Organizational responses to potential industry-level illegitimacy spillovers from scandals : a study of BMW-Group and diesel gate

Organizational responses to potential industry-level illegitimacy spill

overs from scandals – A study of BMW-Group and Diesel gate

Hidde Jeroen van Duuren Student ID: 10221379

19-08-2016 2015/2016

MSc in Business Administration - Strategy Track Universiteit van Amsterdam Supervisor: dr. A.E. Kourula

Statement of originality

This document is written by; Hidde J. van Duuren, 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 in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

Table of Contents

1.4 Objectives and structure of the paper ... 8!

2. Literature review ... 9!

2.1 Corporate social responsibility ... 10!

2.2 Legitimacy and illegitimacy ... 14!

2.3 Spillover effect ... 17! 2.4 Theoretical framework ... 18! 2.4.1 Forms of legitimacy ... 18! 2.4.2 Legitimation strategies ... 19! 3. Methodology ... 20! 3.1 Research design ... 20! 3.2 Sampling ... 20! 3.3 Data gathering ... 21! 3.4 Data analysis ... 24! 3.5 Evaluation of quality ... 26! 4. Results ... 26!

4.1 BMW-Group press releases ... 27!

4.1.1 Direct and indirect related ... 27!

4.1.2 Forms of legitimacy ... 28!

4.1.3 Legitimation strategies ... 29!

4.2. LexisNexis database ... 32!

4.2.1 Direct and indirect related ... 32!

4.2.2 Forms of legitimacy ... 32!

4.2.3 Legitimation strategies ... 33!

4.3 Additional unexpected findings ... 36!

5. Discussion ... 36! 5.1 Responses ... 37! 5.2 Forms of legitimacy ... 38! 5.3 Legitimation strategies ... 40! 6. Conclusion ... 42! 6.1 Summary ... 42!

6.2 Contribution and suggestions for further research ... 44!

References ... 46!

Appendices ... 49!

Apendix A – Analysis BMW-Group press releases ... 49!

Abstract

The Volkswagen diesel gate scandal had a massive impact in the international automotive industry. This scandal raised a more general question how organisations respond to these potential industry-level illegitimate spill overs. Despite the central role of legitimacy within organizational contexts, we know little of the subtle meaning-making processes through which potential industry-level illegitimate spill overs are legitimated in contemporary society (Vaara, Tienari & Laurila, 2006). To answer the question the design of this research has a qualitative approach, where a single case study on the BMW-Group and the diesel gate scandal is used to accomplish the partly explorative nature of this thesis (Eisenhardt, 1989, Saunders & Lewis, 2012). BMW-Group press releases and international news articles from LexisNexis database were analysed as responses to the scandal. Two forms of legitimacy, pragmatic and moral, received most attention and six strategies were used to retain organisational legitimacy. The main contribution is the opening of new thoughts to be further explored.

1. Introduction

1.1 Background

The ‘diesel gate’ scandal of the Volkswagen Group affected the German automotive industry hard and stock markets plummeted. It had this massive impact because it flew in the face of recent developments in the industry; in light of climate change and air quality concerns over the last decades, the principles of corporate social responsibility, and especially of environmental responsibility, have steadily acquired priority on the agendas of stakeholders. The Volkswagen Group, owner of Volkswagen, Audi, Porsche, Skoda, Seat, Bentley, Lamborghini, Bugatti, Scania and MAN, practiced environmental irresponsible behaviour by manipulating emission test of their vehicles – the facts hardly need reiteration. It was not just the blatant disregard of environmental standards, but the deliberate attempt to deceive that caused a worldwide uproar. The uproar, unsurprisingly, had serious economic consequences for the Volkswagen Group.

More surprising, however, was the fact that Volkswagen Group was not the only group affected by the scandal, although none of the other players were accused of complicity or of manipulating their own emission test. As it was, the whole automotive industry suffered for the illegitimate action of one industry player and all stocks dropped. As the manipulation of emissions became public and was dubbed the ‘diesel gate’ scandal, shock waves were sent not just through Volkswagen Group, but through the whole automotive industry – Volkswagen’s fraud caused problems of legitimacy on an industry-wide scale.

1.2 Research problem

This recent diesel gate scandal in the automotive industry is just one example of a wider phenomenon in which organisations are negatively affected by the actions of other organisations in the industry, but the organisations affected are without the means to directly tackle the actual problem(s) – they are not responsible, they were not actors, but they still suffer from the corrosion of legitimacy and its economic consequences.

These problems, of course, were caused by the illegitimate action of Volkswagen Group; their behaviour was clearly not aligned with the socially constructed system of norms, values, beliefs, and definitions (Suchman, 1995), not in line with principles of corporate social responsibility and environmental responsibility. Their illegitimate actions, their protrusive neglect of corporate social responsibility had far-reaching ramifications for the wider organisational field of the automotive industry.

These effects have been observed before in other cases; Haack claimed that in an organisational context, illegitimate action can not only create a spill over effect that can be both horizontally and vertically distributed to sub brands within the organisation (Haack et al., 2014); illegitimacy could also create a horizontal spill over to organisations active in the same organizational field – an effect visible in the ‘diesel gate’. This serious threat of potential illegitimate spill over whenever a scandal occurs, such as Volkswagen Group’s recent one, demands the attention of other players in the industry as they seek to mitigate or prevent the negative economic effects of these potential spill overs. Hence, it is interesting to study organizational responses whenever the threat of illegitimate spill over in the industry occurs.

In doing so this thesis will seek to contribute to Suchman’s (1995) legitimacy theory and tries to gain a better understanding of legitimation strategies (Vaara, Tienari & Laurila, 2006) – because it is in terms of legitimacy that a spill over effect occurs, a negative effect in our example. The changing conditions of globalization have put the two traditional modes of legitimacy, pragmatic and cognitive legitimacy, under pressure. Consequently, organisations are often required to establish a third form of legitimacy, moral legitimacy (Suchman, 1995; Palazzo & Scherer, 2006), a form that is relatively understudied. Taking this form of legitimacy into account would not only extend Suchman’s legitimacy theory, but this thesis can also provide organisations with a better understanding of the phenomenon and help by offering ways to respond to a spill over effect in practice.

1.3 Research gap

Legitimacy is widely used as a basic concept to explain the effort put into corporate social responsibility by the organisation itself (Grunig, 1979) and it refers, in an organisational context, to the approval of actions by the environment of an organisation (Suchman, 1995). An understanding of legitimacy in the organisational context has mainly been restricted to that of a single organisation; legitimacy, within this limited scope, is the transferral of the legitimate spill over effect to other single business units. In a globalizing society however, scandals of individual businesses increasingly have an effects on the legitimacy of the entire industries – spill overs should no longer just be studied on the level of single companies, but instead on that of whole industries. Legitimate spill over occurs when the legitimacy of a primary subject alters the legitimacy of a ‘cognitively related’ secondary subject without altering the former’s legitimacy (Kostova and Zaheer, 1999). This effect can account

for illegitimate action of the primary subject as well, a phenomenon neglected within legitimacy literature.

Despite the central role of legitimacy within organizational contexts, we know little of the subtle meaning-making processes through which organizational phenomena are legitimated in contemporary society (Vaara, Tienari & Laurila, 2006). This lack of focus on illegitimacy and industry-level spill over demands at least some further research. In order to come to terms with this legitimation process, a detailed analysis of contemporary business news needs to be taken into account, together with a thorough examination of the particular mix of discourses associated with them

(Vaara, Tienari & Laurila, 2006). That is exactly what this thesis sets out to do. This thesis’ research question is therefore fairly straightforward: how do organisations respond to potential industry-level illegitimate spill overs created by scandals?

1.4 Objectives and structure of the paper

This research project has as its objective to deepen our understanding of legitimacy, and, more specifically, our understanding of the legitimation strategies used by organizations when the threat of potential illegitimate spill over in an industry occurs. The project is executed by means of a single case study; it will examine the BMW-Group during and after the diesel gate scandal and investigate their response in this specific context. In doing so, this project tries to provide insights into several processes; for one, it seeks to illuminate responses to potential illegitimate action; secondly, it will indicate the forms of legitimacy and legitimation strategies used to retain the legitimacy of the organisation. In turn, the findings here will lead us to new questions; questions that will be detailed in the discussion. It is an explicit aim of this

thesis to extrapolate propositions from this single case that can potentially point out gaps in current legitimacy theory in need of investigation.

The next chapter provides an over- and review of the relevant literature. This review consists of four sub frames; corporate social responsibility; legitimacy and illegitimacy; spill overs; and lastly a theoretical framework. Following this literature review is a chapters that sets forth the methodology and discusses the research design, sampling, data gathering, data analysis and quality evaluation of this thesis before the discussion and the results are presented. The thesis ends with a conclusion, contribution and suggestions for further research.

2. Literature review

To understand the illegitimate effect in the context of the organization the model of Matten & Moon (2008) is used to illustrate the influence. In this model they refer to a synonym of the organization, the cooperation, to set out the context of the organization. The revelation of the illegitimate action by an organisation other than the organisation investigated creates a change in the organizational field of the latter organisation. The illegitimate shock, as an external action affecting the organisation, causes a change in the organisational field and has its influence on the organisational context. This can lead to a change in the corporate social responsibility practice of the organisation towards a more explicit or implicit redefinition of their motivation as a response to the illegitimate action of the other organisation to retain their own legitimacy.

To understand the response of an organisation to potential industry-level illegitimate spill overs arising from scandals, this thesis starts by analysing three main concepts. A scandal can be seen as a form of corporate social irresponsibility, so as a

first concept, corporate social responsibility and irresponsibility will be discussed. The aim of corporate social responsibility practices is to legitimize the activities of an organisation. Since a scandal is obviously not an attempt to legitimize activities, legitimacy and illegitimacy will subsequently be defined. The overarching effect of the scandal on other industry players is, the spill over effect, is the third concept that merits attention. Having finished a discussion of those three concepts, this review concludes with a theoretical framework.

2.1 Corporate social responsibility

Although Corporate Social Responsibility (CSR) as a concept was developed during the second half of the twentieth century, there still exists some confusion around the definition of CSR (Dahlsrud, 2008). The first modern definition of CSR took shape in the early 1950’s. Back then there was a social responsibility for individual entrepreneurs to “follow those lines of action which are desirable in terms of the objectives and values of our society” (Carroll, 1999). The definition that time was mostly concerned with social responsibility as a concept; corporate was not part of the concept yet, nor was it part of the definition. This focus on individual businessman slowly shifted towards a more corporate view. But still, even now, the definition, though formulated with an individual businessman in mind, applies equally to the corporate setting.

In the 1960’s the development of the definition of CSR started to incorporate the corporate dimension. Davis (1960) started a trend, by redefining CSR in such a way that it is concerned with decisions and actions taken for reasons at least partially beyond the organization's direct economic or technical interest. The development of the definition shows that not just an individual businessman but also an organisation

should do more than just pursuing their economic goals. This redefinition reflects that organisations have some responsibility to the society as a whole. The growing attention for corporate social responsibility resulted in an increase in the amount of scientific papers on this topic (Dahlsrud, 2008). Sadly, as is so often the case, all these papers come with slightly different definitions of, and different standpoints in regard to, corporate social responsibility.

One’s definition of CSR today depends on the different dimensions one includes in the definition (Dahlsrud, 2008). For this thesis the definition that covers a wide scale of dimensions and which is widely used, seems most appropriate because CSR as a concept is not investigated for this thesis but solely applied. Dahlsrud (2008) shows that the definition of the Commission of the European communities, 2001, is most frequently used and this definition is therefore taken up in this thesis. CSR, the definition runs, is a concept whereby companies integrate social and environmental concerns in their business operations and in their interaction with their stakeholders on a voluntary basis (Dahlsrud, 2008).

The more voluntaristic CSR dimensions in recent years caused a development of CSR towards what is called corporate social performance (CSP), an inclusive and global concept that embraces corporate social responsibility, responsiveness, and the entire spectrum of socially beneficial activities of businesses (Carroll, 1999). Where Milton Friedman (1970) argues that the only responsibility of a firm is to increase its profits while the field of corporate social responsibility developed, in CSP the economic aspect is just one of the many responsibilities an organisation has. Different stakeholders of organisations desire different objectives and values, and therefore corporate social responsibility is a multi-dimensional concept.

The definition of CSR helps us to understand social responsible behaviour and suggests that corporate social irresponsible behaviour is simply the opposite of CSR.

Irresponsibility, as distinct from responsibility, is, in the CSR literature, wrongly implied to be the opposite of responsibility; rather its opposite is the failure for organisations to act responsibly (Lange & Washburn, 2012). When organisations do not act responsibly by violating the expectations of the observer, the pressure on organizations to change their socially irresponsible behaviour rises (Coombs, 1998). The charges of corporate irresponsibility are legitimate charges when behaviour goes against standards commonly accepted by a society; socially irresponsible behaviour is irresponsible to the extent observers perceive it as such (Coombs, 1998; Lange & Washburn, 2012). A line, therefore, cannot clearly be drawn between responsible and irresponsible behaviour, as one does, for instance, when taking a legal perspective; any distinction is rather subject to the perception of the organization's environment, and hence to some extent subjective.

The demand for corporate social responsibility can differ for each industry or for each country an organisation operates in. That can explain the confusion in the literature about corporate social responsibility. Dahlsrud (2008) analysed 37 different definitions of corporate social responsibility and concluded that the existing confusion is not so much about the definition but more about the construction of corporate social responsibility within its context. The context dimensions of corporate social responsibility include environmental, social, economic, stakeholder and voluntariness (Dahlsrud, 2008). Standpoints on these dimensions towards CSR can differ as contexts vary.

The different approaches and standings towards corporate social responsibility do exist and differ among organisations. Two processes of CSR, implicit and explicit

corporate social responsibility, are two general approaches towards CSR practice (Matten & Moon, 2008). The implicit corporate social responsibility consists of values, norms, and rules and relate to the more required and obligated issues of the organisation. Explicit corporate social responsibilities are voluntary programs and strategies, which enables the corporation to address issues as being part of the corporation. So there is a difference in the process of CSR practices undertaken by organizations.

These implicit and explicit CSR processes show some similarities with the stakeholder and voluntariness dimension of Dahlsrud (2008), where the stakeholder dimension is more obligated CSR practice than the voluntariness dimension. The distinction between levels of CSR practices is even further defined into four categories of CSR; economic, legal, ethical and philanthropic (Carroll & Shabana, 2010). These categories address the motivation of organisations and can be useful to identify specific kinds of benefits that flow back to companies, as well as society, in their fulfilment of CSR. Economic and legal can be aligned with the implicit and stakeholder dimension whereas the philanthropic and ethical level can be more voluntariness and explicit.

CSR in the context of environmental, social, economic, stakeholder and voluntariness dimensions is a concept whereby companies integrate social and environmental concerns in their business operations and in their interaction with their stakeholders on a voluntary basis (Dahlsrud, 2008). Carroll & Shabana (2010), show that CSR practice can be better understood by defining the four layers of CSR; economic, legal, ethical and philanthropic, which align with the implicit and explicit process of corporate social responsibility of Matten & Moon (2008). What corporate social responsibility reflects for a corporation is the “social imperatives and the social

consequences of business success” (Matten & Moon, 2008). The slight adjustment of CSR towards corporate social performance suggest organisation can differ in their “social imperatives and the social consequences of business success”. The environment of the organisation determines CSR and corporate social performance. You can refer to legitimacy as a basic concept to actually measure the corporate social performance of the organisation and understand the effect of CSR effort in the organisation’s environment (Grunig, 1979). The aim of CSR practices is to legitimize the activities performed by the organisation to increase the organisation’s acceptance by society (Palazzo & Richter, 2005).

2.2 Legitimacy and illegitimacy

In the institutional theory, the link of legitimacy with CSR can be made as the concept of legitimacy suggest that CSR practice can be used to narrow the legitimacy 'gap' between how the organization wishes to be perceived and how it actually is (Campbell, 2000). In the literature a set of definitions has been developed that all try to define the concept of legitimacy (Parsons & Jones, 1960). Parsons & Jones (1960) describe legitimacy as “the appraisal of action in terms of shared or common values in the context of the involvement of the action in the social system”. Haack, Pfarrer & Scherer (2014) state “legitimacy refers to the degree to which beholders perceive an organization as being congruent to social norms and standards”. A more overarching definition of legitimacy is that “Legitimacy is a generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions.” (Suchman, 1995). In the organisational context, legitimacy can be understood as the approval of actions by the environment of an organisation.

The approval of actions represents the reaction of observers to the organization as they see it, which means that legitimacy is possessed objectively, yet can be created subjectively (Suchman, 1995). In that sense, to establish legitimacy and understand how legitimacy is created can be very complex. Legitimacy can be better understood by making a distinction between pragmatic, moral and cognitive legitimacy. Suchman (1995) describes pragmatic legitimacy as a form of exchange legitimacy where an organisation gains approval for policies that meet up with the expectation of that policy to the observer. Moral legitimacy is concerned with ethics and reflects the question of ‘Is it the right thing to do?’ (Thomas and Lamm, 2012). At last, cognitive legitimacy is about the approval of inevitable action of the organization.

Being clearly defined, legitimacy helps us to better understand the concept and improves our understanding of illegitimacy, which is the complete opposite legitimacy. Countering Parsons & Jones (1960), illegitimacy is the negative appraisal of action in terms of shared or common values in the context of the involvement of the action in the social system. These are actions that are not desirable proper or appropriate within some socially constructed system of norms, values, beliefs, and definitions. An organization that is accused of illegitimate action does not have approval of action by the environment. They can be accused of pragmatic, moral or cognitive illegitimacy by whoever observes the organisation.

There are two theoretical fields that have a different approach towards legitimacy, the strategic view and the institutional view. From the institutional perspective, the assumption is that organisations are active in a homogeneous context and therefore have limited influence on creating legitimacy (Haack et al. 2014). So legitimacy is out of the organisation’s control and the organisation can create

legitimacy only by aligning with the homogeneous context. In contrast with the institutional assumption of homogeneous context, the strategic view sees legitimacy as a source of advantage that can be shaped by the firm (Suchman, 1995). The fact that the organisation’s context is homogeneous does not necessarily mean that it cannot be a source of advantage as well. Dacin, Oliver, & Roy (2007) acknowledge both the strategic and institutional functions of legitimacy, but take legitimacy as a broader concept. That broader concept pertains the extent to which a firm's structures and activities appear to conform to social norms, values, and expectations of the firm's economic and social environment. This comes back to the CSR practices occupied by the organisation.

The controversial beginning of CSR has changed to a now general acceptance of CSR as a legitimate business function, which is outlined along with a discussion of the role of corporate social performance (McWilliams, 2000). The traditional concepts of CSR are mainly based on pragmatic and cognitive legitimacy (Scherer & Palazzo, 2011). CSR, in a national context, is building on the assumption that organisations preserve their legitimacy by following the national rules of the game (Scherer & Palazzo, 2011). While in a country organisations could build upon a clear-cut division of labour and could profit of a stable legal and moral context for their operations, the process of globalization casts doubt upon the efficiency and legitimacy of these established roles and responsibilities (Scherer & Palazzo, 2011). The changing condition of globalization pressures these forms of legitimacy (Palazzo & Scherer, 2006). Because of the pressure on pragmatic and cognitive legitimacy by the globalization processes, organisations are often required to establish the third form of legitimacy, moral legitimacy (Suchman, 1995).

The link between legitimacy theory and CSR is that CSR practice can be used to narrow the legitimacy 'gap' between how the organization wishes to be perceived and how it actually is (Campbell, 2000). “Legitimacy is a generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions” and can be better understood by making a distinction between pragmatic, moral and cognitive legitimacy (Suchman, 1995). Dacin, Oliver, & Roy (2007) acknowledge both the strategic and institutional functions of legitimacy but take legitimacy as a broader concept and it is this broader concept that will be used for this thesis. The traditional concepts of CSR are mainly based on pragmatic and cognitive legitimacy but, as outlined above, globalization has made moral legitimacy imperative (Scherer & Palazzo, 2011).

2.3 Spillover effect

Central for this thesis is the potential industry-level illegitimate spill overs. The spill over effect needs to be defined specifically for the context of this thesis. So far we discussed corporate social responsibility, irresponsibility and legitimacy, illegitimacy as separated concepts and aimed to gain better understanding of these concepts. This thesis, however, is focussed on the short-term response of an organisation to a

potential industry-level illegitimate spill overs arising from scandals. Spill over refers to the extent to which a message influences beliefs related to attributes that are not contained in the message (Ahluwalia, Unnava & Burnkrant, 2001). Put this in context of this thesis we define the spill over effect as the extent to which illegitimacy influences the beliefs related to attributes that are not contained in the message.

Legitimate spill over occurs when the legitimacy of a primary subject alters the legitimacy of a ‘cognitively related’ secondary subject without altering the former’s legitimacy (Kostova and Zaheer, 1999). This related subject could be either a subsidiary that operates at the same organisational level or a subsidiary down the hierarchy under the umbrella of the parent company. These two situations are called horizontal and vertical legitimate spill over (Haack et al., 2014). In the organisational field, organisations are operating at the same level and therefore horizontal spill over might occur.

2.4 Theoretical framework

2.4.1 Forms of legitimacy

Corporate social responsibility efforts can be understood as an attempt to create legitimacy for an organisation. The environmental, social, economic, stakeholder and voluntariness, dimensions of corporate social responsibility, and the socially constructed system of norms, values, beliefs, and definitions for legitimacy need to be aligned. This alignment is what corporate social responsibility reflects, “social imperatives and the social consequences of business success” (Matten & Moon, 2008). The socially constructed system of norms, values, beliefs, and definitions for legitimacy is assumed to be stable and homogeneous (Haack et al., 2014). We acknowledge that this might not always be the case. However, because this thesis deals with the organisation’s response on the shock in the organisational field and not on the socially constructed system itself, it is necessary to make this assumption.

According to Suchman (1995) there are three forms of legitimacy that can be obtained; pragmatic, moral and cognitive. Suchman (1995) describes pragmatic legitimacy as a form of exchange legitimacy where an organisation gains approval for

policies that meet up with the expectation of that policy to the observer. This form can be most expected when the responding organisation is trying to prove that they did not commit the wrongs the other organisation is accused of. Moral legitimacy is concerned with ethics and reflects the question of ‘Is it the right thing to do?’ (Thomas and Lamm, 2012). This form of legitimacy might not directly relate to the illegitimate action and is expected to be more about the approval of business as a whole. At last, cognitive legitimacy is about the approval of inevitable action of the organization (Suchman, 1995). Basically, this means the confession of illegitimate action on the grounds that it was inevitable for survival of the organisation. This form of legitimacy will be least expected, since the companies (within our situation) are not directly accused of illegitimate action, so no confessions can be made.

2.4.2 Legitimation strategies

When looking at the response on potential illegitimate spill over in the industry, the actual aim is to understand the organisations legitimation strategy. Despite the central role of legitimacy, we know little of the subtle meaning-making processes through which organizational phenomena, such as industrial restructuring, are legitimated in contemporary society (Vaara, Tienari & Laurila, 2006). For the legitimation of organisational phenomena five legitimation strategies are identified (1) normalization is to render something legitimate by exemplarity, (2) authorization is legitimation by reference to authority, (3) rationalization is legitimation by reference to the utility or function of specific actions or practices, (4) moralization is legitimation that refers to specific values, and (5) narrativization is legitimation by portraying others as winners, losers, hero's, adversaries, or culprits (Vaara, Tienari & Laurila, 2006). In this thesis these strategies are used to understand the response of an organisation in the short term in the context of potential industry-level illegitimate spill overs from scandals to

retain pragmatic, moral or cognitive legitimacy. This enriches the understanding of how forms of legitimacy are retained with the threat of potential illegitimate spill over and which strategies fit best with each form of legitimacy.

3. Methodology

The methodology is presented in in five separate parts. First, the design of this research project will be discussed. Second, the sampling procedure for this thesis is described. Third, the gathering of data is explained. Fourth, the approach of analysing the found data is described. Fifth and final the evaluation of quality is explicated.

3.1 Research design

This thesis has a qualitative approach, which enables a naturalistic insight to the subject matter as little is known about the phenomenon (Saunders & Lewis, 2012). A case study is used to accomplish the partly explorative nature of this thesis and extend theory on legitimation strategies to potential illegitimate spill over effects (Eisenhardt, 1989). Precisely because of the partly explorative nature of this study and the fact that still little is known about this phenomenon a single case study seems appropriate (Saunders & Lewis, 2012). The aim of this study is not to generalize the findings but more to explore the proposed phenomenon of legitimation strategies to potential illegitimate spill overs.

3.2 Sampling

The case of emission fraud in the automotive industry has the threat of potential illegitimate spill over and therefore it can be used to investigate the central question of how organisations respond to potential industry-level illegitimacy spill overs from

scandals. Lowering the emissions of cars is demanded by government and society to protect the environment. Volkswagen Group could not meet the emission criteria, which resulted in emission fraud called ‘diesel gate’ in September 2015. The Volkswagen Group created a shock by the illegitimate action to fraud with emissions in the organisational field of the automotive industry, especially in Germany. Of the traditional German ‘three’, Audi, Mercedes and BMW, BMW is the only company that has not been named for committing fraud, a fact that makes BMW unique and interesting for this thesis because it provides us with the opportunity to see how BMW responded to the illegitimate action of Volkswagen. This non-probability sampling can be justified because this single case study merits an in-depth investigation, give the explorative nature of the thesis (Saunders & Lewis, 2012).

3.3 Data gathering

This thesis analysis the short-term response on the potential illegitimate spill overs of the ‘diesel gate’ scandal in the automotive industry. Although the scandal is still widely discussed it is not a hot topic in the industry anymore. To define short-term for this thesis, a trend analysis on google has been done to determine when diesel gate and emission, within the automotive industry, were relatively ‘hot’ topics. ‘Hot’ is defined by looking at keywords ‘emission’ and ‘diesel gate’ in the automotive industry and estimating when emission and diesel gate received relatively more attention than they do in the normal situation. That time frame of being a relatively ‘hot topic’ defines the time frame for short-term responses to the scandal. As it is, diesel gate and emission have been hot for two and a half months between the 13th of September till the 2nd of December in the industry.

For this set, timeframe data were gathered from two sources, BMW-Group and the LexisNexis database. For the Group all press releases of the

BMW-Group have been included in this research whether they were related directly to the topic or not. The reason to include all releases was to be able to investigate whether some form of legitimation was done by indirect communication that could be related to the scandal. Excluding communication of BMW-Group would have made this impossible so therefore no exception is made. The list of releases by the BMW-Group can be found in table 1.

Table 1

Articles BMW-Group

Source Title

BMW Group Corporate Communication

BMW Group statement concerning the current discussion of diesel engines.

BMW i Corporate Communication

Emission-free cash transport with the BMW i3. BMW Group Corporate

Communication

Statement Harald Krüger Chairman of the Board of Management of BMW AG Conference Call Interim Report to 30 September 2015 3 November 2015, 10:00 a.m.

BMW i Corporate Communication

The BMW i3 turns two. Time for an interim review. BMW Corporate

Communication

“sport auto Award 2015”: Four BMW models took first place in their category in the magazine’s readers’ choice awards – the BMW M135i, BMW M235i, BMW 335i and BMW M3.

BMW Plant Hams Hall Corporate

Communications

BMW UK engine plant ramps up production of next generation engines.

BMW Media Information

BMW at the Tokyo Motor Show 2015. BMW Group Corporate

Communication

Press conference business domain strategy BMW Motorrad. MINI Press Folder THE NEW MINI CLUBMAN.

MINI Press Folder THE NEW MINI CONVERTIBLE. BMW Group Corporate

communication

BMW Germany unveils world’s largest giant poster for IAA 2015.

BMW Group Corporate communication

Peter Schwarzenbauer Member of the Board of Management of BMW AG for MINI, BMW Motorrad, Rolls-Royce, Aftersales BMW Group Stephan Schaller President BMW Motorrad.

In the database of LexisNexis all English written news was collected for the set timeframe that discussed the terms “emission” in their text and were directly related to the BMW-Group. As the BMW-Group is not directly involved in the scandal, terms like “diesel gate” or “diesel gate” were not used in gathering data because this would not supply data related to the BMW-group. Central is the scandal, which can be related to BMW-Group and is therefore appropriate to use. When we searched for emission more than 3000 related articles came forward. Than we narrowed the results by looking for company related articles which brought us to a total of 21 articles for this thesis. The list of articles out of the LexisNexis database is presented in table 2.

Table 2

LexisNexis articles

Source Title

Business Spectator BMW shares fall on emissions concerns.

Indepedent.co.uk Volkswagen emissions scandal: BMW forced to deny it falsified emissions tests; The car company was responding to the claims in German trade magazine Auto Bild.

The Gold Coast Bulletin BMW disputes emissions cheating claim.

ANSA English Media Service Auto Bild denies accusing BMW of emissions manipulation; Magazine says test values aren't proof of falsification.

just-auto global news GERMANY: BMW denies it has manipulated emission tests.

London Evening Standard BMW denies emissions test manipulation after shares sink 8%.

Express Online Other carmakers to be investigated for emissions cheating.

Press Association Mediapoint BMW denies emission-rigging reports as share price plummets.

Channel NewsAsia US regulator extends diesel emissions probe to other car models: FT.

Channel NewsAsia BMW denies report about emissions manipulations.

Daily Mail BMW dragged into diesel storm.

Xinhua General News Service BMW denies manipulating or rigging emission tests.

India Automobile News BMW says none of its vehicles use defeat devices. Financial Wire EU pollutant emission testing.

Scottish Daily Mail BMW dragged into diesel storm; Luxury car giant faces claims that it was also fiddling green tests. Agence France Presse Daimler denies any rigging of emission tests.

Daily Mail VW scandal hits ami-firm hard.

ANSA English Media Service BMW also rigged emissions tests – Bild.

MailOnline MARKET REPORT: VW scandal hits Ubisense

hard as shares skid 27% on a shock profits warning.

3.4 Data analysis

The content analysis of qualitative data is a subjective interpretation of the content of qualitative data by the systematic process of coding and identifying themes and patterns (Hsieh & Shannon, 2005). The first step in the analysis is to split the data in a direct and indirect link with the diesel gate scandal. That means there will be two groups for each sources which totals four groups. Although the direct related data to the scandal are expected to give more insight in responses of the organisation, there can be found organisational legitimation strategy concerned with the diesel gate in indirect data sources as well. These groups will be used to see whether different strategies of legitimation and forms of legitimacy come forward in different groups, which enriches the understanding of organisational response to potential industry-level illegitimate spill overs from scandals.

As this thesis has a deductive approach, codes and categories are derived from the theory discussed before the starting to analyse the data (Bourque, 2004). With coding, the aim is to assigns a summative or essence-capturing word to specific parts of the data that has been collected (Hsieh & Shannon, 2005). Legitimacy is understood by making a distinction between pragmatic, moral and cognitive legitimacy Suchman (1995). The responses to potential industry-level illegitimate

spill overs from scandals are expected to gain, retain or repair the legitimacy of the organisation. Legitimacy is therefore the central theme in the analysis of the data. So the first more descriptive coding of data will be focused on the forms of legitimacy the organisation is trying to gain, retain or repair. The first three descriptive codes that will be used are “moral legitimacy”, “pragmatic legitimacy” and “cognitive legitimacy”. In the legitimacy literature organisational legitimacy is mostly obtained by moral legitimacy (Suchman, 1995). This descriptive coding can already give some interesting insights in which form of legitimacy topotential industry-level illegitimate spill overs from scandals are mostly aimed for.

The next step, after analysing data for forms of legitimacy, in analysing the data will be more analytical and focus on the different legitimation strategies that were used to obtain each form of legitimacy. Vaara, Tienari & Laurila, (2006) introduced five legitimation strategies for organisational phenomena which are (1) normalization: render something legitimate by exemplarity, (2) authorization is legitimation by reference to authority, (3) rationalization is legitimation by reference to the utility or function of specific actions or practices, (4) moralization is legitimation that refers to specific values, and (5) narrativization is legitimation by portraying others as winners, losers, hero's, adversaries, or culprits. For coding these strategies, the codes normalization, authorization, rationalization, moralization and narrativization were used. The aim is for this thesis is a detailed analysis of organisational response to potential industry-level illegitimate spill overs to examine the particular mix legitimation strategies, which allows to deductive approach for this thesis (Saunders and Lewis, 2012).

3.5 Evaluation of quality

The use of multiple sources in this thesis creates some control over the data that will be analysed and will form triangulation to ensure that the data is telling what it should be telling (Saunders & Lewis, 2012). The danger of using of secondary data in this case study is that the data can be biased and therefore is low in reliability (Saunders and Lewis, 2012). Data developed and handed out by BMW might not objectively describe the actual situation but reflects their intended message. However getting access to corporate data and actually collecting them is not feasible for this study. The use of LexisNexis can be a source of control for the data obtained at the BMW-Group to ensure its reliability.

The generalizability of this research is a particular worry because this research is performing a case study which known for its lack of generalizability, especially a single case study (Saunders and Lewis, 2012). However a case study seems to be the most logical strategy to answer the question in this research. Findings in this specific study might not account directly to other industries. This case study will explain what is going on in a specific setting and can help to build some settings for further research. Insights of this study can be used to study other organisational fields and might lead to conceptualization of findings that can be accounted more generally.

4. Results

In this chapter the results of the analysis are presented. The findings in the data of BMW-Group and the findings in the LexisNexis database will be discussed separately. First to be presented is the direct and indirect related articles to the central topic of illegitimate action, which is based on the title of articles. Second, the results of legitimacy forms is presented whether the focus is on any of these three forms;

pragmatic, moral or cognitive legitimacy. Third, the relation between each form of legitimacy and legitimation strategies is shown to get an overview of how the organisation in this specific case is responding to potential illegitimate spill over. In the appendices A and B a full overview of the findings can be found, the text will present numbers and examples to clarify the findings. As this is only the presentation of results, these findings will be discussed in the discussion chapter of the thesis.

4.1 BMW-Group press releases

4.1.1 Direct and indirect related

For the BMW-Group all press releases for the given time-frame of two and a half months between the 13th of September till the 2nd of December were taken into the data set and analysed. In this period a total of 12 articles press articles were released. These releases differed from statements of the board, to presenting new models or results in the auto motorsport. Not all directly related to the topic so therefore a first analysis was done to see which articles were and were not directly related to the topic. There was one article that was directly responding to the recent scandal of Volkswagen diesel gate.

Example: BMW Group statement concerning the current discussion of diesel engines All other 11 releases were not directly related to the central topic based on the title of the article but might however be a response to the scandal indirectly.

Example: Statement Harald Krüger Chairman of the Board of Management of BMW AG Conference Call Interim Report to 30 September 2015 3 November 2015, 10:00

a.m.

Releases that were not expected to give any response on the current diesel engine scandal were articles that were not related to the core brand of the BMW-group and not threatened by the illegitimate spill over of the scandal.

Example: MINI Press Folder

For the twelve articles one is a direct response to the scandal, eleven were indirectly related and of those eleven articles two were noticed to have no response to the scandal, as they were both releases of MINI.

4.1.2 Forms of legitimacy

For the twelve articles the analyses of different forms of retaining legitimacy has been done. The three forms of legitimacy, pragmatic, moral and cognitive of Suchman (1995), were used to define the retaining legitimacy efforts, which can be related to the central illegitimate action of the diesel gate scandal by Volkswagen. Presenting these results the distinction between direct and indirect related articled need to be kept in mind. The direct related press release of the BMW-Group shows two forms of legitimacy. The release has four pragmatic attempts of retaining legitimacy.

Example: The BMW Group does not manipulate or rig any emissions tests. We observe the legal requirements in each country and fulfill all local testing

requirements.

Example: We support the swift introduction of the new regulations to create clarity for consumers and the industry as quickly as possible.

No form of cognitive attempt to retain legitimacy in the direct related article to the scandal was found. For the indirect related releases findings were similar compared to the direct related release. Seven pragmatic legitimacy efforts, eight moral and one cognitive attempt to retain legitimacy. Cognitive legitimacy efforts did not come forward in the direct related article.

Example: Diesel is one of the cleanest and most efficient drive trains in the industry. Without diesel powertrains, we would not have been able to meet the strict CO2 limits

in Europe.

For all these legitimacy retaining efforts different strategies can be identified within the form of legitimacy that has been tried to retain.

4.1.3 Legitimation strategies

For the different forms of legitimacy different strategies can be used. This last form of analysis present how these three forms of legitimacy were obtained. In the table all frequencies of different strategies for different forms of legitimacy are presented. For the pragmatic legitimacy the direct related article mostly refers the authority so it uses authorization to retain pragmatic legitimacy.

Example: We observe the legal requirements in each country and fulfill all local testing requirements.

Also narrativization and normalization were used as is shown in the table. The indirect related articles show more normalization as core strategy to retain the legitimacy of the BMW-Group.

Example: Customers can choose from 86 models within the BMW Group portfolio that emit less than 120g of CO2 per kilometer in the current EU test cycle. As in the direct relate article, the indirect related articles use narrativization and authorization as other strategies to retain pragmatic legitimacy for the organisation. Rationalization and moralization were not used as strategy to retain pragmatic legitimacy for the organisation.

The retainment of moral legitimacy for the organisation in the direct related article does not show a preference for one legitimation strategy. In the article three strategies are used, moralization, rationalization and normalization. The indirect related articles show a slight preference for rationalization as strategic legitimation choice to retain their moral legitimacy.

Example: Over the past years, we have cut the CO2 emissions of our new car fleet in Europe by more than one third.

Other strategies used for moral legitimacy were narrativization, normalization, moralization and authorization, which means all strategies, were applied to retain moral legitimacy.

Last, the detainment of cognitive legitimacy was only been attempted ones in this analyses. The direct relate article did not show attempts to retain cognitive legitimacy so therefore no strategies were found in this specific case. However, the

indirect articles showed one attempt to retain cognitive legitimacy and used authorization as strategy to do so.

Example: Without diesel powertrains, we would not have been able to meet the strict CO2 limits in Europe.

Other strategies were logically not applied, as cognitive legitimacy has not been a central form of legitimacy to retain. All frequencies of finding in the BMW-Group press releases are found in table 3.

Table 3

BMW frequency table

Legitimacy (Suchman, 1995)

Legitimation strategy (Vaara et al., 2006) Direct Indirect Pragmatic Narrativization Normalization Authorization Rationalization Moralization Denial 1 1 2 0 0 0 1 4 2 0 0 0 Moral Narrativization Normalization Authorization Rationalization Moralization Denial 0 1 0 1 1 0 1 1 1 3 2 0 Cognitive Narrativization Normalization Authorization Rationalization Moralization Denial 0 0 0 0 0 0 0 0 1 0 0 0

4.2. LexisNexis database

4.2.1 Direct and indirect related

For the LexisNexis database all English press releases concerned with “emission” and direct related to the BMW-Group for the given time-frame of two and a half months between the 13th of September till the 2nd of December were taken into the data set and analysed. In this period a total of 21 articles were published. One article was released twice so that meant the analysis issued 20 articles in total. Just like the BMW-Group press releases, the first analyses was done to see which articles were and were not directly related to the topic based on the article titles. Most of the articles were directly related to BMW-Group and the emission scandal.

Example: BMW dragged into diesel storm.

The articles that were not directly related were articles focussed on the scandal but not so much on the BMW-Group and therefore been appointed as indirect related articles for this thesis.

Example: EU pollutant emission testing.

In the total data set of twenty articles, thirteen articles were directly related to the BMW-group and the emission scandal, seven of them were indirectly related.

4.2.2 Forms of legitimacy

For the twenty articles the same analyses, of different forms of retaining legitimacy, as for the BMW-Group press releases has been done. The three forms of legitimacy, pragmatic, moral and cognitive of Suchman (1995), were used to define the retaining legitimacy efforts, which can be related to the central illegitimate action of the diesel

gate scandal by Volkswagen. Presenting these results the distinction between direct and indirect related articles need to be kept in mind. The direct related articles showed a total of thirty-seven legitimacy efforts to retain pragmatic legitimacy.

Example: The company's exhaust treatment systems are active whether on the test bench or on the road, he added.

Another six attempts of retaining legitimacy are found to be moral legitimacy efforts.

Example: We are willing to discuss our testing procedures with the relevant authorities and to make our vehicles available for testing at any time.

For cognitive legitimacy attempts another four were found. Which is interesting as none were found in the direct related article of the BMW-Group press releases.

Example: not familiar with the test mentioned by Auto Bild magazine.

For the indirect related releases findings were different compared to the direct related release. In the indirect related articles only one moral form of legitimacy has been observed. The indirect related articles do not say anything further that can be related to the BMW-Group. For all these legitimacy retaining efforts in the direct related articles different strategies can be identified within the form of legitimacy that has been tried to retain.

4.2.3 Legitimation strategies

For the different forms of legitimacy different strategies can be used. This last form of analysis presents how these three forms of legitimacy were obtained. In the table all

frequencies of different strategies for different forms of legitimacy are presented. For the pragmatic legitimacy the direct related articles mostly refer to examples so it uses normalization to retain pragmatic legitimacy.

Example: vehicle's exhaust systems make no difference between the testing and road environment.

Also narrativization and authorization were used as is shown in the table. As none of the indirect related articles retained pragmatic legitimacy no strategies for indirect articles were found. Rationalization and moralization were not used as strategy to retain pragmatic legitimacy for the organisation, which is similar to the findings in the press releases of the BMW-Group.

The retainment of moral legitimacy for the organisation in the direct related articles does show a preference for one legitimation strategy, moralization. In the articles no other strategy than moralization was applied to retain moral legitimacy. The indirect related articles showed one attempt to retain moral legitimacy and also used moralization as strategy to accomplish that.

Example: We are willing to discuss our testing procedures with the relevant authorities and to make our vehicles available for testing at any time.

There were no other strategies used for moral legitimacy, which means that narrativization, normalization, rationalisation and authorization were not applied to retain moral legitimacy.

Last, the retainment of cognitive legitimacy has been attempted four times in this analyses. The direct related articles showed preference for moralization as strategy to obtain cognitive legitimacy.

Example: was not familiar with the test mentioned by the magazine.

Other strategies were not applied as cognitive legitimacy has not been a central form of legitimacy to retain and only been applied four times. The indirect articles showed no attempt to retain cognitive legitimacy so no strategies were there to be found. An overview of all frequencies of the findings in the articles of LexisNexis is listed in table 4.

Table 4

LexisNexis frequency table

Legitimacy (Suchman, 1995)

Legitimation strategy (Vaara et al., 2006) Direct Indirect Pragmatic Narrativization Normalization Authorization Rationalization Moralization Denial 4 12 6 0 0 15 0 0 0 0 0 0 Moral Narrativization Normalization Authorization Rationalization Moralization Denial 0 0 0 0 6 0 0 0 0 0 1 0 Cognitive Narrativization Normalization Authorization Rationalization Moralization Denial 0 0 0 0 3 1 0 0 0 0 0 0

4.3 Additional unexpected findings

There are two additional notes that need to be made as part of the results founded in the data. The first note which might already been noticed is that there is a lot of overlap between the articles found in the LexisNexis database. That overlap can be explained by the press release of the BMW-Group. The statement that was made by the BMW-Group concerned with the diesel gate scandal is often referred to in the articles found in the LexisNexis database. Second, we need to acknowledge that especially in the LexisNexis data there seems to be a sixth strategy to retain legitimacy that has not been part of the drafted five legitimation strategies introduced in the theory of this thesis. In the analysis it has been coded as “denial” and mostly found in retaining pragmatic legitimacy, which will be discussed in more depth in the discussion chapter of this thesis.

5. Discussion

The objective of this research is to extend the understanding of legitimacy and more specifically the legitimation strategies used by an organization when the threat of potential illegitimate spill over in an industry occurs. Legitimate spill over occurs when the legitimacy of a primary subject alters the legitimacy of a ‘cognitively related’ secondary subject without altering the former’s legitimacy (Kostova and Zaheer, 1999). With the qualitative single case research approach on the BMW-Group and the diesel gate we tried to identify the response to scandals. For the legitimation of organisational phenomena five legitimation strategies were identified and implicated to the case (Vaara, Tienari & Laurila, 2006). In the organisational context,

legitimacy can be understood as the approval of actions by the environment of an organisation (Suchman, 1995). This discussion will first discuss how the response of

an effected organisation should be interpreted. Than we will come forward with the forms of legitimacy and legitimation strategies used to retain legitimacy of the organisation. We aim to make propositions out of this case that will help to point out new gaps in theory that need to be investigated.

5.1 Responses

In the figure of Matten & Moon (2008) we showed the institutional context of the organisation. The scandal created a shock in the organisational field and affects the organisation, which would force the organisation to rethink their approach towards CSR. In this thesis we referred to institutional theory and legitimacy as a basic concept to explain the effort put into CSR (Grunig, 1979). So, for a change in the organisational field, and in this case a negative change, legitimation efforts were expected as a response towards the diesel gate scandal. From the institutional perspective, the assumption is that organisations are active in a homogeneous context and therefore have limited influence on creating legitimacy (Haack et al. 2014). If we would take that perspective the scandal would have influenced the organisational field of the organisation but no response or change in CSR practice would be made because the organisation would have no influence on retaining their legitimacy. However, when taking a look at the results in this specific case the response of the affected organisation is obvious. The statement of BMW-Group is a direct response to the diesel gate scandal and is understood as an attempt to retain legitimacy of the organisation.

In contrast with the institutional assumption of homogeneous context, the strategic view sees legitimacy as a source of advantage that can be shaped by the firm (Suchman, 1995). Although the results of this thesis cannot clearly state that BMW-Group was able to retain their legitimacy while the threat of potential industry-level

illegitimate spill over occurred, the results show that BMW-Group is putting in effort to retain their legitimacy. Based on the insight of these results we would like to point out that an organisation will respond to the illegitimate action of another organisation when the when the threat of potential spill over is there.

The distinction made in the analysis between direct and indirect related articles was to see weather legitimacy on emissions was retained without referring directly to the scandal. In the BMW-Group press releases emission and performance on emission was mentioned several times. Specifically the releases of the BMW i Department referred to that context. For the sake of this thesis all message were included and defined as attempts to retain legitimacy. However, it is hard to say weather the indirect related content was actually an attempt to retain legitimacy as response towards the industry scandal. Although in this thesis we cannot directly draw-up conclusions that legitimacy is also retained indirectly, it might be interesting to do further research and see if scandals change the communication on the related subject. Another note we would like to make is that although in the analysis showed indirect legitimation, the message did not came across in the news articles of the LexisNexis database.

5.2 Forms of legitimacy

With the responses of the BMW-Group, we state that organisations are taking a strategic perspective towards the concept of legitimacy and the potential threath of illegitimate spillovers. In that strategic perspective there are three main forms of legitimacy to retain; pragmatic, moral, and cognitive legitimacy (Suchman, 1995). The changing conditions of globalization have put the two traditional modes of legitimacy, pragmatic and cognitive legitimacy, under pressure. Consequently,

organisations are often required to establish a third form of legitimacy, moral legitimacy (Suchman, 1995; Palazzo & Scherer, 2006).

First we look at the two traditional forms and the definition of spill over in the context of this thesis. The scandal of an industry player is influencing the beliefs related to organisations that are not contained in the message (Ahluwalia, Unnava & Burnkrant, 2001). The two traditional forms of legitimacy can either be a form of exchange legitimacy where an organisation gains approval for policies that meet up with the expectation of that policy to the observer (pragmatic), or get the approval of inevitable action of the organization (cognitive legitimacy) (Suchman, 1995). In the results of the analysis we hardly found any effort to retain cognitive legitimacy. With the definition of cognitive legitimacy, approval of inevitable action of the organization, it seems to make sense that in the case of potential spill overs this will not be retained. Approval of action is more expected when the organisation has confessions to make. Keeping in mind the spill over effect the organisation is not contained in the message of the industry scandal, cognitive legitimacy does not need to be retained.

Pragmatic legitimacy is tried to retain as a response to illegitimate action of other industry players. In the results pragmatic legitimacy was the form of legitimacy that received most attention. This is coherent with the limited findings of cognitive legitimacy. Retaining the approval of policies that meet up with the expectation of that policy to the observer does make sense when kept in mind the position of the effected organisation. The pragmatic legitimacy can be drawn back to the basic concepts of corporate social responsibility. Carroll & Shabana (2010), show that CSR practice can be better understood by defining the four layers of CSR; economic, legal, ethical and philanthropic, which align with the implicit and explicit process of

corporate social responsibility of Matten & Moon (2008). Pragmatic legitimacy focuses on the implicit, or economic and legal, aspects of corporate social responsibility to prove that policies of the organisation meet up with the expectation the observer.

Referring to the basic concepts of corporate social responsibility and the changing conditions of globalization which put the two traditional modes of legitimacy, pragmatic and cognitive legitimacy, under pressure often requires organisations to establish a third form of legitimacy, moral legitimacy (Suchman, 1995; Palazzo & Scherer, 2006). Moral legitimacy received quite some attention in the analysis. Moral legitimacy is concerned with ethics and reflects the question of ‘Is it the right thing to do?’ (Thomas and Lamm, 2012). In terms of corporate social responsibility moral legitimacy is more related to explicit or ethical and philanthropic part of corporate social responsibility. We found that moral legitimacy is an extension of retaining legitimacy as such. This fits the idea of the pyramid of corporate social responsibility of Carroll & Shabana (2010), where economic and legal need to be fulfilled in order to fulfil ethical and philanthropic needs.

5.3 Legitimation strategies

To retain the different forms of legitimacy different legitimation strategies were used. Legitimation strategies for cognitive legitimacy will not be discussed because we found cognitive legitimacy did not receive attention when the threat of illegitimate spillover occurred. The attempt to retain pragmatic legitimacy showed normalization, authorization and narrativization as legitimation strategies already described by

Vaara, Tienari & Laurila (2006). We would like to bring up a fourth strategy that did not come forward in the theory discussed but we will come back to this in the last paragraph this sub frame.

Normalization is to render something legitimate by exemplarity seems to make sense as strategy for the approval of policies to show they meet up with the expectation of the observer, pragmatic legitimacy. Authorization is legitimation by reference to authority. Authorization and normalization can be linked to implicit (Matten & Moon, 2008), or economic and legal, aspects of corporate social responsibility (Carroll & Shabana 2010). Therefore these two strategies seem to be sensible and coherent with the theory discussed. However, narrativization, a legitimation strategy that portrays others as winners, losers, hero's, adversaries, or culprits seems a bit od in the sense that this strategy loses the internal focus. Whereas, pragmatic legitimacy is focussed to gain approval for the activities practiced by the organisation.

Two strategies that were not used for pragmatic legitimacy did come forward in the attempt to retain moral legitimacy. Rationalization is legitimation by reference to the utility or function of specific actions or practices. It almost seemed some form of cognitive legitimacy. The only big difference for rationalisation as strategy to retain moral legitimacy is that the strategy fulfilled legal requirements and had nothing to do with fraud. The strategy of moralization is the legitimation that refers to specific values, and most logical in the attempt to retain moral legitimacy. Overall the use of legitimation strategies is mixed in their attempt to retain legitimacy and seem to be supplementary to each other.

A sixth strategy, denial, additional to the five legitimation strategies defined by Vaara, Tienari & Laurila (2006), comes forward when we look at the response of BMW-Group to potential industry-level illegitimate spill overs created by the diesel gate scandal. In the analysis we’ve already mentioned it and can be found in the appendices as well. What is found in multiple articles is that it often starts with just