Introduction

Institutional pressures play an important role in explaining the proliferation of corporate social responsibility (CSR) in companies facing social and environmental demands from a variety of stakeholder groups (e.g. government authorities, industry organisations, and NGOs) (Bitzer and Glasbergen 2010; Campbell 2007; Hoffman 2001; Joyner and Payne 2002; Matten and Moon 2008; Quazi 2003; Wright et al. 2007).Footnote 1 In general, institutional pressures are understood as: ‘social, legal, and cultural forces outside the firm that exert influence on how managers perceive the environment and eventually shape and determine strategic actions’ (Menguc et al. 2010, p. 285). There is, for instance, evidence of social and environmental pressure from customers to suppliers (Baden et al. 2009) and from investors in the growing socially responsible investment (SRI) industry to the companies in which they invest (Buchholtz and Carroll 2009). At the same time, a wide range of new labels, certifications, guidelines, and multi-stakeholder initiatives has also created an infrastructure for CSR that puts pressure on companies to address the societal impacts of their operations (Waddock 2008). Likewise, governments have launched a large number of ‘soft’ and ‘hard’ policies intended to promote CSR, ranging from awareness-raising campaigns and capacity-building to incentive schemes and legislation (e.g. CSR reporting) (Albareda et al. 2007; Lozano et al. 2008; Steurer 2010).

In general, institutions can be seen as resilient social structures that ‘provide stability and meaning to social life’ (Scott 2010, p. 12). Traditionally, institutional studies of CSR have emphasised tendencies towards a homogenisation process by which companies assumedly conform to changes in the institutional environment (Campbell 2007; Jones 1999; Matten and Moon 2008). Less effort has been made to study the multiple ways in which companies can respond to institutional pressures for CSR; for example, some may adapt more or less unconsciously to new stakeholder demands, whereas others will try to hide non-conformity behind flashy CSR policies (Owen et al. 2001). Even companies operating within the same institutional setting may respond differently to CSR pressures (Bebbington et al. 2009; Greening and Gray 1994; Greenwood and Hinings 1996; Oliver 1991). For instance, Sharma and Vredenburg (1998) identify a variety of pro-active and reactive environmental strategies within the oil and gas industry. Likewise, Criado-Jiménez et al. (2008) argue that Spanish companies respond differently to new requirements for CSR reporting, while Spar and La Mure (2003) report that companies have chosen different strategies in response to activist pressures from the Free Burma Coalition.

This paper explores how institutional pressures for CSR are translated into different company strategies, and seeks to identify the underlying factors that determine the diversity of responses. More specifically, it examines the relationship between institutional CSR pressures and strategic responses among Nordic fashion companies (i.e., clothing companies in Denmark, Norway, Sweden, Finland, and Iceland). We choose the fashion industry because clothing is an important industry, has significant social and environmental impacts and is reported to be increasingly engaged in CSR activities (Allwood et al. 2006; DEFRA 2010, p. 4; MISTRA 2010, p. 7). In addition, little research has been done on CSR among a larger population of fashion companies, even though references to, and case studies of, individual fashion companies are common in the CSR literature (e.g. Primark, H&M, Gap, Adidas, Reebok, C&A, Gildan, Levis, Stormberg, and Nike) (Amazeen 2011; Ansett 2007; Arnold 2003; Turcotte et al. 2007; DeTienne and Lewis 2005; Doorey 2011; Frenkel and Scott 2002; Graafland 2002; Hoivik and Melé 2009; Islam and Deegan 2010; Jones et al. 2009; Knight and Greenberg 2002; Locke and Romis 2007; Wright et al. 2007; Yu 2008).

To-date institutional theory has proven to be inadequate in explaining the heterogeneity found in organisations confronted with similar institutional pressures (Delmas and Toffel 2008); to fill this gap, our research contributes to the literature in several ways. By examining homogeneity/heterogeneity in responses to institutional pressures and the underlying determinants, we hope to make a small contribution to this discussion and more broadly to the on-going debate between structure and agency in institutional theory (Heugens and Lander 2009). In general, the structuralism found in much institutional theory is challenged if fashion companies sharing similar internal and external characteristics demonstrate a high degree of variance in their responses to CSR pressures. Conversely, the concept of agency lacks explanatory power if different types of fashion companies facing different institutional environments adopt similar CSR approaches.

Further, our analysis helps to explain the observable divergence of responses to stakeholder demands for CSR (McWilliams and Siegel 2001). Above all, it differs from existing institutional studies of CSR by moving beyond passive conformance across all institutional conditions and various types of resistance behaviour (Oliver 1991). In line with other streams of CSR literature, the analysis adds opportunity-seeking as a category of strategic response that covers pro-active attempts to use CSR initiatives to translate institutional pressures into business benefits and competitive advantages (Burke and Logsdon 1996; Porter and Kramer 2006). The inclusion of opportunity-seeking as well as conformance and resistance provides a more complete picture of the span of strategic responses available to companies facing institutional pressures. The empirical tests of the new framework also provide evidence of the prominence of various strategic responses within a specific field, an aspect that as yet has received far too little attention. Attempts to construct items based on Oliver’s (1991) original work, for example, are sparse (Clemens and Douglas 2005; Clemens et al. 2008).

Context: The Fashion Industry and CSR

The importance of the fashion industry in many countries is attested to by the numbers: worldwide, clothing and textiles represent approximately 7 % of world exports, and the sector employs more than 20 million people (Allwood et al. 2006; DEFRA 2010; MISTRA 2010). Fashion is also a highly globalised industry: the large majority of fashion manufacturing takes place in low-cost regions in Asia and Africa, with China as the biggest exporter (Langhelle et al. 2009; Laudal 2010). In fact, around 70 % of the clothes imported to the EU come from developing countries (Laudal 2010). Such globalisation, however, comes at a price: the global network of retailers, wholesalers, agents, contractors, and sub-contractors makes the fashion supply chain highly complex and difficult to control (Emmelhainz and Adams 1999; Giesen 2008; MISTRA 2010): ‘(…) the clothing industry is infamous for having supply chains that are difficult to keep track of. One pair of jeans may be ordered from a brand by a small, Norwegian retailer, the brand hires an agent to find a factory, a few large factories receive the order, and further subcontract it’ (Langhelle et al. 2009, p. 15).

Fashion is about both function and aesthetics; that is, we wear clothes to meet physical/functional needs, adhere to social norms, demonstrate power and indicate group relationships (Allwood et al. 2006). Increasingly, however, fashion also seems to be about environmental degradation, hazardous chemicals, low wages, violation of workers’ rights and child labour. In recent decades, as discussed in more detail below, a large number of fashion companies have been subject to stakeholder criticism about the societal impacts of their business activities.

In terms of environmental impacts, clothing has a relatively large environmental footprint compared to that of other products, although its impact depends, to a large extent, on the type of textiles used (Chapman 2010). For instance, the extensive use of pesticides and water makes conventional cotton production one of the most polluting crop yields in the world (Giesen 2008). Its negative impact is amply illustrated by the ecological disaster in the Aral Sea, the capacity of which has been reduced 40 % by the massive amounts of water used for cotton plants (Allwood et al. 2006; DEFRA 2010; Fletcher 2008; Giesen 2008). Chemicals used in clothing and textile manufacturing also cause serious harm to the local communities and the environment. Hence, Greenpeace (2011a, b) recently launched a campaign against some of the world’s largest fashion companies after hazardous chemicals were found in a number of major brands.

Not only does fashion create negative environmental impacts when transforming raw materials to final product, but the use phase also generates a number of negative environmental externalities. For instance, washing and tumble drying is a key contributor to fashion’s total environmental footprint. Fashion also creates a great deal of waste: each year, an average UK consumer sends 30 kg of clothing and textiles to the landfill, and only 15 % of disposed clothing and textiles are recycled and reused (Allwood et al. 2006). Likewise, in Sweden an average person discards approximately 22 kg of clothing and textiles every year, and only 17.5 % is recycled (MISTRA 2010).

In terms of the social impacts of the fashion industry, poor labour conditions at supplier factories constitute a serious problem (Allwood et al. 2006). Fashion manufacturing is, to a large extent, outsourced to low-cost regions, and even though labour costs constitute only a small part of the total retail price, an analogy has been drawn between the ideal fashion company and a ship that stops where wages are lowest (Giesen 2008; Langhelle et al. 2009). Because of the globalisation of the fashion industry and the widespread outsourcing from developing countries, abuses such as underpayment, poor health and safety standards, long working hours, physical and verbal abuse, as well as child labour are widely reported in the fashion supply chain (Giesen 2008). In fact, one Norwegian study concludes that within the 5 years preceding the research, nine out of 13 textile companies discovered problems with working conditions among suppliers (Blomgren 2011). As a result, a number of high-profile fashion brands have been attacked—whether or not they directly control the factories involved—for not using their bargaining power in the supply chain to solve social problems. The apparel and footwear industries, particularly, frequently appear in negative news reports on sweatshops and human rights abuses (Sethi 2003). The best known example is Nike, which in the last decade has been vigorously criticised by Non-Governmental Organisations (NGOs), university students and the media alike (ibid.). Even today, the fashion industry is regularly accused of poor social and environmental practices in such reports as ‘Offside’, ‘Fashion Victims’, ‘Let’s Clean Up Fashion’, ‘Killer Jeans’, and ‘Stitched Up’, which all address poor labour conditions among fashion suppliers in developing countries (Alam et al. 2008, 2011; Connor and Dent 2006; McMullen and Maher 2009; Riddselius and Maher 2010).

These negative social and environmental impacts, as well as widespread public criticism, have stimulated attempts to rethink the current way of doing business in the fashion industry (see e.g. DEFRA 2010, pp. 13–23). For instance, over past decades, a large number of CSR standards, organisations and initiatives have appeared (and sometimes disappeared), including Trendsetters, Apparel Industry Partnership, Better Cotton Initiative, World Wide Responsible Apparel Production (WRAP), Clean Clothes Campaign, Fair Wear Foundation, DressCode, Oeko-Tex, AFIRM, Made-By, Sustainable Apparel Coalition, and Better Cotton Initiative (Ählström and Egels-Zandén 2008; Turcotte et al. 2007; Emmelhainz and Adams 1999; Fletcher 2008; Frenkel and Scott 2002; Giesen 2008; Nordic Fashion Association 2009; Wright et al. 2007). Most recently, the UN and the Nordic Initiative Clean and Ethical (NICE) have joined forces to make the fashion industry the first-ever sector-specific initiative under the UN Global Compact.

A number of individual niche producers and large fashion companies have also taken action to address sustainability issues (Fisher et al. 2008). The literature cites a number of pro-active fashion companies although their actual commitment to CSR is often difficult to validate, including Marks and Spencer, Levis, Monsoon, Patagonia, People Tree, American and Efird (A&E), New Look, Nau, American Apparel, Kuyichi, Brown, Noir, Stormberg, and Misericordia (Bekefi and Epstein 2008; Fletcher 2008; Giesen 2008; Goworek 2011; Hoivik and Melé 2009; Park and Lennon 2006). These firms have introduced a variety of initiatives to deal with the negative social and environmental impact of fashion, for example, by introducing fur-free policies; promoting fair trade cotton; launching traceability programmes in the supply chain; experimenting with new, recycled and reused materials; estimating sustainability impacts of material choices; introducing social and environmental guidelines for suppliers; developing environmental performance measures; improving product packaging; introducing labelling schemes for environmentally friendly laundry and drying; developing training and education materials and establishing networks for sustainable clothing and textiles (DEFRA 2010).

Nevertheless, despite several initiatives to ameliorate the negative social and environmental impacts of fashion production and consumption, the fashion industry still faces significant social and environmental challenges. Ethical clothing, for instance, still only constitutes a small part of the total market for ethical products and services. Therefore, it remains highly relevant to examine how companies cope with the social and environmental challenges that emerge throughout the chain, from extraction of raw materials to final disposal/recycling/reuse of the clothing. There is a need for both small, incremental changes and larger, more transformational/systemic changes (Fletcher 2008, p. xiii) and for ‘hard’ technological solutions and ‘soft’ cultural changes (ibid, p. 43). Initiatives to create a new and more sustainable fashion industry involve, for instance, development of new materials and production technologies, substitution of hazardous chemicals with environmentally friendly alternatives, improvement of recycling and reuse of clothes, and changes in consumption behaviour (Allwood et al. 2006).

Institutional Theory and CSR: Perspectives on Strategic Responses to Institutional Pressures

The linkage between institutional pressures and strategic responses is a key issue in institutional theory. Traditionally, institutional theory has been accused of being rather deterministic and leaving little room for agency (Bitzer and Glasbergen 2010; Greening and Gray 1994; Matten and Moon 2008). Organisations are expected to become more homogeneous as a result of coercion from powerful actors (coercive isomorphism), imitation of other organisations which are perceived to be successful (mimetic isomorphism), and professionalisation of jobs and work methods (normative isomorphism) (DiMaggio and Powell 1983; Greening and Gray 1994; Heugens and Lander 2009; Tan and Wang 2011). In the literature, institutional pressures can be traced back not only to the concrete demands of market and non-market stakeholders such as customers, regulators and local communities but also to broader, more abstract categories such as ‘forces’, ‘templates’, ‘scripts’, ‘cultural repertoires’, ‘cultural frameworks’, ‘social facts’, and ‘shared meaning systems’, which have no easily identifiable sender (Greenwood and Hinings 1996; Hoffman 2001; Brønn and Vidaver-Cohen 2009; Delmas and Toffel 2008; Bebbington et al. 2009; Lee 2011; Oliver 1991).

However, even though the structuralist perspective on institutional theory remains popular, more agency-oriented researchers have stressed that the institutional environment is not a prison that makes agency impossible but may actually be seen as a springboard for a variety of organisational behaviours (Heugens and Lander 2009, p. 63). From this viewpoint, institutional pressures may inspire organisations to be either conforming/resistant, passive/active, preconscious/controlling, impotent/influential, or habitual/opportunistic (Oliver 1991). Hence, one company may consciously adopt a pro-active approach to institutional pressures, while another adheres slavishly to new expectations from the outside environment. Powerful organisations, such as multinational corporations (MNCs), may even be able to shape the institutional environment (Matten and Moon 2008; Tan and Wang 2011).

Oliver’s (1991) seminal article ‘Strategic Responses to Institutional Pressures’ makes a key contribution to the literature by making more room for agency in institutional theory. Oliver (1991) distinguishes among five types of strategic response, which represent increasing levels of resistance to institutional requirements: (1) Acquiescence refers to conscious or unconscious adherence to rules, values and norms within the institutional environment; (2) compromise represents partial conformance with institutional requirements through negotiations with the company’s stakeholders; (3) avoidance concerns company attempts to fend off the necessity of conformance; for example, by hiding non-compliance with institutional requirements; (4) defiance refers to companies that resist conformance by challenging the institutional rules and values; and (5) manipulation involves a company’s purposeful attempts to affect institutional demands and expectations. Oliver’s work has subsequently inspired a number of different studies and remains a widely cited contribution to the debate between structure and agency in institutional theory (e.g. Clemens and Douglas 2005; Clemens et al. 2008; Etherington and Richardson 1994; Goodstein 1994). Although, in general, these studies do not break radically with Oliver’s original contribution, they have stimulated dialogue on potential adaptations and regroupings of resistance strategies (Clemens et al. 2008; Clemens and Douglas 2005; Etherington and Richardson 1994). For example, there has been discussion on whether ‘manipulation’ can actually be described as the most active resistance strategy (Clemens and Douglas 2005, p. 121I).

In the CSR literature, researchers have generally paid relatively little attention to the institutional conditions that affect CSR (Campbell 2006, 2007; Jackson and Apostolakou 2010; Lee 2011). For example, according to Campbell (2007), little of the CSR literature explores ‘whether institutional conditions affect the tendency for firms to behave in socially responsible ways’ (p. 948). Increasingly, however, institutional theory seems to be taking hold in CSR research (Bertels and Peloza 2008; Campbell 2007; Doh and Guay 2006; Jones 1999; Li et al. 2010; Matten and Moon 2008). The studies often emphasise tendencies toward homogenisation, but there have also been some applications of Oliver’s (1991) work on strategic responses to institutional pressures; for example, in a study of how impending environmental regulation may inspire different company strategic responses within the steel industry (Clemens et al. 2008).

What seems to distinguish mainstream institutional theory from the CSR literature is that the latter is open to responses beyond conformance and resistance. In fact, over the years, a number of CSR scholars have discussed typologies for the span of CSR approaches (Aragón-Correa 1998; Bhattacharyya 2010; Buysse and Verbeke 2003; Henriques and Sadorsky 1999; Jermier and Forbes 2003; Garcés-Ayerbe et al. 2012; Lee and Rhee 2007; Noci and Verganti 1999; Räsänen et al. 1994; van Tulder et al. 2009). Even as early as the 1970s, McAdam (1973) distinguished among four social responsibility philosophies: fight all the way, do only what is required, be progressive and lead the industry. Later in that decade, Carroll (1979) identified four different categories of social responsiveness: reaction, defence, accommodation, and pro-action. More recently, van Tulder et al. (2009) have proposed a distinction among inactive, reactive, active and pro-/interactive CSR approaches, while Heikkurinen (2010) operates with five levels of CSR: passive, reactive, pro-active, entrepreneurial, and creative. Mirvis and Googins (2006) distinguish among five stages of corporate citizenship associated with a corresponding number of issue management approaches: defensive, reactive, responsive, pro-active, and defining. O’Higgins (2010) makes a distinction among sceptical, pragmatic, engaged and idealistic company configurations, while Lee (2011) lists four different CSR strategies: obstructionist, defensive, accommodative and pro-active. Additional CSR classifications developed by other authors include: ‘do minimum harm/do maximum good’ (Wheeler et al. 2003), ‘bolt on/built in’ (Grayson and Hodges 2004), and ‘responsive/strategic (Porter and Kramer 2006).

In general, whereas mainstream institutional theory seems preoccupied with the span between conformance and resistance behaviour, the CSR literature allows for more pro-active, value-creating responses to institutional pressures. It recognises, for instance, that companies may invest in beyond-conformance CSR in an attempt to reap first-mover advantages if they foresee stricter future institutional constraints that also represent business opportunities (Nidumolu et al. 2009). Misani (2010) also makes a distinction between convergent CSR (CSR practices already adopted by other companies in the industry) and divergent CSR (unique CSR used to gain competitive advantage). Therefore, it is proposed that opportunity-seeking—understood here as pro-active attempts to use CSR initiatives as a means to translate institutional demands into business benefits and competitive advantages—should be added as a new type of strategic response to institutional pressures. CSR not only involves conformance with requirements or fighting demands from institutional constituents, but it also recognises that some companies may move beyond conformance either because of ideological beliefs or because they believe in the business opportunity presented by being among the companies that define the industry’s future. In the words of Bekefi and Epstein (2008, p. 43): ‘Companies can become leaders in corporate sustainability by developing pro-active strategies that create opportunities and increase profits rather than using only reactive strategies that respond to government regulation, industry standards, or consumer protests.

Hypothesis Development

Institutional pressures from the environment influence perceptions and actions within the organisation (Menguc et al. 2010; Pache and Santos 2010). These pressures may be direct (e.g. NGO campaigning against the company) or indirect (e.g. NGO affecting business practices through political lobbying) (Chua and Rahman 2011), and they may also be explicit (e.g. new regulatory requirements) or implicit (e.g. generally accepted business principles within the industry). The responses to these diverse pressures are determined by a wide range of internal and external factors, including competitive advantage expectations, the degree of legal coercion, environmental uncertainty, the interconnectedness of the institutional field, and the diffusion of institutional expectations (Garcés-Ayerbe et al. 2012; Oliver 1991, pp. 167–168). For instance, Oliver (1991, p. 166) argues that conformance is more likely to occur when institutional pressures do not threaten the autonomy of the organisation. This argument is also found in the CSR literature, where it has been argued that top-down CSR requirements from buyers to suppliers may be counterproductive and foster resistance behaviour at the supplier level (Baden et al. 2009, pp. 432–433; Boyd et al. 2007, p. 346).

At the most general level, however, we expect conformance to be the dominant reaction to institutional pressures for CSR, whereas only few companies try to move significantly beyond or below requirements (Hypothesis 1). Without denying the role of strategic agency, we believe this to be well in line with institutional theory, since the perspective would lack much explanatory power if conformance were the exception, rather than the rule. If all possible behaviours (homogeneity/heterogeneity, conformance/resistance, harmony/conflict, compliance/non-compliance, etc.) are equally likely and accepted within the boundaries of the institutional perspective, the theory is effectively emptied of meaning and becomes virtually impossible to verify. Furthermore, if institutions broadly represent the ‘the rules of the game’ it is reasonable to expect an element of consensus even though some divergence can be expected. The hypothesis also seems to be in line with recent empirical evidence from the CSR literature. For instance, a study of environmental strategies among 197 Belgium companies indicates that the companies are reactive or they adopt a pollution-prevention approach, while only a few companies qualify as environmental leaders (Buysse and Verbeke 2003). Moreover, more recent management surveys indicate that only a minority of companies takes a pro-active/strategic approach to CSR (Deloitte 2011; McKinsey 2001). Our first hypothesis, therefore, can be summarised as follows:

Hypothesis 1 (H 1 )

Conformance behaviour will be the most common response to institutional pressures for CSR, and only a minority of companies will choose opportunity-seeking and resistance.

The intensity of pressures for CSR is also likely to influence the CSR strategy. For instance, over the years a number of NGO campaigns have been successful in shaping company policies and behaviour. In the context of fashion, when the CEO of Timberland suddenly received 65,000 emails from Greenpeace supporters as part of a campaign against the deforestation of the Amazon, it motivated the organisation to take concrete steps to improve the traceability of its leather sourcing (Swartz 2010; Greenpeace 2009). More generally, the anti-sweatshop campaigns against major fashion brands in the 1990s also instigated a number of companies introducing codes of conducts and other supply chain initiatives (Kozlowski et al. 2012, p. 18). Last, animal rights groups have long fought fashion companies and other organisations that directly or indirectly support the fur industry. In consequence, a number of fashion designers, manufacturers, and retailers are now taking steps to exclude fur products from their offerings to consumers.

We thus expect fashion companies that perceive strong CSR pressures from stakeholders to be more likely to be opportunity-seekers when it comes to CSR (Hypothesis 2). This hypothesis is in line with Lee (2011, p. 292), who argues that companies facing strong external pressure for CSR are more likely to be pro-active when it comes to CSR. We also believe this hypothesis to be well in line with existing empirical CSR literature. For instance, Kassinis and Vafeas (2006) find a positive relationship between pressures from community groups and plant level environmental performance among polluting industries in the United States. Likewise, Darnall et al. (2010) document a positive relationship between growing stakeholder pressures and pro-active environmental practices, although this relationship is moderated by company size. Garcés-Ayerbe et al. (2012) and Murillo-Luna et al. (2008) also show support for the hypothesis that increasing stakeholder pressures perceived by managers stimulate pro-active environmental strategies. Last, Buysse and Verbeke (2003) identify some, admittedly complex, linkages between perceived pressures from primary and secondary stakeholders and the adoption of environmental strategies. We propose:

Hypothesis 2 (H 2 )

Strong institutional pressures for CSR will increase the likelihood of opportunity-seeking behaviour.

Perceived stakeholder influence, in general, is expected to influence response strategies. For instance, a fashion company highly dependent on a single retailer is more likely to comply with the demands of this stakeholder compared to a company that sells its products through thousands of retailers. As noted by Pache and Santos (2010, p. 458): ‘When organizations depend on key institutional referents for resources, such as funds, staff, or license to operate, they are likely to comply with what these stakeholders expect from them to secure access to these key resources.’ As an example, Darnall et al. (2010) argue that small companies rely more on support from community stakeholders, thus becoming responsive to local concerns and demands. On the contrary, a company that consider a stakeholder group to be of marginal importance will instead be more likely to adopt resistance strategies because firms can choose strategies more freely when they feel less dependent on stakeholders for survival (DiMaggio and Powell 1983, p. 154; Oliver 1991, pp. 163–164).

Generally, we assume that opportunity-seeking companies are more likely to see stakeholders as being influential, in contrast to companies that pay less attention to stakeholders (Hypothesis 3). A company that is aware of the various voices from constituents is believed to be more likely to transform these signals from the environment into concrete actions. On the contrary, a company with a narrower stakeholder mind set is considered to be more likely to neglect or even develop an adversarial attitude towards concerns from ‘irrelevant’ stakeholder groups, which, in turn, are likely to stimulate resistance strategies. The hypothesis bears similarity to Henriques and Sadorsky’s (1999) study, in which findings suggest that companies that perceive stakeholders as being highly important are more likely to be environmentally pro-active, whereas companies that consider stakeholders to be less important are more often reactive in terms of environmental commitment. Thus,

Hypothesis 3 (H 3 )

High perceived stakeholder influence will stimulate opportunity-seeking, whereas low perceived stakeholder influence will increase the likelihood of resistance strategies.

The fact that institutional pressures may be inconsistent, conflicting or contending—which makes it difficult to meet one demand without ignoring another—gives the organisation an element of discretion in determining responses (Greenwood and Hinings 1996; Pache and Santos 2010). For instance, a MNC facing different home country and host country standards can choose between different response strategies and tactics (Tan and Wang 2011). In general, Oliver (1991, p. 162) expects that multiple conflicting demands will tend to inspire non-conformance behaviour. The absence of consensus and shared worldviews in the environment makes it difficult for organisations to conform to all requirements and may also increase internal awareness of an issue and the palette of strategic alternatives (Ibid. 163). In the case of fashion, an example could be that costly compliance with environmental regulation may conflict with the interest of the owners who want to increase profits. Moreover, environmental concerns of environmental NGOs may conflict with the priorities of fast fashion consumers who may be unwilling to compromise their consumer and consumption patterns for environmental friendliness. Last, the long, global and complex fashion supply chain involves many actors who may not all share an interest in sustainable fashion (Kozlowski et al. 2012, p. 25). In other words, the views of the institutional constituents cannot –at least not in the short run—be expected to go together. Therefore, the organisation may be inspired to consider non-conformance strategies, e.g. by conforming with requirements of one constituent while ignoring those of another. For instance, Oliver (1991) provides an example of an oil company that ignores public demands for cleaning up after an oil spill in order to please its shareholders.

It is worth noticing, however, that conflicts exist not only amongst stakeholder groups but also within stakeholder groups. The broad stakeholder categories may underestimate the heterogeneity of individual stakeholder groups (Kassinis and Vafeas 2006, p. 145). For instance, various consumer segmentation studies show that there is only a small group of dedicated ethical consumers, whereas the mainstream consumer attaches only relatively little importance to social and environmental issues (Cowe and Williams 2000). Likewise, a generic stakeholder group such as ‘the local community’ in reality has a wide range of interests that are not necessarily aligned. However, thus far, the role of within-group conflicts on strategic responses to institutional pressures has rarely been explored. Kassinis and Vafeas (2006) conclude, for instance, that heterogeneity within stakeholder groups (income, pro-environmental preference and population density) has an impact on the environmental performance.

In sum, we hypothesise that companies will be more inclined to adopt non-conformance when stakeholder groups give CSR different priorities. Moreover, the likelihood of non-conformance strategies is also expected to increase when there is little consensus about the importance of CSR within the individual stakeholder group (consumer, shareholder, or employees). In other words, heterogeneous institutional pressures are expected to promote heterogeneity in corporate responses.

Hypothesis 4 (H 4 )

Conflicting institutional pressures between and within stakeholder groups will increase the likelihood of non-conformance response strategies, whereas consistent institutional pressures will promote conformance behaviour.

An overview of the relationship between institutional pressures and strategic responses is presented in Fig. 1. In short, stakeholder orientation (H3), CSR pressure intensity (H2), and CSR pressure consistency (H4a and H4b) are all expected to shape strategic responses to institutional pressures. Overall, however, it is assumed that conformance remains the dominant CSR response strategy (H1). In the following section, we will describe how we have measured the key concepts.

Fig. 1
figure 1

A strategic response model to CSR pressures

Method

Sample and Data Collection

The analysis is based on survey responses from a random sample of 400 fashion companies based in Nordic countries (Denmark, Norway, Sweden, Finland, and Iceland) that have more than five employees. A sample overview of the companies participating in the survey is provided in Table 1. The companies have been selected from publicly available databases in the five countries. The data collection was carried out using telephone interviews during autumn 2011 by an external data provider that also translated the survey into local languages. Since most of the companies had no CSR department, the target group identified for the survey was top management or, alternatively, staff responsible for personnel/HR and environment. The overall response rate was 20.48 %,Footnote 2 and company selection was designed to ensure a 50–50 split between fashion creators (designers/manufacturers) and fashion sellers (agents, wholesalers/retailers). In the sample used for analysis, however, 20.2 % of the companies were involved in many different steps of the value chain (e.g. designing clothing products and selling it through owned retail stores), and are thus labelled ‘mixed’

Table 1 Sample overview

Measuring Institutional Pressures

Translating the complex phenomenon of institutional pressures into a limited number of measurable indicators raises several issues. For instance, it is sometimes unclear in the literature how concrete pressures from identifiable organisations are distinguished from broader and more abstract institutional pressures. According to Lee (2011), institutional theory tends to focus on a supra-individual level of analysis while still empirically assuming that institutional forces are channelled by stakeholders. In general, a stakeholder can be defined as: ‘any group or individual who can affect or is affected by the achievement of the organization’s objectives’ (Freeman 1984, p. 46). As regards the identification of the stakeholders that exert pressure on the company, existing measures of institutional pressures are generally adapted to the research topic under study, meaning that stakeholder differences can be studied as sources of institutional pressures. Thus, when studying coercive pressures, Liu et al. (2010) look at customers and suppliers; Teo et al. (2003) focus on customers, suppliers and parent corporations; and Liang et al. (2007) emphasise local government, industry associations and competitive conditions. Likewise, in terms of normative pressures, Liang et al. (2007) consider customers, suppliers and government as relevant stakeholder groups; Teo et al. (2003) emphasise the role of suppliers, customers, and professional/trade/industry organisations; and Liu et al. (2010) look at suppliers, customers, and competitors. All three sets of authors, however, focus on competitors when studying mimetic pressures. The topic-specific approach to selecting stakeholders, therefore, although advantageous for aligning theory with the research question, makes it difficult to reach consensus on a commonly accepted and comparable approach to measuring institutional pressures.

This paper adopts a stakeholder-based approach to such measurement that has two expectations: (1) institutional pressures will be channelled through the company’s stakeholders and (2) failure to comply with rules, norms and habits will be sanctioned or rewarded by these institutional constituents. As noted by Lee (2011): ‘(…) stakeholders play a critical mediating role between institutional environments and organisations (…)’ (p. 287). The stakeholder groups included in this study are customers, owners, suppliers, employees, public authorities, the local community, and competitors. The selection of stakeholders is based on Clarkson’s (1995) description of primary stakeholders whose support is crucial for company survival. Evidence indicates that real-life managers also tend to give priority to these stakeholder groups when asked to articulate their views on the role of business in society (Pedersen 2010, 2011). Moreover, we add competitors to the list even though the literature has often ignored this stakeholder category (Spence et al. 2001). However, institutional theory considers competitors (at least the successful ones) as a potential source of imitation (mimetic isomorphism) which they will model themselves after (see above) (DiMaggio and Powell 1983, p. 152).

In the survey of Nordic fashion companies, institutional pressures for CSR were measured as the average pressure exerted by seven stakeholder groups operationalised as the following question: ‘To what extent do the groups below set social and environmental requirements (CSR) to the company?’ (on an 11-point scale from 0 = no demands and 1 = few demands to 10 = very high demands). The question was repeated for each stakeholder group and a total average was calculated by excluding ‘no demands’ stakeholders and adjusting for the number of stakeholders included.

This total average pressure served as the basis for testing our H2. For each stakeholder category, the respondents were also asked to indicate company-stakeholder dependencies; that is, the relative importance of the stakeholder to the company. More specifically, respondents were asked the following question, which was repeated for each stakeholder: ‘To what extent are the groups below able to affect the company?’ (on an 11-point scale of 0 = no influence and 1 = little influence to 10 = very significant influence). Stakeholder influence was then measured by the average influence exerted on the company by all stakeholders excluding those with no influence. If stakeholder influence was present, then we investigated whether a high perceived average influence would lead to opportunity-seeking behaviour (see H3).

To obtain a measure of the consistency of institutional pressures, respondents were asked to what extent each stakeholder group’s social and environmental requirements were in sync with other expectations (e.g. price, quality), as measured by the following question: ‘To what extent are [stakeholder requirements] consistent with other requirements (e.g. price, quality)?’ (on a 10-point scale from 1 = no/very little consistency to 10 = complete consistency). A total score for within-group consistency was calculated based on the average consistency across all stakeholders, with a low value signalling little average consistency and a high value, high average consistency. For between-group consistency, we calculated a score reflecting the range from minimum to maximum stakeholder pressure for a company by subtracting the minimum perceived stakeholder pressure from the maximum perceived stakeholder pressure across all stakeholders. A low value means more consistent pressure and a high value more conflict. We then used this measure of within-group consistency and between-group consistency to test H4. In the case that a stakeholder exerted no pressure on the company, within-group and between-group consistency were not considered.

Measuring Strategic Responses to CSR Pressures

In this paper, we propose a new scale of strategic responses to institutional pressures that distinguishes between three broad groups of strategic responses: (1) resistance (attempts to avoid compliance with institutional pressures), (2) conformance (adaptation to institutional environment requirements), and (3) opportunity-seeking (a beyond-conformance behaviour that exceeds external expectations). The model thus moves beyond the previous categorisations of strategic responses in institutional theory which tend to cover only conformance and resistance strategies. It is, however, very much in line with the categorisations found in the CSR literature (see earlier section). Our three categories can then be sub-divided into a number of more specific strategies: rejecters (non-compliers), negotiators (partial compliers), conformists (compliers), anticipators (prognostic compliers), and definers (innovative CSR). Grouping the responses of companies into five strategies does not try to underestimate the actual complexity in behaviours when it comes to CSR. It is important to note that the same company may adopt multiple strategies in response to institutional pressures from different stakeholders (Clemens and Douglas 2005). Some companies may adopt a negotiator strategy towards one stakeholder and a conformist strategy towards another. Moreover, responses may also depend on the CSR issue in question. A company may go to great length to fight social exclusion, while at the same time being more or less indifferent when it comes to reducing environmental footprint.

Table 2 Strategic Responses to CSR pressures

In this study, corporate response to CSR pressures mirrors the most frequent response to stakeholders that the respondents themselves deem to be relevant. The respondents indicated their response strategy by answering the following question for each stakeholder: ‘If demands are made for social and environmental responsibility (CSR), what is the company’s typical reaction?’ Respondents had to choose one of five responses corresponding to the five categories outlined above: (1) We try to go beyond existing and planned CSR requirement, (2) We try to predict CSR requirements and meet these beforehand, (3) We try to live up to all the CSR requirements that we are faced with, (4) We try to relax CSR requirements to the extent possible, and (5) We try to avoid having to meet CSR requirements to the extent possible.

To build a dependent variable on response strategy, we excluded stakeholders that had no perceived influence on the company. We then counted how often a response strategy occurred for a company and assigned the most often-used response strategy to CSR pressures as the dominant response for that company. In 30 cases, the most frequent response was split equally between two or more categories, and thus we weighted the response to CSR pressures by the influence of the individual stakeholder. We could then determine the dominant response strategy based on stakeholder importance. This procedure generated six variables: one ordinal that included all strategic responses from −2 rejecters and −1 negotiators over 0 conformers to 1 anticipators and 2 definers, and dummies for each of the five individual response strategies. It also identified response strategies for 340 companies, descriptive statistics for which are given in Table 3.

Table 3 Descriptive statistics by total sample and by stakeholder

Measuring CSR

Based on our assumption that opportunity-seeking fashion companies will demonstrate higher levels of CSR than do resistant organisations, we added CSR performance as an extra variable to test response strategy. Measuring CSR, however, creates a number of challenges. First, because CSR has often been criticised as a blurry concept (Gjølberg 2009a; Lozano et al. 2008; van Tulder et al. 2009), it has been difficult to reach agreement on a commonly accepted list of indicators that can be used to quantify and measure it (Jackson and Apostolakou 2010). In the words of Gjølberg (2009a), ‘CSR is, due to the definitional disagreements in academia and the wide variety of practices labelled “CSR” in the corporate world, an elusive concept which to a certain extent defies quantification’ (p. 20). Hence, in their attempts to operationalise CSR and related concepts, academics and practitioners include different social and environmental dimensions (Accenture 2010; Castka et al. 2004; Chan 2010; Deloitte 2011; Haanaes et al. 2011; Husted and Allen 2006; Jackson and Apostolakou 2010; Li et al. 2010; Park and Stoel 2005; SHRM 2007).

In this study, respondents were asked to indicate the company’s CSR effort along five dimensions: (1) environmentally friendly products, processes, and/or services; (2) employee health, safety, and well-being; (3) customer health, safety, and well-being; (4) social and environmental conditions in the supply chain (e.g. code of conduct, social auditing); and (5) philanthropy and investments in the local community. These dimensions were measured by the following question: ‘How would you evaluate the company’s effort (policies, activities, documentation, communication) within the following areas?’ (on an 11-point scale from 0 = no initiatives and 1 = very little effort to 10 = very comprehensive efforts). We then analysed the intensity of CSR adoption among Nordic fashion companies by adding up all five dimensions to produce a CSR score that varied from 0 = no initiatives to 50 = very comprehensive efforts. This score enabled us to derive a measure of strategic responses to CSR pressures for the five dimensions individually and in total.

Control Variables

Although the likelihood that size influences strategic responses to institutional pressures is debatable, the greater visibility of large companies on the business landscape may make them more inclined to conform to institutional pressures in order to avoid damage to reputation and brand (Clemens and Douglas 2005). They may also have more resources to invest in CSR, and the scale of their operations facilitates the establishment of a formal CSR structure (Jackson and Apostolakou 2010). For instance, a Nordic fashion giant like H&M can more easily hire a CSR expert compared to an SME fashion designer that is still struggling to gain a foothold in the industry. On the other hand, large powerful organisations may also have more resources for resisting stakeholder pressures for more CSR, suggesting that they may not be the first to adopt a pro-active approach to CSR. Smaller companies, in contrast, may be more dependent on support from local stakeholders (Darnall et al. 2010). Likewise, larger organisations are often more mechanistic and rigid, whereas smaller companies are less bureaucratic and more innovative (Darnall et al. 2010; Spence and Bourlakis 2009). In terms of CSR, large companies are more likely to use formalised and codified CSR instruments, standards and agreements; however, small and medium enterprises (SMEs) tend to work more informally (Fitjar 2011; Graafland and van de Ven 2006). All these observations imply that large companies may adopt more passive strategies, whereas smaller enterprises may demonstrate a higher level of discretion by choosing more active strategies (Clemens et al. 2008). Whichever the perspective, size, measured by number of employees, can be expected to shape the type of response chosen by Nordic fashion companies; accordingly, it was included in this study as a control measure.

Another factor that may potentially play a role in the choice of strategic responses is company age. Because organisations at the beginning of their life-cycle are likely to be driven more by efficiency than by legitimacy concerns (DiMaggio and Powell 1983), relatively young organisations can be expected to demonstrate more response variety than do well-established organisations that are more likely to conform to institutional pressures within the field. We therefore introduced company age into the analysis as a control variable that might influence how fashion companies respond to institutional pressures for CSR.

We also controlled for country differences in CSR responses. Despite talks of a Nordic approach to CSR (Albareda et al. 2007; Gjølberg 2009b; Midttun et al. 2006), research evidence also exists for differences in CSR approaches among Nordic countries. For instance, Gjølberg (2010) argues that the CSR approaches of Nordic governments differ in focus and underlying justification. Likewise, a recent survey concludes that Sweden ranks significantly higher than does Denmark in the quality and maturity of CSR reporting (KPMG 2011). We therefore controlled for the impact of country differences by grouping the fashion companies in the five Nordic countries according to response strategy patterns while also taking into account the smallness of the sub-sample for Iceland (n = 40). An ANOVA and corresponding post estimation (Wald test) clearly indicated that Sweden and Finland showed similar patterns of response strategies and differed statistically significantly from Denmark, Iceland, and Norway. We therefore created a dummy variable for which a value of 1 represented Sweden and Finland and a value of 0, Denmark, Iceland, and Norway.

Analytical Results

We estimate our regressions using the following model, in which the strategic response to CSR pressures is the dependent variable and variables along the hypotheses serve as independent variables:

$$ {\text{SR}} = \beta_{0} + \beta_{ 1} P + \beta_{ 2} {\text{SI}} + \beta_{ 3} {\text{SC}} + \beta_{ 4} {\text{SV}} + \beta_{ 5} {\text{CO}} + \beta_{ 8} {\text{C}} + \varepsilon $$

where SR is a matrix for both our measure of strategic response to CSR pressure and CSR performance and may be either ordinal, binary (dummies) or metrically scaled. P denotes institutional pressure, SI is a measure of average stakeholder influence, SC is a matrix of within- and between-stakeholder consistency, SV is stakeholder view, and CO is a country dummy. C is a matrix of control variables (size and age). ε is a matrix of idiosyncratic error terms, and the βs are the coefficients to be estimated.

Depending on the nature of SR, we employ either an ordered probit, logit or OLS model. In the first stage, we investigate the effects of the independent variables on all response strategies (ordinal variable = probit model). In the second step, we estimate the effects of the independent variables on each individual response strategy (e.g. conformism) against all others (dummies = logit models). In the third step, we estimate the effects of the set of independent variables on CSR performance (OLS). We also carry out a fourth step to compare the response strategies of conformers versus opportunity-seekers, conformers versus resisters and opportunity-seekers versus resisters (logit model). The coefficients’ (βs) interpretation depends on the analysis method. For ordered probit and OLS, a coefficient close to zero means no effect on the strategic response, a positive value means a positive effect and so forth. This is different with regard to the logit models, presented coefficients represent the so-called odd ratios, where a value of one means no effect, above one a positive effect on the response strategy, and below one a negative effect. All analyses are performed using STATA SE 11.2.

Table 3 presents the descriptive statistics and outlines the key variables. The results clearly indicate that conformance is, in fact, the most common response to CSR pressures, regardless of whether the unit of analysis is the total number or individual stakeholder groups. H1 is thus supported. Nevertheless, strategic responses to institutional pressures overall seem rather stakeholder insensitive, although differences do emerge, such as between public authorities and competitors. Even more noteworthy, Table 3 indicates that it is primarily owners who set CSR requirements and their pressures are generally more consistent and in alignment with company strategy than are the pressures from other stakeholders. These results somewhat challenge the conventional view that owners are interested mainly in profits and that CSR is driven primarily by pressures from such external parties as customers, community groups and regulators.

The results for the regression analyses of strategic responses are given in Table 4, providing the coefficients and cut-offs for all five response strategies (column 1), the odds ratios for the individual responses against all other responses (columns 2–6) and the OLS estimates for CSR performance (column 7). Figure 2a, b then plots the odds ratios (from table columns 2–6) and their corresponding confidence intervals for the variable of most interest, institutional CSR pressure and the consistency within stakeholder groups. Both Table 4 and Fig. 2a show that increased pressures for CSR encourage opportunity-seeking strategies. The data thus support H2. The results thereby indicate that increased pressures, strong signals from stakeholders, lead companies to take a pro-active stand on CSR. This observation is further supported by the existence of a positive relationship between perceived institutional pressures and overall CSR performance.

Table 4 Estimates on CSR response and CSR performance
Fig. 2
figure 2

a Institutional CSR pressure: Odds ratios for strategic CSR response with 95 % confidence interval. b Consistency within stakeholder: Odds ratios for strategic CSR response with 95 % confidence interval

As regards H3, stakeholder influence in general seems unrelated to response strategy, as the growing influence of stakeholders in general does not seem to have an impact on whether companies choose an opportunity-seeking, conformance or resistance strategy. Although there is a positive relationship between stakeholder influence and overall CSR performance (see Table 3, column 7, OLS estimates), the analytical results provide no support for H3, which predicts that increased stakeholder influence will stimulate conformance behaviour.

In terms of H4, the level of conflict between stakeholder groups does not seem to have an impact on response strategies, although conflicts between stakeholders overall seem to have a positive impact on CSR performance. Hence, it is impossible to identify a clear pattern of between-group consistency and strategic responses. However, consistency within stakeholder groups does seem to have an impact on CSR responses. In line with H4, increasing consistency within stakeholder groups motivates fashion companies to conform to requirements, thus reducing the likelihood of their being opportunity-seekers and resisters (see Fig. 2b). To give a concrete example, when different customer segments are consistent in their social and environmental demands, the company seems to be more inclined to conform to these requirements. On the other hand, mixed signals from a stakeholder group may influence the company to consider more active strategies (opportunity-seeking or resistance). The analysis also shows that the overall CSR performance of companies experiencing high levels of between-group and within-group consistency is relatively high. Apparently, consistent stakeholder requirements and the tendency to favour conformance over opportunity-seeking and resistance tend to drive up average CSR performance. Hence, overall, the analysis does support H4 in terms of within-group consistency.

The difference between fashion companies that select different response strategies does becomes more apparent in Tables 5 and 6. Table 5 shows the mean values for each independent variable by response strategy, categorised into three groups—conformists, opportunity-seekers, and resisters. This table also reports the results of t tests of the conformist group against one of the other groups, indicating that, in contrast to conformists, opportunity-seekers perceive significantly higher institutional pressure and less than average consistency within stakeholders. In terms of resisters versus conformists, however, only the average consistency within stakeholder groups is significantly lower, most probably due to the small number of observations for resisters (n = 17). As for Table 6, different strategies are compared against each other, further supporting the finding that increasing CSR pressures promotes opportunity-seeking strategies, whereas within-stakeholder group consistency induces conformance behaviour.

Table 5 Descriptive statistics: conformists against opportunity-seekers and resisters along the hypotheses
Table 6 Logit estimates—CSR responses against each other

Discussion of Findings

The results of our analysis largely support institutional theory by documenting conformance to stakeholder pressures as the dominant organisational response among Nordic fashion companies, regardless of stakeholder group (customers, owners, employees, suppliers, governmental authorities, or competitors). The majority of companies see CSR responses to institutional pressures as largely a compliance game rather than a strategic opportunity to differentiate the company from its competitors. This finding is in accordance with a recent McKinsey (2001) survey, which concluded that a few companies (around 30 %) are pro-active in managing sustainability. Interestingly, conformance seems to be the most common strategy, irrespective of stakeholder group, indicating that company responses to institutional CSR pressures seem to be less dependent on the identity of the constituent formulating the demands and requirements. The results raise the question as to whether future research should pay more attention to the nature of institutional pressures (coercive/non-coercive, direct/indirect, explicit/implicit, positive/negative, etc.) rather than to their origin.

The results of our study among Nordic fashion companies indicate that increased pressures for CSR stimulate opportunity-seeking behaviour. The finding expands the institutional literature’s general expectation that firms are more likely to behave responsibly if they experience strong pressure from regulators, NGOs, industry organisations and so forth (Campbell 2006, pp. 934–935). In other words, we find that some companies use social and environmental pressures as an opportunity to improve social and environmental performance beyond what stakeholders require. Similar results are observed in other contexts. For instance, Murillo-Luna et al. (2008, p. 1238), based on an analysis of 240 industrial firms, argue that ‘(…) the greater the environmental demand perceived by managers, the more solutions the firm tends to adopt beyond the mandatory environmental requirements established by the authorities, and even beyond market or society expectations.’ These results may help to explain why some companies become very active within the field of CSR after experiencing significant pressures from stakeholders. The Brent Spar and later the Niger Delta incident, for example, are thought to have inspired self-reflection and organisational changes in Shell (Jermier and Forbes 2003; Preuss 2011, p. 24).Footnote 3

With respect to the criticisms of Oliver’s (1991) typology as lacking ‘(…) predictive power when discussing responses to conflicting demands (…)’ (Pache and Santos 2010, p. 456), our results indicate that stakeholder conflict/consistency does in fact influence the choice of strategic responses. In particular, consistency within stakeholder groups seems to inspire conformance behaviour. Conformance strategies are less common when stakeholder demands are inconsistent, possibly because such inconsistency gives rise to a variety of responses, that is, mixed signals from stakeholders can be subject to multiple interpretations, and companies experiencing stakeholder inconsistency between CSR requirements and other demands may choose from a broader arsenal of response strategies. The finding bears similarities with Goodrick and Salancik (1996) who argue that growing uncertainty of institutional standards increase discretion to choose between different practices.

What are the implications of the results from this study? First, the conformance-mentality in the fashion industry makes it questionable whether systemic changes will take place unless multiple stakeholders (policy makers, media, NGOs, consumer groups, etc.) are able to mobilise significant pressure for sustainable fashion. At least, this will be the case unless the leading players in the industry (such as H&M, Bestseller and IC Company) all adopt pro-active CSR approaches that are subsequently imitated by the jungle of smaller fashion designers, manufacturers and retailers as a standard reaction to uncertainty (DiMaggio and Powell 1983). The good news, however, is that the dominance of conformance strategies may create market opportunities for innovative fashion companies that are trying to differentiate themselves by addressing CSR.

Second, the results from the study highlight the importance of stakeholder activism in promoting CSR as increasing pressures seem to stimulate beyond-conformance behaviour. Therefore, the stakeholder environment does seem to play an important role in promoting CSR behaviour, and more concerted efforts by multiple actors (policy makers, media, NGOs, etc.) may be a way forward in promoting sustainable changes in the fashion industry. If more stakeholders set CSR demands, more companies within the fashion industry are likely to consider alternative designs, input materials, manufacturing methods, distribution channels, etc. Individually, codes of conduct, public procurement, responsible leadership training and education, NGO activism, ethical consumerism, etc., may be insufficient in transforming dominant business practices but collectively these efforts may heighten the industry awareness of social and environmental issues and raise the bar for CSR.

Third, the results highlight the importance of internal stakeholder consistency when it comes to promoting conformance with CSR requirements. This is a challenge for the future of sustainable fashion, as there is still only a minority of consumers demanding sustainable fashion. Even within the market of ethical consumerism, ethical clothing still seems to play only a marginal role (Co-operative Bank 2010). Moreover, it has been argued that companies setting CSR demands for their own suppliers still give priority traditional buyer requirements (prize, quality, and delivery time) (Baden et al. 2009). Thus, the future of sustainable fashion seems to require an element of consistency in the minds and deeds of the key stakeholder groups. To promote consistency, CSR has to be better embedded into existing and new technologies, structures, systems, and mind-sets.

Conclusions and Limitations

In the face of a growing recognition in CSR research into the role of institutional factors (Li et al. 2010, p. 639), this paper aimed to deepen understanding of CSR behaviour in institutional contexts. More specifically, it explored how fashion companies in Nordic countries respond to pressures from key institutional constituents. The paper therefore went beyond institutional theory’s usual typologies of strategic responses that focus on conformance and resistance behaviour by also paying attention to opportunity-seeking behaviour, including pro-active attempts to translate institutional pressures into strategic benefits. The inspiration for expanding the model of strategic responses to include opportunity-seeking behaviour was the CSR literature, which is rich in typologies of CSR strategies that range from being reactive to pro-active. The CSR literature has long recognised that companies may adopt a more pro-active approach to institutional pressures rather than simply meeting or resisting expectations from the outside environment.

Methodologically, we formulated and tested a number of hypotheses on the relations between institutional pressures and strategic responses (resistance, conformance, and opportunity-seeking). Generally speaking, we anticipated that strategic responses to CSR would be dependent not only upon the source of the pressure, but also its intensity and consistency. This perspective bears similarities to that of Oliver (1991), who argues that, ‘[o]rganizational responses to institutional pressures toward conformity will depend on why these pressures are being exerted, who is exerting them, what these pressures are, how or by what means they are exerted, and where they occur’ (p. 159).

The results of our study indicate that CSR remains a compliance game in the Nordic fashion industry: fashion companies tend to conform to stakeholder requirements rather than considering resistance and opportunity-seeking strategies. Hence, conformance remains the dominant response category, irrespective of stakeholder group. However, as stakeholder pressures increase, more companies are considering opportunity-seeking as a response to stakeholder demands. The analytical results also indicate that companies experiencing consistent demands within stakeholder groups tend to select conformance strategies, implying that consistent stakeholder requirements make conformance a safe haven for fashion companies.

The paper of course has limitations. Most particularly, it develops a stakeholder-oriented approach to measure the strategic responses to institutional pressures. The measurement framework has at least two limitations. First, the list of potential stakeholder groups as mediators of institutional pressures could potentially be much longer (see e.g. Henriques and Sadorsky 1999; Hoffman 2001). However, all models are simplifications of what they model (Jackson and Carter 2000), and a trade-off must be made between the completeness of a measurement tool and its practical usability (as well as the costs of implementation). Second, as previously pointed out, the stakeholder approach to the measurement of institutional pressures is not necessarily sufficient to capture broader, taken-for-granted institutional pressures at the macro-level. For example, Schaefer (2007) notes that coercive isomorphism can be the result of both ‘(…) pressures from other organisations on which a focal organisation is dependent and by pressures to conform to cultural expectations of the larger society’ (p. 508). The measurement framework used in this study is more suited to capturing the first type of pressure rather than to the second. In addition, just like a company can adopt different response strategies, stakeholders are not just passive transmitters of institutional pressures but may have their own agendas which create ‘noise’ when analysing the linkages between institutional pressures and company responses.

In addition, because the survey explicitly addressed CSR, there is a risk that companies with few or no CSR activities may have been unwilling to answer the survey, just as responding companies may be inclined to give socially desirable responses (Weaver et al. 1999). To minimise these potential biases, the questions for the survey were phrased as neutrally as possible. The survey was also designed so that the questions explicitly addressing CSR issues were intentionally placed in the last section, with the business environment, and stakeholder relations in general addressed in the first section. It is nevertheless impossible to rule out the possibility that some companies may have over-reported strategic responses to CSR pressures. For this reason, as an additional test of response, we also included a question on CSR performance along five dimensions. Nevertheless, even though inactive companies may be under-represented in the study, the survey data indicate that the majority of respondents reported compliance rather than opportunity-seeking behaviour and that most companies seem to have only relatively low CSR performance.

Finally, the study focuses only on a single industry and has limited geographic coverage. As regards the former, although a single industry focus has previously been applied in the study of strategic responses to institutional pressures, certain industries—because they have particular characteristics—may favour certain types of strategic responses to institutional pressures (Clemens and Douglas 2005). For instance, high-impact industries in general tend to be more active on the CSR agenda (Jackson and Apostolakou 2010). In terms of geography, organisational practices (including CSR) can be expected to vary across countries and cultures because of differences in institutional structures (Campbell 2007; DiMaggio and Powell 1983; Gjølberg 2009a; Kostova and Roth 2002; Matten and Moon 2008; Tan and Wang 2011). Hence, companies in the Nordic region may share some characteristics that are not easily transferred to other geographic settings. For instance, Nordic countries have been characterised as strong welfare states that tend toward a consensus orientation, cross-sector collaboration and partnerships (Albareda et al. 2007; Gjølberg 2010). As regards CSR, companies in Nordic countries are generally recognised for their contributions to sustainable development (Birkin et al. 2009) and their relatively high scores on SRI evaluations, CSR reporting, CSR memberships and CSR management standards, which may result from the homogeneity of institutional traditions in these well-advanced welfare states (Midttun et al. 2006).

It should also be noted that strategic responses are not set in stone; rather, they are fluid, which makes it difficult to develop a single measure for a fashion company’s response to CSR pressures. For instance, resistance in one area may be contrasted with opportunity-seeking in another area. As Lawrence (2010) notes in his examination of company responses to demands from non-market stakeholders, it is possible to find elements of different strategies within the same company. Moreover, an examination of institutional pressures and strategic responses at a single period of time does not allow for the variation of CSR pressures over time (see e.g. Schaefer 2007). Nor does it account for changing strategies over time (Lawrence 2010), which may shift from refusal to conformance (Turcotte et al. 2007, p. 34), from scepticism to ambivalent acceptance (Frenkel and Scott 2002, p. 34), from ignorance to strategic compliance (Lee and Rhee 2005, p. 394) or from reactive to pro-active (Perez-Aleman and Sandilands 2008, p. 30). For example, Nike initially resisted stakeholder demands for transparency in the supply chain but later became one of the first major fashion companies to release a global factory database (Doorey 2011). In sum, the dynamics of institutional pressures and corporate responses cannot be covered in a one-time survey, meaning that more qualitative longitudinal studies are needed in the future to deepen our understanding of the relation between CSR pressures and corporate responses. The fast-growing literature on institutional logics may serve as a fruitful theoretical approach for a more fine-grained analysis of the dynamics between the characteristics of institutional pressures (complexity, ambiguity, incomparability, etc.), the type and evolution of organisational responses, and the mediating role of internal organisational structures/systems as well as external environmental conditions (see e.g. Greenwood et al. 2011).