputation (Minor & Morgan, 2011). Also, firms try to reduce the likelihood to become a target of activism by embracing policies in favour of the NGO’s requests (Baron, 2003). Hence, CSR is becoming a ‘must’ in the corporate world and accordingly, more companies than ever implement CSR activities into their strategies (Aguilera, Rupp, Williams, & Ganapathi, 2007; Bhattacharya & Sen, 2004).
However, Pedersen (2006) states that when a company engages in CSR it does not mean the firm is no longer vulnerable to NGO activism. In a similar vein, Minor & Morgan (2011) argue that when a firm engages in CSR but still behaves unethically towards society and the environment, the company could experience consequences even more unfavourable than when it did not engage in CSR at all. Such arguments result in a growing interest among managers in the consequences of CSR (Mcwilliams et al., 2006). However, apart from the study of Rehbein, Waddock, & Graves (2004), who found that shareholder activists take action against companies with poor CSR performance, little is research is done after the effect of a company’s stakeholder-related practices on activist pressure.
Therefore, this research aims to identify the effect of CSR performance on companies’ likelihood to become the target to NGO activism. Research will be conducted in the fashion industry, because this industry is often seen as “a significant contributor to global environment and social issues” (Kozlowski et al., 2012, p. 17). Hence, this study aims to answer the following research question: What is the effect of CSR performance on fashion companies’ likelihood to become the target of NGO activism?
This paper is developed as follows. We start with a review of the literature in which the causes of stakeholder activism, with NGO activism in particular, are discussed. Secondly, corporate social responsibility is defined and the consequences of these actions are examined. Based on this, hypotheses are developed about the effect of CSR on NGO activism in the fashion industry. Furthermore, we discuss the theoretical framework and data collection methods used for this study. Accordingly the hypotheses are tested and the results are evaluated. Finally, we provide managerial implications for the unexpected outcomes and give directions for future research.
2. LITERATURE REVIEW
In today’s business environment companies more often become the target of activism (Spar & La Mure, 2003). Due to the centrality of businesses, which connects firms with wider society (e.g. through the use of human and natural resources and contribution to welfare and pollution) actors in the surrounding environment expect companies to share principles with them (Sjöström, 2008). As a result, companies receive a significant amount of
internal and external actors that pressure them to behave more socially responsible (Aguilera et al., 2007; Porter & Kramer, 2006). Especially community groups, environmentalists, human right organizations and charitable organizations use stakeholder group action to influence companies’ practices (Rowley & Moldoveanu, 2003) and to adjust their social impact (Guay, Doh, & Sinclair, 2004; K. Rehbein et al., 2004). Consequently, stakeholder activism has become a widespread issue in the corporate world (den Hond & de Bakker, 2007), making this concept particularly interesting to investigate.
This section provides an overview of the literature on stakeholder activism and corporate social responsibility. First stakeholder activism is defined and the central role of NGOs is discussed. Moreover, this review sheds light on corporate social responsibility, the motives to engage in CSR activities and the associated consequences for companies.
2.1. Stakeholder activism
A stakeholder is, by definition, “any group or individual who can affect or is affected by the achievement of the organization’s objectives” (Freeman, 1994, p. 46) and activism refers to the players “that seek to change the practices of a firm” (Degli Antoni & Sacconi, 2013, p. 216). Traditionally, the practices of a firm were primarily regarded to make value for its shareholders (Hatcher, 2002; Pedersen, 2006). Nevertheless, Freeman (1984) coined the stakeholder theory, arguing that for organizations to be successful in the current and future environment the needs of non-financial stakeholders should be taken into account as well (Mcwilliams et al., 2006). Lamberti & Lettieri (2009) suggest that there are six groups of stakeholders (shareholders, employees, customers, suppliers, community and the environment1) that affect or are affected by the business practices of a firm and for that reason address specific responsibilities. Sjöström (2008) discusses that these stakeholders long tried to influence company practices by means of shareholder proposals, which are “written demands that shareholders can forward to a company and that will be voted upon in the annual general meeting” (p. 145). However, such proposals rarely resulted in actual changes of corporate behaviour (Sjöström, 2008). Therefore,
stakeholders started to use other tools to push for change; they seek to obtain compliance through controversial activities (Maistruau & Bonardi, 2014). These controversial activities generally include the development of harm on companies in order to reach a desired goal (Baron & Diermeier, 2007). Maistruau & Bonardi (2014) state that such “harm can take a variety of forms: the disruption of operations, lawsuits, boycotts, protests, public criticism, reputational damage and criticism against individual executives” (p6).
Moreover, there is an increase in stakeholders groups that aggressively put pressure on firms regarding a social or environmental issue (Porter & Kramer, 2006). They do this by making business practices a matter of public debate and therewith enforce companies to adjust their social practices (den Hond & de Bakker, 2007). The stakeholders have the power to influence companies since they are commonly the providers of resources (Konrad, Steurer, Langer, & Martinuzzi, 2006). Firms for instance need to be accepted by communities and various other stakeholders in order to do business in the first place (Porter & Kramer, 2006). Thus when a company fails to meet the interest of its stakeholders its performance may be negatively affected (Konrad et al., 2006). To illustrate this, Sirgy (2002) has developed the corporate performance model in which he shows that “the success of a company is determined by the extent to which the managers of internal stakeholders meet the needs of external stakeholders, coordinate with one another and manage the corporate image in the eyes and minds of distal groups” (p. 159). Therefore we can say that a good relationship between stakeholders and firms often leads to corporate survival and growth (Sirgy, 2002).
Nevertheless, as it is nearly impossible for stakeholders to change the social practices of companies individually, they often join forces through non-governmental organizations (NGOs) (Martens, 2002). At NGOs, activists organize around shared beliefs and principles and use whatever tools they have to obtain compliance (Spar & La Mure, 2003). Moreover, NGOs are able to use their independent status as advisors and therewith seek to promote change in company policies (Guay et al., 2004). Because NGOs are rapidly growing in popularity and scope (Reid & Toffel, 2009) the focus of this study will specifically be on the role of this actor in stakeholder activism.
2.2. NGO activism
A notable development over the past decades is the increasing number NGOs (Doh & Guay, 2006). Along with the growth in number, NGOs also received more power and influence, enabling them to affect business practices to a significant extent (Guay et al., 2004). Through the increased globalization and developments in technology it has become relatively easy to mobilize and activate people, which enabled the NGOs to successfully pressure companies to change their practices (Leipold, 2000). The success of the actions taken by NGOs encouraged similar organizations to take more action as well (Leipold, 2000). As a result, the number of NGOs, their influence and power continue to grow and companies start to change their practices in line with the NGOs desired goals.
2.2.1. Private politics
According to Sasser, Prakash, Cashore, & Auld (2006), NGOs can influence the policy making of firms in two ways: via public politics and/or through private politics. Public politics is the traditional way in which NGOs target the state with the aim “to influence legislation, regulation and judicial interpretations to institutionalize new set of norms” (Reid & Toffel, 2009, p. 1159). On the contrary, at private politics, activists try to institutionalize a new set of norms without reliance on the law or government (Baron, 2003) and rather use consumer and public pressure to affect the social behaviour of companies (Spar & La Mure, 2003). Nevertheless, in public politics NGOs found themselves very easily blocked by law or the government and therefore shifted to private politics (Baron & Diermeier, 2007). This shift is also a result of the changing role of the government. It was formerly the government’s task to regulate social responsibilities of business organizations (Egels-Zandén & Hyllman, 2006). Their role, however, diminished when organizations started to move business functions outside national borders. Hence, NGOs began to fill this governance gap (Egels-Zandén & Hyllman, 2006) and take up the responsibility to address issues related to social and environmental business performance (den Hond & de Bakker, 2007). Sasser et al. (2006) argue that this role fits NGOs well because “they enjoy a moral authority, derived from its perceived independence, veracity, reliability and representativeness” (p.5). Thus, with their
independent status, NGOs increasingly moved from public to private politics in order to influence the policy making of firms (Baron & Diermeier, 2007; Spar & La Mure, 2003).
2.2.2. Target selection
The increased level of NGO activism has resulted in concerned managers, willing to understand how firms get involved in private politics (Rehbein et al., 2004). With the aim to understand this, it is important to determine how NGOs select their target (Baron, 2003). A few studies have addressed this topic.
First, Spar & La Mure (2003) state that NGOs select a company at which they can address an issue in the most inexpensive and effective way. Therefore, the targeted company preferably is the one with “the greatest potential to affect the problem and the greatest susceptibility to external pressure” (Spar & La Mure, 2003, p. 81). Furthermore, Baron (2003) indicates that companies producing negative externalities, such as harm to the environment or with poor working conditions in the supply chain, belong to the firm characteristics that make firms susceptible to NGO activism. Moreover, further empirical research conducted by Baron & Diermeier (2007) indicates that activists take action for an issue where something really can be changed and select a company that has high valuation (expensively priced stock in comparison to other companies in the same industry), high probability of responsiveness and low costs of a campaign. These authors also clarify that activists target a specific company even when they want to influence an entire industry. One successful campaign may result in other companies to adjust their strategies with the aim to avoid becoming the next target and therewith work as a domino effect throughout an industry (Baron & Diermeier, 2007).
Moreover, after selecting a target, NGOs use consumer and public pressure to harm the firm (Spar & La Mure, 2003). Especially under the threat of controversies such as boycotting campaigns (Degli Antoni & Sacconi, 2013) which damages the company’s public image and thereby threatens its reputation (King, 2008), firms are motivated to take the NGOs seriously and to generate new policies more consistent with the NGO’s interests (Reid & Toffel, 2009). Accordingly, more firms implement policies that decrease their vulnerability to
become the target of private politics (Baron, 2003). Previous research suggests that companies do this in the form of increased investment in corporate social responsibility (Maistruau & Bonardi, 2014).
2.3. Corporate social responsibility
Business organizations increasingly implement corporate social responsibility (CSR) activities into their strategies (Aguilera et al., 2007). These activities involve companies to act responsible towards society and the environment beyond what they are legally obligated to (Vallaster, Lindgreen, & Maon, 2012). The academic literature contains many definitions of, CSR, however, consistent with Aguinis & Glavas (2012) we define CSR as: “context-specific organizational actions and policies that take into account stakeholder’s expectations and the triple bottom line of economic, social and environmental performance” (p. 933). Such organizational actions and policies can be very diverse, ranging from policies within the firm, such as the adjustment of production methods and ethical labour practices, to actions outside the firm, such as environment protection plans and investments in local communities (Aguilera et al., 2007). Thus, CSR actions and policies involve those that reach beyond the company’s own interest and obligations (Vallaster et al., 2012).
Degli Antoni & Sacconi (2013) indicate that within the CSR literature, there lies a particular focus on the firm-stakeholder relationship. They argue that this is because a firm is considered to behave to the norm of CSR, when it acts responsibly towards its stakeholders (Degli Antoni & Sacconi, 2013). Pedersen (2006) states that the increase in CSR “breaks with the notion that shareholders are the only important constituents and that
shareholder wealth is the only relevant criteria for evaluating company behaviour” (p. 138). Meaning that CSR moves away from the traditional view of a company, in which it was expected to put the needs of shareholders first (Hatcher, 2002). Nevertheless, to be successful in the current and future environment not just shareholders but multiple stakeholders have to be taken into account (Mcwilliams et al., 2006) and if firms do not meet stakeholder demands, they increasingly face controversies of NGOs and activist groups (Hatcher, 2002).
To deal with these changes companies often translate stakeholder demands into tailored CSR policies (Lindgreen & Swaen, 2010). However, there are also consequences of a firm’s CSR engagement. These will be discussed in the next section.
2.3.1. Consequences of CSR
Companies that implement CSR practices can benefit in term of gaining a positive company image, better support behaviour and favourable stakeholder attitudes (Du, Bhattacharya, & Sen, 2010). However, even though CSR is often implemented to generate a better firm-stakeholder relationship, companies still experience quite some challenges (D’Amato, Henderson, & Florence, 2009). First of all, the various stakeholders have different expectations of the firm’s responsibilities and when they perceive the CSR initiatives to be window-dressing, they are likely to act against the company instead (Vallaster et al., 2012). In a similar vein, Strike, Gao, & Bansal (2006) discuss the situation when companies behave responsible through CSR activities on one hand, but on the other hand still act irresponsible in other business practices. They provide an example of Nike, one of the major fashion brands that implemented CSR activities in its strategy whilst acting irresponsibly along its supply chain: “the international human right organization ‘Global Exchange’ has reported that Nike employees in developing countries are forced to work excessive hours, are not paid enough to meet their most basic needs, and are subject to violent intimidation if they speak out about labour abuses. At the same time, Nike claims to be committed to alleviating poverty by improving the well being of disadvantaged adolescent girls in the developing world. The company also donated one million US dollars to relief organizations providing aid to the victims of the December 2004 tsunamis” (Strike et al., 2006, p. 850). This example to indicates that companies can generate value when they act responsible, but can destroy it when they act against the norm of CSR (Strike et al., 2006).
Likewise, Minor & Morgan (2011) state that when a company engages in CSR but still does harm to the environment or society, it could experience worse consequences than when it did not implement any CSR activities in the first place. Thus, engagement in CSR activities alone do not take away the firm’s risk to become the target of NGO activism (Pedersen, 2006).
The previously mentioned arguments raise the question whether firms really do well by doing good (Mcwilliams et al., 2006). Therefore, this research aims to identify the effect of CSR performance on the presence of NGO activism. How we conduct this research will be discussed in the theoretical framework.
3. THEORETICAL FRAMEWORK
Now that researchers have identified and defined CSR activities, there is a need to study the strategic role of CSR practices (Mcwilliams et al., 2006). The consisting literature on CSR largely concludes that CSR provides companies with insurance-like benefits and that it tempers stakeholder responses to negative events (Kayser, 2014; Minor & Morgan, 2011). Additionally, the current literature indicates that activists value firms that implement practices labelled as ‘socially responsible’ and boycott firms that do not show ethical concerns for its stakeholders (Degli Antoni & Sacconi, 2013). Therefore, we aim to test whether this really is the case, or if managers should reconsider CSR, given its vulnerability to controversies as we previously discussed. Rehbein et al. (2004) began this process by investigating whether activists examine a company’s stakeholder-related practices when they make target decisions. They specifically indicate how little is still known about the relationship between a company’s stakeholder-related practices and activist pressure. In their empirical research the authors, however, only focused on shareholder activism. They found that shareholder activists take action against companies (with the intention of altering the social behaviour) with poor social performance. This means that shareholder activists are more likely to target companies with lower CSR performance. We build on their findings by focusing on another stakeholder, namely NGOs. Hence, our hypotheses are based on the following proposition: companies where the level of CSR performance is low are more likely to become the target of NGO activism.
3.1. Fashion industry
An industry that experienced a growing demand of companies to behave socially responsible is the fashion industry (Perry & Towers, 2009). Kozlowski et al. (2012) state that “the fashion industry has been increasingly
under the spotlight as a significant contributor to global and environmental issues” (p. 33). Accordingly, fashion companies frequently become the target of NGO activism. Reasons can be found in the fact that businesses in the fashion industry deal with a highly competitive environment of downward price pressure, shorter product life cycles fragmentation and globalization of production, and consumer-led and buyer-driven supply chains (Perry & Towers, 2009). The industry is characterized by elements that run contrary to the notions of ethics, such as rapid changes in styles, novelty and mass consumption (Shaw, Hogg, Wilson, Shiu, & Hassan, 2006). Moreover, the production of fashion products involves intensive labour, which makes it considerably more prone to (work condition) scandals (Dijken, 2007). On top of that, within the fashion industry the public perception of what quality entails is shaped by image rather than how the product is created (Hilton et al., 2004). Accordingly, the CSR behaviour of companies in the fashion industry is increasingly examined in today’s market; when a company behaves unethically, this will be highlighted by activists and the company will find itself victim of bad press (Shaw et al., 2006). This leads to the formulation of the first hypothesis:
Hypothesis 1: Fashion companies with low CSR performance are more likely to become the target of NGO
activism than fashion companies with high CSR performance.
3.2. Contractor & supply chain
Within the fashion industry, a substation amount of NGO action is aimed at corporate performance on environmental and ethical areas in the contractor & supply chain (C&S) (Saicheus, Knox, & Cooper, 2012). Ever since consumers became more fashion conscious, the simple product designs transformed into mass productions at which apparel companies shifted to low cost supply chains2 with shorter lead times (Turker & Altuntas, 2014). By making use of the lower manufacturing costs outside national borders, fashion companies are able to quickly respond to the changing market and meet the reduced prices for fashion products (Bruce, Daly, & Towers, 2004).
2 The term supply chain is used to describe “the flow of goods from the very first process encountered in the production of a product, right through the final sale to the end consumer” (Bruce et al., 2004, p. 152).
Therefore, in the last thirty years apparel companies largely moved to less developed countries to have their goods manufactured (Turker & Altuntas, 2014). However, in these less-developed countries social impacts such as the work conditions and rights are still major issues of concern (Kozlowski et al., 2012). Hence, the changing circumstances and the new ways of doing business in the fashion industry – as fast and cheap as possible – negatively effect the environment and society at large (Turker & Altuntas, 2014), especially those on the bottom of the contractor & supply chain (Kozlowski et al., 2012).
Nevertheless, now that the supply chains became larger and more complex, concerns about the associated environmental and social issues started to be raised by NGOs (Perry & Towers, 2009). Accordingly, NGOs increasingly press for the human social and economic rights and firms are encouraged to move into this direction as well (Robinson, 2003). Companies often respond to this with the adoption of sustainable strategies and practices (Turker & Altuntas, 2014). Thus, now that the work conditions and human rights in the C&S have become a major point of concern for companies – next to NGO activism – we are particularly interested in exploring whether there is an effect of social C&S performance (SC&S performance) on social C&S NGO activism (SC&S activism).
Hypothesis 2a: Fashion companies with low SC&S performance are more likely to become the target of SC&S
activism than fashion companies with high SC&S performance.
There are some significant issues with the production and use of apparel that have an impact on the environment. Kozlowski et al. (2012) state that “the environmental impacts associated with the production and use of apparel throughout its lifespan include wastewater emission, solid waste production and significant depletion of resources from the consumption of water, minerals, fossils fuels and energy” (p. 19). Nevertheless, there is a growing role in corporate environmental sustainability, which stakeholder indicate as one of the most important aspects in a company’s CSR efforts (Babiak & Trendafilova, 2011). Accordingly, companies that perform poorly on environmental sustainability in the contractor or supply chain have the potential to fall target of controversies or
incidents (Sustainalytics, 2012). This leads to an extension of the second hypothesis in which we explore whether there is an effect of environmental C&S performance (EC&S performance) on environmental C&S NGO activism (EC&S activism).
Hypothesis 2b: Fashion companies with low EC&S performance are more likely to become the target of EC&S
activism than fashion companies with high EC&S performance.
4. METHOD
4.1. Database research
Research is conducted by means of database research. Data is drawn from the Sustainalytics database of the years 2010-2012 Sustainalytics is a research firm specialized in environmental, social and governance (ESG) research and analysis for companies all over the globe. For each company analysed, Sustainalytics generates a company report on its corporate responsibility performance. The reports are based on items extracted from multiple sources, such as financial accounts, company documentation, databases, media reports and ongoing contact with managers (Surroca, Tribó, & Waddock, 2010). This means of research is chosen as all items are assessed by independent experts, making the collected data less susceptible to social desirability bias than other data collection methods, such as survey collection (Sustainalytics, 2012).
4.1.1. Sample
Companies that fall under the scope of this research are fashion companies with apparel as a primary focus in the year 2010. Sustainalytics released 3535 company reports of this year. The companies are classified into sectors according to the Global Industry Classification Standard (GICS). Firms in the fashion industry are classified in the GICS industry groups ‘consumer durables and apparel’ and ‘retailing’. Nevertheless, these groups cover companies beyond the scope of this research, therefore the company description of the 299 selected companies are carefully read. Hence, only the companies that operate as an apparel retailer are included – otherwise the
controversies could be undertaken for other issues than those related to apparel. Therefore shopping mall operators and broad line retailers are left out of the sample. Also, manufacturing companies are excluded since these firms bring their products on the market through the brands that are already in the sample. Moreover, the companies also have to be evaluated by Sustainalytics in the years 2011 and 2012 (in reports of these years the number of controversies are identified) and their company reports have to be complete (i.e. no missing or inaccurate scores). This results in a sample of 41 companies from 13 different regions.
4.2 Measures
4.2.1. Independent variable
The predictor variable is CSR performance, which consists of the company’s ESG (environmental, social and governmental) score of 2010 – the year before the controversies took place. ESG-scores have become industry standard for evaluating corporate social responsibility (Auer, 2014). In the company reports each of the companies receive a separate environmental, social and governance score (ranging from 0 to 100, at which 0 is the lowest score). The Sustainalytics rating process is conducted on the indicator level. For each of the three dimensions of CSR, a comprehensive set of generic and sector-specific indicators is analysed and scored. The set of indicators relevant for a company is determined by the specifics of its business model and value creation chain. Thus is varies per industry as does the weighting of the individual indicators that is required to build environmental, social and governance scores. We look at the total ESG score for the first hypothesis and the S-score and E-S-score for contractor- and supply-chain performance at the second hypotheses.
4.2.2. Dependent variable
In order to measure the presence of NGO activism, the number contractor and supply chain-related controversies or incidents of the companies are indicated. These include, for example, controversies or incidents in the supply chain amongst contractors related to poor labour standards, health and safety concerns, the use of child or forced labour, freedom of association and strikes and lockouts. Only the controversies or incidents in which NGOs
played a role are included in this study. Also, we aim to measure whether the level of CSR performance has a predictor effect the presence of NGO activism. Therefore we only look into the controversies that took place in 2011, the year after the measured level of CSR performance. These controversies are drawn from the company reports of 2011 and 2012.
4.2.3. Control variables
As the set of indicators and their weight Sustainalytics used varies per industry, we have taken industry as a control variable in this study. Moreover, Rehbein, Waddock, & Graves (2004) provide evidence and that larger companies have a higher chance of falling target of activism. Hence, market capitalization in million USD and number of employees are used to indicate the size of the company.
5. ANALYSIS & RESULTS 5.1. Descriptives
This study is completed with data of eight interval variables: total ESG score, social contractor & supply chain performance (SC&S performance), environmental contractor & supply chain performance (EC&S performance), total number of NGO controversies (total controversies), total number of social contractor & supply chain NGO controversies (SC&S controversies) and total number of environmental contractor & supply chain NGO controversies (EC&S controversies), market capitalization in million USD, number of employees and two categorical variables: industry and presence of NGO activism. The variable presence of NGO activism is ‘present’ when the company became target of at least one controversy. The variable total number of controversies is the sum of social- and environmental controversies a company experienced. For the purpose of this analysis numeric codes were created for the categorical variables. The control variable industry was recoded into ‘1’ for retailing and ‘2’ for Consumer Durables & Apparel. Moreover to get a clear overview whether NGO activism was present or absent, the dependent variable is also coded with ‘0’ for absent and ‘1’ for present.
Table 1 Descriptive statistics categorical variables
Variable Levels N %
Industry Retailing 23 56.1
Consumer Durables & Apparel 18 43.9
NGO activism Absent 26 63.4
Present 15 36.6
Table 2 Descriptive statistics interval variables
Variable N Min. Max. Mean Std. Dev.
ESG score 41 36 70 49.90 10.561 SC&S performance 41 17 100 46.20 23.686 EC&S performance 41 38 90 53.44 13.495 Total controversies 41 0 5 .78 1.275 SC&S controversies 41 0 4 .59 1.072 EC&S controversies 41 0 2 .20 .459
Market capitalization in million USD 41 1194 76483 12085.02 16146.822 Number of employees 41 751 161000 36210.88 41365.075
The sample consists of 23 companies classified in the industry ‘Retailing’ and 18 companies in the industry ‘Consumer Durables & Apparel’. In total 41 companies were analysed of which 36.6% of the companies had to deal with NGO activism in 2011. Apart from the categorical variables, eight interval variables were used in this study. The descriptive statistics of eight interval variables are outlined in Table 2. Notable is that the total number of NGO controversies that were present at companies is fairly low (ranging between zero and five controversies per company). Especially for the number of EC&S controversies, it can be noted that NGO activism barely occurred (maximum of two controversies per company). The mean for the controversies indicates that there were more SC&S controversies (.59) than EC&S controversies (.20).
5.2. Correlations
The correlations between the variables are identified by the Pearson correlation coefficient (r). Figure 3 provides an overview of the correlation matrix at which the significant correlations at a five percent level (0.05) are
marked with one asterisks (*) and the significant correlation a one percent level (0.01) are market with two asterisks (**).
First it can be noted that there is a significant positive correlation between total ESG score and total controversies r = .520, p < .01 (p = .000). This means that when a company’s score on CSR performance increases, also the number of controversies increases.
Moreover, there is also a significant positive correlation r = .494, p < .01 (p = .001) between SC&S performance and SC&S controversies, meaning that companies that score high on SC&S performance also experience more SC&S controversies. Nevertheless, there is no significant correlation p > .01 (p = .152) found between EC&S performance and ES&S controversies.
Furthermore, the correlation matrix indicates that market capitalization in million USD is better used as a control variable. There is a correlation between market capitalization and ESG score r = .385, p < .05 (p = .013), meaning that larger companies score better on CSR. Nevertheless, there is no correlation between number of employees and ESG score p > .05 (p = .077). In addition, this variable does correlate with EC&S performance p < .01 (p = .003) but does not correlate with SC&S performance p > .01 (p = .060). Therefore the variable market capitalization is used as control variable size.
Table 3 Correlation matrix
**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).
1 2 3 4 5 6 7 8 9 10
1. ESG score Pearson’s r 1 .905** .800** .520** .593** .061 .234 .385* .279 .386*
2. SC&S performance Pearson’s r .905** 1 .658** .425** .494** .029 .098 .296 .260 .299 3. EC&S performance Pearson’s r .800** .658** 1 .577** .588** .228 .472** .446** .137 .385* 4. Total controversies Pearson’s r .520** .425** .577** 1 .938** .587** -.080 .266 .260 .816** 5. SC&S controversies Pearson’s r .593** .494** .588** .938** 1 .270 -.071 .255 .299 .728** 6. EC&S controversies Pearson’s r .061 .029 .228 .587** .270 1 -.055 .144 .022 .566**
7. Industry Pearson’s r .234 .098 .472** -.080 -.071 -.055 1 .068 -.301 -.162
8. Market capitalization in million
USD (size) Pearson’s r .385
*
.296 .446** .266 .255 .144 .068 1 .412** .101
9. Number of employees (size) Pearson’s r .279 .260 .137 .260 .299 .022 -.301 .412** 1 .123 10. Presence NGO activism Pearson’s r .386* .299 .385* .816** .728** .566** -.162 .101 .123 1
Table 4 Distribution values
Skewness Kurtosis
Total controversies 1.728 2.430
SC&S controversies 1.816 2.341
EC&S controversies 2.375 5.362
Table 5 Test of normality
Kolmogorov-Smirnova Shapiro-Wilk
Statistic df Sig. Statistic df Sig.
Total controversies .364 41 .000 .677 41 .000
SC&S controversies .415 41 .000 .614 41 .000
EC&S controversies .494 41 .000 .475 41 .000
Market capitalization in million USD .250 41 .000 .648 41 .000
a. Lilliefors Significance Correction
5.3. Normality
To check if the distribution of the data is approximately normal, normality tests were conducted among the dependent variables total controversies, SC&S controversies and EC&S controversies. First, we looked at the values of skewness and kurtosis (see table 4 in the appendix). All three variables are positively skewed (1.728; 1.816; 2.375) which indicates that there are too many low scores in the distribution. Moreover, all values of the kurtosis are positive (2.430; 2.341; 5.362), indicating a pointy and heavy-tailed distribution. The skewness and kurtosis values (all far from zero) indicate the variables are likely not to be normally distributed. Hence, a normality test is conducted and the Shapiro-Wilk test p < .05 (p = .000) is indeed significant. This means that the distributions are different from a normal distribution (i.e. none of the variables are normally distributed).
Accordingly, data was transformed by taking the logarithm of the set of numbers (with a constant added to cover up the value of zero). However, since the range of the values lies relatively close to one another and contains many values of zero, the data remains non-normal. Therefore, the data for these three variables is transformed into categorical variables: for the variable total controversies we already conducted a binary variable to indicate the number of companies that had to deal with controversies, the variables SC&S NGO controversies and EC&S controversies were also transformed into binary variables, at which the values were coded with ‘0’ when controversies were absent and with ‘1’ when controversies were present. Thus, the outcome variables were transformed from continuous to categorical variables. Therefore, hypotheses were tested using logistic regression analyses, which does not make any assumptions of normality, linearity and homogeneity of variance for the independent variables.
5.4. Logistic regression analysis
Multivariate logistic regression analyses were done to test the hypotheses proposed in the literature review. These analyses compute the probability that a company with a particular score on CSR performance is a member of the modelled category, namely presence of NGO activism. The variables were included in the regression equation using the hierarchical method in which control variables (industry and size) were entered in the analysis (model 1) before the predictor variable (model 2), whose effect we were primarily concerned with. Based on this, the significant model is selected.
The first hypothesis is: ‘Fashion companies with low CSR performance are more likely to become the target of NGO activism than fashion companies with high CSR performance’. For this hypothesis we tested the effect of CSR performance (ESG score) on NGO activism (total number of NGO controversies).
Table 7 Classification Table
Observed
Predicted
NGO activism Percentage
Correct Absent Present
Step 1 Total controversies Absent 23 3 88.5
Present 6 9 60.0
Overall Percentage 78.0
The results of the logistic regression are shown in Table 6. The Omnibus Tests of Model Coefficients shows the chi-square statistic (X 2), which indicates whether this data fits the test. There are no significant results in model 1 (see Table 6) p > .05 (p = .446), whilst model 2 yields a chi-square of 9,982 which is significant p < .05 (p = .019). This means that model 2, with the variable ESG score, better fits the data than model 1 without this variable. Additionally, the classification model (see Table 7) documents the validity of the predicted probabilities (Peng, Lee, & Ingersoll, 2010). The overall predictive ability of this model is 78%, which indicates that 78% of the outcome was correctly predicted by the model. The model also shows that the prediction for absence of controversies was more accurate than
Table 6 Logistic regression
Model 1 Model 2
B Wald Sig. B Wald Sig.
(Constant) -1.165 3.638 .056 -7.184 7.909 .005
Size .000 .531 .466 .000 .126 .723
Industry .738 1.167 .280 1.520 3.071 .080
ESG score .114 6.648 .010
X 2 df Sig. X 2 df Sig.
Omnibus Test of Model Coefficients 1.615 2 .446 9.982 1 .019
Nagelkerke R square .053 .296
that for presence of controversies (60%). Moreover, the statistical significance of individual regression coefficients (i.e. B) is tested using the Wald statistic. Only ESG score has a significant effect p < .05 (p = .010), meaning that CSR performance is a predictor of the presence of NGO activism. Besides, Nagelkerke R square = .296 identifies how much of the variance of the dependent variable is explained by the predictor variables, meaning that 29.4% of NGO activism is explained by CSR performance. Thus, we can conclude that there indeed is a relation between CSR performance and NGO activism. However, there is a positive correlation, meaning that hypothesis 1 is not supported.
Hypothesis 2a is: ‘Fashion companies with low SC&S performance are more likely to become the target of SC&S activism than fashion companies with high SC&S performance’. Hence, the effect of SC&S performance on the number of SC&S controversies is tested. The results of the logistic regression are shown in Table 8. The Omnibus Test of Model Coefficients shows that the first model (at which the control variables are part of the equation without the predictor variable) does not fit the data p > .05 (p = .320). Nevertheless, the second model at which the predictor variable ‘SC&S performance’ is also part of the equation, the chi-square of 8,872 is significant p < .05 (p = .031). Meaning that model with the predictor variable SC&S performance better predicts the presence of SC&S controversies. The overall predictive ability of this model is 68.3% (see Table 9), meaning that 68.3% of the outcome was correctly predicted by the model. The model also shows that the prediction for absence of SC&S activism (86.2%) was more accurate than that for presence of SC&S controversies (25%). Moreover, the Wald statistic indicates that only SC&S performance has a significant predictor effect on the presence of SC&S controversies p < .05 (p = .019). Furthermore, Nagelkerke R square = .277 indicates that 27.7% of SC&S controversies is explained by SC&S performance. Therefore, we can conclude that there is a
Table 8 Logistic regression
Model 1 Model 2
B Wald Sig. B Wald Sig.
(Constant) -1.640 5.865 .015 -3.948 8.518 .004
Size .000 1.486 .223 .000 .234 .629
Industry .719 .942 .332 1.133 1.758 .186
SC&S performance .044 5.519 .019
X 2 df Sig. X 2 df Sig.
Omnibus Test of Model Coefficients 2.281 2 .320 8.872 3 .031
Nagelkerke R square .077 .277
Table 9 Classification Table
Observed
Predicted
SC&S controversies Percentage Correct Absent Present
Step 1 SC&S controversies Absent 25 4 86.2
Present 9 3 25.0
Overall Percentage 68.3
relation between the variables. However, similar to our results for hypothesis 1, both variables are positively correlated, meaning that hypothesis 2a is not supported.
Hypothesis 2b is: ‘Fashion companies with low EC&S performance are more likely to become the target of EC&S activism than fashion companies with high EC&S performance’. For this hypothesis we tested the effect of EC&S performance on EC&S controversies. Table 10 documents the results of the logistic regression. Similar to the previous hypotheses, the first model does not yield any significant results p > .05 (p = .265).
Table 10 Logistic regression
Model 1 Model 2
B Wald Sig. B Wald Sig.
(Constant) -2.632 8.165 .004 -7.375 7.426 .006
Size .000 1.952 .162 .000 .022 .882
Industry .961 1.016 .314 1.963 2.817 .093
EC&S performance .081 4.060 .044
X 2 df Sig. X 2 df Sig.
Omnibus Test of Model Coefficients 2.656 2 .265 6.941 3 .074
Nagelkerke R square .105 .260
Table 11 Classification Table
Observed
Predicted
EC&S controversies Percentage Correct Absent Present
Step 1 EC&S controversies Absent 33 4 97.1
Present 5 3 28.6
Overall Percentage 85.4
Nevertheless, for this hypothesis we find that the data for the second model (with the predictor variable ‘ES&C performance’ in the equation) does not fit the data either p > .05 (p = .074). So there is no effect of the predictor variables, taken together, on the dependent variable. The overall predictive ability of the model is 85% (see Table 11), which very accurately predicts the absence of EC&S controversies (97.7%). Nevertheless, the Wald statistic indicates that EC&S performance is the only significant predictor of the presence of environmental C&S NGO activism p < .05 (p = .044). This predictive ability, however, is very weak. Moreover, Nagelkerke R square = .260 indicates that 26% of EC&S controversies is explained by EC&S performance. However, the correlation that is found
between the variables is positive, but was not significant. Therefore, we find no support for hypothesis 2b.
6. DISCUSSION
In this section we discuss the results of our study. First we evaluate the general effect of CSR performance on NGO activism in the fashion industry. Followed by the results of this effect in the contractor & supply chain, where we look at the social and environmental performance. Furthermore, we discuss the scientific and managerial implications. Lastly, we outline the potential limitations and give directions for future research.
6.1. The effect of CSR performance on NGO activism
Building on the study of Rehbein, Waddock, & Graves (2004) we proposed that companies where the level of CSR performance is low are more likely to become the target of NGO activism. Our findings indicate that CSR performance has a significant effect on the presence of NGO activism. However, we find that this effect is significant in the positive direction, opposed to the hypothesized expectation. This implies that companies with good CSR performance experienced more NGO activism than companies with low CSR performance. This is a notable outcome given that NGO activism involves NGOs that push companies to adjust their strategies and behave in more socially responsible ways. Nevertheless, there are various clarifications for these findings. As Pedersen (2006) already argued, engagement in CSR does not necessarily mean that companies are no longer vulnerable to becoming the target of NGO activism. When a company scores high on CSR performance, NGOs can become sceptical about the true motives of the organization. In a study completed by Yoon, Gürhan-Canli, & Schwarz (2006) it is found that the effectiveness of CSR activities is mediated by the perceived sincerity of motives. Thus, when NGOs sense ambiguous motives,
CSR has the potential to backfire. This is for instance the case when companies engage in CSR primarily out of economic self-interest (Graafland & Ven, 2006; D’Amato, Henderson, & Florence, 2009). In a similar vein Du, Bhattacharya, & Sen (2010) state that it is a challenge for companies to minimize stakeholder scepticism when engaging in CSR. These authors argue that the communication message that plays an important role in this; “stakeholders quickly become leery of CSR motives when the company aggressively promotes its CSR efforts” (p. 9).
Moreover, this study only took NGO activism into consideration after the level of CSR performance was measured. Several authors have suggested that companies respond to NGO activism with an increased investment in CSR (Auer, 2014; Maistruau & Bonardi, 2014). Therefore, clarifications for the contra-predicted outcome can also be found in a two-way causality, in which CSR promotes NGO activism and NGO activism promoted CSR.
Finally, another factor that plays a role in our finding is that we specifically looked into the fashion industry. This industry is a unique case as companies deal with various factors that run contrary to the notions of CSR (Moedas, 2007; Shaw et al., 2006). We will discuss this in more detail in the next section.
6.2. Contractor & supply chain
CSR is built on the notion that, in addition to financial responsibilities to shareholders, firms should assume social and environmental responsibilities to stakeholders (Perry & Towers, 2009). In the fashion industry, the focus of the social and environmental responsibilities is especially in the supply chain, where stakeholders have increasingly been raised their concerns about (Saicheus et al., 2012). Therefore, we decided to look specifically into social and environmental performance of firms, and the effect on NGO activism within the
contractor & supply chain. In this study we divided the C&S into two sections: social and environmental.
Starting with the social side of the C&S, we found that CSR performance is a significant predictor of NGO activism. Nevertheless, similar to the general effect of CSR performance on NGO activism, this effect is significant in the positive direction, opposed to the hypothesized expectation. This implies that companies with good SC&S performance experienced more SC&S activism than the companies with poor SC&S performance.
For the C&S, additional clarifications can be found in the difficulties companies face in managing and monitoring their supply chain and accordingly, in ensuring quality standards along the chain (Turker & Altuntas, 2014). Especially because the fashion industry is very labour intensive, it is more prone to working condition scandals (Dijken, 2007). As a consequence, fashion companies are frequently associated with incidents that occur at the bottom of the supply chain (Kozlowski et al., 2012). An example can be found in the building collapse of textile factory ‘Rana Plaza’ in Bangladesh in 2013, with a death-till of 1,127 employees. After this incident, NGOs (i.e. Clean Clothes Campaign) linked 23 fashion companies to the factory collapse because they had their products manufactured in that factory, being part of those company’s C&S (Kayser, 2014). NGOs held the fashion companies responsible for the incident and made them victim of bad press. Actually, what the activists wanted was to raise public awareness for the issue, which they did this by seeking linkages with the companies known to the public (Saicheus et al., 2012). Due to the high complexity of the supply chain in the fashion industry, which generally is very long and has large number of parties involved, it is relatively easy for NGOs to link fashion companies with incidents. However, when such incidents occur, NGOs do not necessarily target a company to push it to adjust its policies alone, but also to get attention for the social and environmental issues within the fashion industry as a whole (Kayser, 2014; Saicheus et al.,
2012). When the latter is the case, there may be other firm characteristics that weight higher for NGOs to select a specific target, such as the firm’s potential to affect the problem and its susceptibility to external pressure (Spar & La Mure, 2003). In a similar vein, Eesley & Lenox (2005) suggest that environmental advocacy groups tend to target firms regardless of their environmental performance. They illustrate this with an example of DuPont, a chemical giant, which finds itself the target of powerful stakeholder actions even though the company is a relatively good environmental performer given its size and industry. However, for the environmental aspect of the C&S, we find no support for the effect of CSR performance on NGO activism. This is likely due to the low number of environmental NGO controversies that were reported by Sustainalytics in 2011.
6.3. Managerial implications
Many companies try to avoid activism with an increased investment in CSR (Baron, 2003; Maistruau & Bonardi, 2014; Mcwilliams et al., 2006; Turker & Altuntas, 2014). However, results of this study indicate that companies with good CSR performance still become the target of NGO activism. Therefore, the management of fashion companies should be aware that they are operating in an industry that is very prone to work condition scandals (Dijken, 2007). When operating in such industries, companies should be aware that CSR leadership alone does not work to avoid NGO activism (Vallaster et al., 2012). Managers should, for instance, carefully manage a company’s multiple identities (i.e. actual identity, communicated identity and conceived identity) to decrease the likelihood of becoming the target of NGO activism (Vallaster et al., 2012).
Moreover, fashion companies should come together to address the social and environmental issues. Dijken (2007) discussed that Nike, GAP and Inditex have formed the “Global Alliance”, in which they work together “to promote the concept of CSR in
developing countries by examining issues in the C&S such as child labour, forces labour, safety compensation etc.” (p. 146). Fashion companies become less vulnerable when they make alignments with other companies in the industry and stand stronger towards potential actions taken by NGOs.
Additionally, companies should search for new ways of being sustainable along the supply chain. For instance by improving sustainability-related communications with stakeholder groups (Saicheus et al., 2012). Managers should also be aware that CSR is becoming an industry norm. Therefore the traditional fashion companies have to adjust their standards not to avoid to activism, but simply to not loose market share to its competitors.
6.4. Scientific implications
Given the increase in NGOs that pressure firms to act more socially responsible, in combination with the substantial growth of CSR initiatives (Baron, 2003; Lindgreen & Swaen, 2010; Maistruau & Bonardi, 2014), we felt it was important to start the process of understanding how CSR effects NGO activism. The literature generally suggested that companies should respond to activism with CSR initiatives (Baron, 2003; Kayser, 2014). However, there was a need to examine the strategic role of CSR companies (Mcwilliams et al., 2006). This research is a stepping-stone to further investigations on the effect of CSR performance on stakeholder activism.
6.5. Limitations and future directions
There are some limitations of this research. First of all, this study relies on the data from company reports of Sustainalytics. These reports provide us with a limited amount of data on the company’s performance. In order to back-up the current study more qualitative research is needed. Moreover, this study is completed with a sample size of 41 companies, which is a
relatively small sample given the excessive amount of fashion companies around the globe. Hence, to make the study more generalizable, research should be completed with a larger sample. Furthermore, future research should also incorporate other elements to assess the relationship between CSR and NGO activism. It would be especially interesting to investigate the relationship with perceived sincerity of motives as moderator to indicate whether this plays a role in the success of the CSR activities, as Yoon et al. (2006) found in their study. Besides, empirical research should be completed including more companies characteristics as indicated by Rehbein et al. (2004), such as the company’s community relationship, progressiveness, visibility and potential to change.
Moreover, because we did not perform an experimental design – in which we manipulated the independent variable – we cannot be sure about the direction of the relationship between CSR performance and NGO activism. Consequently, we cannot exclude the possibility that certain companies were already targeted by NGOs prior to the moment we measured their CSR performance. For future research, we therefore recommend an experimental or longitudinal design so causality can be inferred from multiple measurement periods.
Lastly, in this research we only focus on the fashion industry and discuss that this affect our outcomes as well. For that reason, it would be interesting to replicate the study with companies from various industries and compare the results in order to indicate whether the relationship between CSR performance and NGO activism is really industry specific. We also leave this challenge to future research.
7. CONCLUSION
The main goal of this study was to examine the effect of CSR performance on companies’ likelihood to become the target of NGO activism. Because companies in the fashion industry
have increasingly been under the spotlight as significant contributors to global environment and social issues, research is conducted in this industry.
The current literature generally suggests that CSR activities provide insurance-like benefits and that it tempers stakeholder responses to negative events. Hence, we proposed that companies where the level of CSR performance is low are more likely to become the target of NGO activism. However, in contrast to this expectation, findings indicate the opposite: fashion companies that perform high on CSR performance actually experience more NGO activism than fashion companies with low CSR performance. This is a notable outcome given that NGO activism involves NGOs that push companies to adjust their strategies and behave in more socially responsible ways. A similar relationship between CSR performance and NGO activism occurs at the SC&S, whereas no relationship is found for the EC&S.
Clarifications for these findings include that the effectiveness of CSR activities is often mediated by the perceived sincerity of motives. NGOs can become sceptical about the true motives of an organization, especially when a company actively promotes its CSR efforts. Moreover, explanations for the contra-predicted outcome can be found in a two-way causality between CSR performance and NGO activism and the fact that this study is conducted in the fashion industry. In this industry companies deal with numerous factors that run contrary to the notions of CSR.
Nevertheless, future research should look into the relationship between CSR performance and NGO activism more specifically, using an experimental design or longitudinal research design. In the mean time, managers of fashion companies must be aware that good CSR performance alone does but not take away the risk to become the target of NGO activism. Therefore, fashion companies should carefully manage their multiple identities and work together to address issues related to the society and environment.
REFERENCES
Aguilera, R. V, Rupp, D. E., Williams, C. A., & Ganapathi, J. (2007). Putting the S Back in Corporate Social Responsibility: A Multilevel Theory of Social Change in
Organizations. Academy of Management Review, 32(3), 836–863.
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. doi:10.1177/0149206311436079
Auer, B. R. (2014). Do Socially Responsible Investment Policies Add or Destroy European Stock Portfolio Value? Journal of Business Ethics. doi:10.1007/s10551-014-2454-7 Babiak, K., & Trendafilova, S. (2011). CSR and Environmental Responsibility: Motives and
Pressures to Adopt Green Management Practices. Corporate Social Responsibility and Environmental Management, 18(1), 11–24.
Baron, D. P. (2001). Private Politics, Corporate Social Responsibility, and Integrated Strateg. Journal of Economics & Management Strategy, 10(1), 7–45.
doi:10.1111/j.1430-9134.2001.00007.x
Baron, D. P. (2003). Private Politics. Journal of Economics & Management Strategy, 12(1), 31–66. doi:10.1162/105864003321220724
Baron, D. P., & Diermeier, D. (2007). Strategic Activism and Nonmarket Strategy. Journal of Economics & Management Strategy, 16(3), 599–634.
Bhattacharya, C. B., & Sen, S. (2004). Doing Better at Doing Good: When, Why and How Consumers Respons to Corporate Social Initiatives. California Management Review, 47(1), 9–25.
Bruce, M., Daly, L., & Towers, N. (2004). Lean or agile: A solution for supply chain management in the textiles and clothing industry? International Journal of Operations & Production Management, 24(2), 151–170. doi:10.1108/01443570410514867
D’Amato, A., Henderson, S., & Florence, S. (2009). Corporate Socil Responsibility and Sustainable Business: A guide to Leadership Tasks and Functions (pp. 1–91).
Degli Antoni, G., & Sacconi, L. (2013). Social responsibility, activism and boycotting in a firm-stakeholders network of games with players’ conformist preferences. Journal of Socio-Economics, 45, 216–226. doi:10.1016/j.socec.2013.05.016
Den Hond, F., & de Bakker, F. G. a. (2007). Ideologically Motivated Activism: How Activist Groups Influence Corporate Social Change Activities. Academy of Management Review, 32(3), 901–924. doi:10.5465/AMR.2007.25275682
Dijken, F. Van. (2007). Corporate social responsibility: market regulation and the evidence. Managerial Law, 49(4), 141–184. doi:10.1108/03090550710836562
Doh, J. P., & Guay, T. R. (2006). Corporate Social Responsibility, Public Policy, and NGO Activism in Europe and the United States: An Institutional-Stakeholder Perspective. Journal of Management Studies, 43(1), 47–74.
Du, S., Bhattacharya, C. B., & Sen, S. (2010). Maximizing Business Returns to Corporate Social Responsibility (CSR): The Role of CSR Communication. International Journal of Management Reviews, 12(1), 8–19. doi:10.1111/j.1468-2370.2009.00276.x
Eesley, C. E., & Lenox, M. J. (2005). Secondary Stakeholder Actions and the Selection of Firm Targets. Adademy of Management Proceedings, 1, 1–38.
Egels-Zandén, N., & Hyllman, P. (2006). Exploring the Effects of Union – NGO
Relationships on Corporate Responsibility: The Case of the Swedish Clean Clothes Campaign. Journal of Business Ethics, 64(3), 303–316.
Freeman, E. (1994). The Politics of Stakeholder Theory: Some Future Directions. Business Ethics Quarterly, 4(4), 409–421.
Graafland, J., & Ven, B. van de. (2006). Strategic and moral motivation for corporate social responsibility. Journal of Corporate Citizenship, 111–124. Retrieved from
http://www.ingentaconnect.com/content/glbj/jcc/2006/00002006/00000022/art00012 Guay, T., Doh, J. P., & Sinclair, G. (2004). Non-Governmental Organizations, Shareholder
Activism, and Socially Responsible Investments: Ethical, Strategic, and Governance Implications. Journal of Business Ethics, 52(1), 125–139.
Hatcher, M. (2002). New corporate agendas. Journal of Public Affairs, 3(1), 32–38. Hilton, B., Choi, C. J., & Chen, S. (2004). The Ethics of Counterfeiting in the Fashion
Industry: Quality, Credence and Profit Issues. Journal of Business Ethics, 55(4), 345– 354. doi:10.1007/s10551-004-0989-8
Kayser, S. A. (2014). Guilt by Association : The Cost of Corporate Social Responsibility and Activist Pressure after a Catastrophe (pp. 1–37).
King, B. G. (2008). A Political Mediation Model of Corporate Response to Social Movement Activism. Administrative Science Quarterly, 53(3), 395–421.
Konrad, A., Steurer, R., Langer, M. E., & Martinuzzi, A. (2006). Emperical Findings on Business: Society Relations in Europe. Journal of Business Ethics, 63(1), 89–105. Kozlowski, A., Bardecki, M., & Searcy, C. (2012). Environmental Impacts in the Fashion
Industry: A Life-cycle and Stakeholder Framework. Journal of Corporate Citizenship, 2012(45), 16–36. doi:10.9774/GLEAF.4700.2012.sp.00004
Lamberti, L., & Lettieri, E. (2009). CSR Practices and Corporate Strategy: Evidence from a Longitudinal Case Study. Journal of Business Ethics, 87(2), 153–168.
Leipold, G. (2000). Campaigning: A Fashion or the Best Way to Change the Global Agenda? Development in Practice, 10(3 & 4), 453–460.
Lindgreen, A., & Swaen, V. (2010). Corporate Social Responsibility. International Journal of Management Reviews, 12(1), 1–7. doi:10.1111/j.1468-2370.2009.00277.x
Maistruau, E. A., & Bonardi, J.-P. (2014). How Much Does Negative Public Exposure on Environmental Issues Increase Environmental Performance? (pp. 1–40).
Martens, K. (2002). Mission Impossible? Defining Nongovernmental Organizations. International Journal of Voluntary and Nonprofit Organisations, 13(3), 271–285. Mcwilliams, A., Siegel, D. S., & Wright, P. M. (2006). Corporate Social Responsibility :
Strategic Implications. Journal of Management Studies, 43(1), 1–18.
Minor, D., & Morgan, J. (2011). CSR as Reputation Insurance: Primum Non Nocere. California Management Review, 53(3), 40–60.
Moedas, C. A. (2007). Globalisation and regionalisation in the clothing industry: survival strategies for UK firms. International Journal of Entrepreneurship and Small Business, 4(3), 291. doi:10.1504/IJESB.2007.013253
Pedersen, E. R. (2006). Making Corporate Social Responsibility (CSR) Operable: How Companies Translate Stakeholder Dialogue into Practice. Business and Society Review, 111(2), 137–163.
Peng, C.-Y. J., Lee, K. L., & Ingersoll, G. M. (2010). An Introduction to Logistic Regression Analysis and Reporting. The Journal of Educational Research, 96(1), 3–14.
doi:10.1080/00220670209598786
Perry, P., & Towers, N. (2009). Determining the antecedents for a strategy of corporate social responsibility by small- and medium-sized enterprises in the UK fashion apparel
industry. Journal of Retailing and Consumer Services, 16(5), 377–385. doi:10.1016/j.jretconser.2009.05.003
Porter, M. E., & Kramer, M. R. (2006). Strategy & Society Strategy & Society. Harvard Business Review, 11, 1–15.
Rehbein, K., Waddock, S., & Graves, S. B. (2004). Understanding Shareholder Activism: Which Corporations are Targeted? Business & Society, 43(3), 239–267.
doi:10.1177/0007650304266869
Rehbein, K., Waddock, S., & Graves, S. B. (2004). Understanding Shareholder Activism: Which Corporations are Targeted? Business & Society, 43(3), 239–267.
doi:10.1177/0007650304266869
Reid, E. M., & Toffel, M. W. (2009). RESPONDING TO PUBLIC AND PRIVATE
POLITICS : CORPORATE DISCLOSURE OF CLIMATE CHANGE STRATEGIES. Strategic Management Journal, 30, 1157–1178. doi:10.1002/smj
