Repairing legitimacy after crisis in a controversial industry : comparison after oil spills in developed and developing countries

(1)

Repairing Legitimacy After Crisis In A

Controversial Industry

Comparison after oil spills in developed and developing

countries

Master Thesis

MSc. Business Studies – International Management Supervisor: Dr. F. Ciulli

Second reader: Dr. J. Lindeque

Student: Willem Egbert Plasterk Student ID: 6181813

(2)

2 Statement of originality

This document is written by Student Willem Plasterk 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.

(3)

3 Abstract

The aim of the current study is to examine how western MNEs (multinational enterprises) operating in controversial industries attempt to restore damaged organizational legitimacy after a legitimacy-threatening event, more specifically if there is a difference in the response between developing and developed countries. The oil industry is chosen as the representative controversial industry, with major oil spills as representative legitimacy-threatening events. An embedded multiple-case study design is adopted with a total of four cases, of which two of a major oil spill in a developed country (the United States) and two in a developing country (Nigeria). Six strategies of repairing organizational legitimacy (Suchman, 1995) are used to identify differences and similarities in the strategies MNEs adopt in their attempt to restore lost legitimacy. MNE press releases and reports, NGO statements, and national and international newspaper articles are used in combination with the computerized software Nvivo to code and identify the strategies employed by the parties responsible for the spills. The results indicate a significant difference in strategies adopted by western MNEs operating in developed and developing countries; those operating in the former primarily, but not exclusively, adopt the two restructure strategies create monitors and watchdogs and disassociate (Suchman, 1995) to repair legitimacy. The latter on the other hand neglect these strategies and exclusively adopt several normative strategies of legitimacy reparation (Suchman, 1995), mainly deny, justify, and explain. Overall, MNEs operating in developing countries are more proactive in their attempts to restore organizational legitimacy, whilst MNEs operating in developing countries appear to react only after negative publicity in the media.

(4)

4

Introduction

The topic of organizational legitimacy has gained an increase in scholarly attention in the last decades. Defined as “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, p. 574), legitimacy is considered as a vital resource for business organizations (Ashforth & Gibbs, 1990).

Multinational enterprises (MNEs) operating internationally face numerous cross-country differences. The importance of legitimacy is particularly significant for firms who can be identified as such. The extent of the stakeholder support for an organization determines the firm’s legitimacy (Samkin, Allen, & Wallace, 2010). A firm needs legitimacy to survive regardless of the context of the industry or type of firm, and if it is not able to gain, maintain, and repair such legitimacy, it is eventually doomed to fail (Suchman, 1995; Human & Provan, 2000). MNEs operating on a global scale can be expected to face a surplus of issues regarding legitimacy compared to firms operating domestically. This is due to the complexity and differences in institutional environment the organization operates in. An

(6)

6

institutional environment is defined as “the set of fundamental political, social and legal ground rules that establishes the basis for production, exchange and distribution” (Davis & North, 1971, p. 6–7).

However, in some industries the concept of legitimacy bears more importance than in others. Especially in industries that are controversial such as the alcohol, tobacco or oil industry, companies have their legitimacy challenged on a regular basis (Palazzo & Scherer, 2006). Controversial industries can be seen as industries that produce “products, services or concepts that for reasons of delicacy, decency, morality, or even to fear to elicit reactions of distaste, disgust, offense or outrage when mentioned or when openly presented” (Wilson & West, 1981). The struggle for companies in such controversial industries to gain and repair legitimacy is ever more present.

Therefore, a key challenge faced by an MNE is the establishment and maintenance of legitimacy in the multiple and heterogeneous institutional environments they are operating in (Kostova & Zaheer, 1999). In this study the definition of an MNE is presented by Ghoshal & Westney (1993, p. 4) as a “specific organizational form that comprises entities in two or more countries, regardless of legal form and fields of activity of those entities, which operates under a system of decision-making permitting coherent policies and a common strategy through one or more decision-making centers, in which the entities are so linked, by ownership or otherwise, that one or more of them may be able to exercise a significant influence over the activities of the others, and in particular, to share knowledge, resources, and responsibilities with others”. In their study of legitimacy, Kostova & Zaheer (1999) state that MNEs introduce an element of complexity due to their international nature. Especially MNEs in controversial industries face issues regarding legitimacy. However, international management literature on topics such as legitimacy has been dominated by a narrow set of institutional ideas focusing on firms operating within borders of institutional environments,

(7)

7

whilst the very nature of these MNEs question the validity of these perspectives (Kostova, Roth, & Dacin, 2008). Kostova et al., (2008) propose that whilst legitimacy from an institutional perspective can be gained through isomorphism, MNEs engage in “actor specific manipulation and negotiation of their status aimed at social construction of their acceptance and approval. Achieving legitimacy in this context makes companies less, not more, similar” (Kostova et al., 2008, p. 1000). Thus it is difficult to achieve legitimacy through isomorphism for MNEs.

Beside the establishment and maintenance of legitimacy, restoring organizational legitimacy when social rules and norms have been violated in a legitimacy-threatening event is also of importance (Beelitz & Merkl-Davies, 2011). However, even though legitimacy in general is a popular topic in extant management literature, the focus often lies largely upon the act of gaining legitimacy, not repairing it (e.g. Elsbach & Sutton, 1992; Zimmerman & Zeitz, 2002; Rothstein, 2009; Reast, Maon, & Lindgreen, 2013). Yet, repairing legitimacy can be of vital importance for firms who have invested substantially in capital abroad, and require social and societal support to be able to successfully continue to conduct their business. Still, the fact that the general scientific focus on the subject of legitimacy is aimed at establishing legitimacy is reflected in a statement made by Suchman: “legitimacy acquisition strategies outnumber legitimacy maintenance and legitimacy repair strategies combined. This pattern both reflects the biases of existing legitimacy literature and indicates the need for further research” (Suchman, 1995, p. 599).

Whilst 20 years have passed since the previous statement was made, it is still valid today, and provides potential for the current research. The societal role of MNEs is gaining important each year, and legitimacy of MNEs with it. This increases the need to examine strategies adopted by MNEs to restore damaged organizational legitimacy, which has major unfulfilled research potential. This study makes several contributions to the literature. First of

(8)

8

all it contributes to literature regarding MNEs operating in controversial industries. Second, it contributes to the legitimacy literature, specifically on the limited legitimacy reparation aspect.

Another important aspect of this research is the international aspect. The institutional framework in which MNEs are embedded in host countries have a major influence upon firm strategies (Peng, 2002). Due to the international aspect of MNEs the strategies employed are often dependent upon subsidiary location. By creating a distinction in institutional framework through examination of crisis in a developed and developing country, the effects of host country institutional framework upon legitimation strategies can be studied. The difference between mostly formal and mostly informal institutions between these two types of countries can be expected to lead to several differences in legitimacy reparation strategy employed by MNEs. As there is little consensus on the definition of institutions and institutional theory, the concept of institutions used in this study is drawn from the notion of New Institutional Economics (NIE) which recognizes institutions as intermediaries through which information and transactions costs are reduced (Harris, Hunter & Lewis, 1995). The NIE is an economical perspective that attempts to extend classical economic theory by focusing on social norms and rules (Rutherford, 2001), which is central in the concept of legitimacy.

The third contribution made of for the international business literature, as the study is aimed at differences in institutional environment and its effect upon adopted legitimacy reparation strategies. Specifically, the distinction between western MNEs operating in developed and western MNEs operating in developing countries will be related to the strategies adopted to repair damaged legitimacy after a crisis of legitimacy.

(9)

9

How do MNEs in controversial industries attempt to repair organizational legitimacy after suffering damage due to a major crisis, and how does the location of the crisis (developed country versus developing country) affect strategies employed?

This study uses a multiple embedded case study design, in which four cases of crisis of legitimacy will be examined in relation to legitimacy strategies adopted by western MNEs operating in both developed and developing countries. The oil industry is adopted as the representative controversial industry, with major oil spills as the cause of the crisis of legitimacy.

Based on the findings the study provides several managerial implications for western MNEs attempting to repair damaged organizational legitimacy. First of all, the recommended adaptation of legitimacy reparation strategies is dependent upon the location of the crisis. When operating in a developed country the strategic restructuring strategies should be central, and adaptation of several normative reparation strategies should be done with caution due to possible negative effects these strategies can have upon firm legitimacy. Additionally when operating in developing countries the normative deny strategy should be combined with other strategies as it might not be effective as a long-term strategy.

The structure of the thesis will be as follows. It will begin by presenting literature on the topic of legitimacy, its relation to controversial industries, and strategies of legitimacy reparation. It will then specify the key challenges faced by MNEs regarding legitimacy, and the aspect institutional environments. Afterwards the research question and propositions are presented, after which the methodology is illustrated. Next, the results of the analysis will be presented. Finally, the results will be reviewed and theoretical and managerial implications will be drawn in the discussion.

(10)

10 2.

Literature

review

The following sections will illustrate the key constructs adopted in this study and the research gap leading to the formulation of the research question and the propositions. First the concept of legitimacy will be presented, followed by a detailed description of legitimacy reparation strategies, which is central in this thesis. Furthermore, the connection between legitimacy and controversial industries will be presented, afterwards the relationship between legitimacy and MNEs and key legitimacy issues that arise for those firms will be examined. Next, the institutional environment of developed and developing countries will be discussed. The literature review concludes with the research question and propositions based on these sections.

2.1 Legitimacy

“Legitimacy, a social judgment of acceptance, appropriateness, and desirability, enables organizations to access other resources needed to survive and grow”

(Zimmerman & Zeitz, 2002, p 414).

The concept of legitimacy has been present in business literature for several decades (Weber 1978; Parson, 1960), generally Weber is generally given credit for introducing legitimacy into sociological theory, and hereby into organizational studies (Johnson, Dowd, & Ridgeway, 2006; Ruef & Scott, 1998; Suchman, 1995; Deephouse & Suchman, 2008). Legitimacy is related to the ability of a firm or organization to meet social expectations for the environment in which it is embedded, public welfare, economic development and humanistic, political and civil rights (Dacin, Oliver, & Roy, 2007).

The topic of legitimacy gained an increase in academic interest in the past two decades, this onset was due to Suchman (1995). In the mid to late ’90s we can perceive a sharp increase in references made to legitimacy, also with the broader management literature.

(11)

11

This increase in attention subsequently led to more understanding and refinement of subjects, sources, processes and consequences of legitimacy (Deephouse & Suchman, 2008). In his article, Suchman (1995) focused on similarities and differences of strategic and institutional approaches to legitimacy. He defined three distinct types of legitimacy which gave the concept more explanatory power. The three types of legitimacy that he defines are: pragmatic, moral, and cognitive (Suchman, 1995). Pragmatic legitimacy is based on self-interested calculations of a firm's most direct actors; it occurs when the interest of its members are satisfied (Suchman, 1995). Suchman (1995) states that, in its most basic form, pragmatic legitimacy is a type of exchange legitimacy where firms gain support for their policies based on the expected value of that policy to those actors, however more complex forms also exist (Suchman, 1995). Moral legitimacy can be related to ethics, as it is not based on judgments of increased benefits, but rather on ‘doing the right thing’. These judgments typically concern opinions on whether firms’ activities promote social welfare, but are obviously not entirely without self-interest for the company (Suchman, 1995). The majority of statements made on organizational legitimacy refer to moral type of legitimacy (Claasen & Roloff, 2011). Finally, cognitive legitimacy refers to acceptance of the organization as being necessary or inevitable based on the taken-for-granted character of the firm.

Further distinctions can be made between external and internal legitimacy. External legitimacy relates to the acceptance of organizations or organizational activities by their social environment, whereas internal legitimacy focuses on acceptance of organizations by their internal actors (Suchman, 1995; Kumar & Das, 2007). There is a multitude of reasons for firms to seek these kinds of legitimacy. The most fundamental reason is that legitimacy is essential for a firm’s survival (Human & Provan, 2000). First of all, legitimacy enhances both stability and comprehensibility, which in turn also enhance each other. Incidental damage to legitimacy consequently also damages the stability of the firm and the comprehensibility of

(12)

12

its actions. Also legitimacy leads to persistence, making legitimate firms to require little ongoing investment to reach collective mobilization (Suchman, 1995). Legitimacy can help solve the collective action problem (Olson, 1965), and get individuals together to reach goals. Additionally, legitimacy leads to an increase in understanding which makes firms more predictable and more trustworthy (Suchman, 1995). Being perceived trustworthy is important to all firms, but above all to new firms or firms venturing into foreign locations for the first time. This is due to the liability of newness (Freeman, Carroll, & Hannan, 1983). Firms face challenges of being accepted by their environment for either the activity they enact, or for gaining validity for the firm as being a legitimate organization. In case of the oil industry, this can occur when firms enter new environments for natural resource seeking activities and hereby have to gain legitimacy for the activity. Suchman refers to gaining validity as the “task of (a) creating new, allegiant constituencies and (b) convincing preexisting legitimate entities to lend support” (Suchman, 1995, p. 587).

There are a multitude of possible causes and categories of sources that can engender damage to legitimacy. In his article on corporate brand reputation and brand crisis management, Greyser (2009) explores how to recognize corporate brand crisis that threatens an organizations reputation, and offers guidance in suggesting organizational actions. He offers a categorization of nine sources of reputational threats. These are: (1) product failure, (2) social responsibility gap, (3) corporate misbehavior, (4) executive misbehavior, (5) poor business results, (6) spokesperson misbehavior and controversy, (7) death of symbol of company, (8) loss of public support, and (9) controversial ownership (Greyser, 2009). Interesting is that for corporate misbehavior he provides among others the example of the Exxon Valdez oil spill, which is one of the selected cases.

(13)

13 2.2 Legitimation strategies

Even in the best situation, it is impossible for an organization to fully satisfy all actors and stakeholders it is involved with (Suchman, 1995). However, managers can significantly influence the way in which a firm’s activities are perceived as legitimate. In his article, Suchman (1995) presents strategies for gaining, maintaining and repairing the pragmatic-moral-cognitive legitimacy trichotomy. Whilst most companies do not seek to satisfy all the three types of legitimacy concurrently, the distinction enables managers to adapt the best practice for their current purpose. Even though the focus of this study will be on repairing legitimacy, it is important to know the distinction of such strategies compared to those aiming at gaining or maintaining legitimacy. In this section gaining and maintaining strategies will be briefly explained, after which repairing strategies will be examined in more detail.

The process of gaining or building legitimacy is a proactive enterprise; managers are aware of the need of legitimacy and have plans of gaining it (Suchman, 1995). There are three types of legitimacy building strategies: effort to conform, effort to select and effort to manipulate. These three strategies are employed in order to create legitimacy and acceptance. It is interesting to note that managers most often choose to conform, as it can be achieved by simply adjusting their own strategy, and does not contest the established institutional logics of the newly entered environment (Meyer & Rowan, 1991). However, as repairing is a responsive act, it is only possible for gaining legitimacy initially.

When aiming to maintain legitimacy firms are often facing a smaller challenge as it is depicted as far easier in the literature when compared to gaining or repairing legitimacy (Suchman, 1995). However Suchman (1995) presents three aspects that can make maintaining legitimacy problematic. First of all, audiences are heterogeneous. Legitimacy is a relationship between firm and stakeholders, and it is therefore not something a firm can simply buy. Rather it can be seen as “a condition reflecting cultural alignment, normative

(14)

14

support, or consonance with relevant rules or law’s” (Scott, 1995, p. 45). Second, stability often entails rigidity. The tendency for isomorphism and taken-for-grantedness reduce a firm's responsiveness to shifting conditions. Finally, institutionalization has the tendency to generate its own opposition. Legitimation projects attract attention, which can be hostile and criticize either the whole industry or an institutionalized firm

2.2.1 Repairing legitimacy

Repairing legitimacy is as stated the central focus of this thesis. The importance of the capability of firms to repair damaged legitimacy cannot be overstated. When faced with a crisis of legitimacy, organizations have to engage in legitimacy repairing behavior (Samkin et al., 2010). A crisis of legitimacy is an “unplanned event that results in considerable negative media publicity that causes harm to the organization (Samkin et al., 2010, p. 25).

The task of repairing legitimacy is similar to gaining legitimacy in many aspects. The main difference between the two is that repairing legitimacy is a reactive response to an unexpected crisis (Suchman, 1995). Such a crisis can occur due to an sudden decline in support of the firm, caused by negative publicity due to unexpected and unpredictable events. Once the firm recognizes the damage, it is often too late to address it with regular legitimation strategies. This can turn achieved legitimation successes into a constraint for management, as they suddenly have to recover stakeholder support (Suchman, 1995). Additionally, a legitimacy crisis may have a significant effect on the relation with associates. More specifically, managers may dissolve relations with trustworthy external partners in an attempt to avoid guilt by association (Suchman, 1995). These crisis tend to end in a reinforcing feedback loops as all involved actors tend to distance themselves in a similar manner. In practice this withdrawal of support most often results in even worse performance failures as resource flows are interrupted (Sutton & Callahan, 1987). Moreover, the negative loop intensifies if the resources that are cut of not only supply material resources but also

(15)

15

functions as a source of legitimacy. This indicates how a legitimacy crisis that initially affects the companies will affect many actors related to them as well.

There are several things firms can do to successfully repair damaged legitimacy. In fact, firms can achieve their goals by simply using legitimacy-building strategies as long as they still remain somewhat credible and connected to stakeholders and relevant actors (Suchman, 1995). Generally however, the firm should begin with addressing the issue at hand before using more general legitimacy strategies. The main reason why firms ought to address the issues first is to construct a perceived boundary between past actions, and stakeholder perceptions of the firm’s principles in general (Suchman, 1995). The literature identified three broad strategies firms can adopt when attempting to repair their legitimacy: propose normalizing accounts, restructure, and ‘don’t panic’ (Suchman, 1995). The following section will examine these strategies for repairing legitimacy in more detail. In Table 1 all legitimation repairing strategies as presented by Suchman (1995) are depicted.

Table 1 Repairing legitimacy strategies Repairing legitimacy strategies

General Normalize

Restructure

Don’t panic

Pragmatic Deny (normalize)

Create Monitors (restructure)

Moral Excuse/Justify (normalize)

Disassociate (restructure)

Cognitive Explain

2.2.1.1 Normalize

The first broad strategy is to formulate a normalizing account, separating the threat from the larger overall perception of the firm (Giacalone & Rosenfeld, 1989, as cited by Suchman, 1995, p. 598; Marcus & Goodman, 1991). There are four specific types of normalizing strategies managers can use to repair legitimacy. The first option is to deny the issue or

(16)

16

responsibility of the issue at hand, with the goal to relieve the surrounding concerns, or until the firm can accumulate a compensating payment (Suchman, 1995). This strategy can be used when managers aim to repair pragmatic concerns. As previously mentioned pragmatic legitimacy is based on self-interested calculations of a firm's most direct actors; it occurs when the interest of its members are satisfied. The major downside to this strategy however is that unless the denials are genuine, later discovery may destroy long-term legitimacy (Suchman, 1995).

Therefore, instead of denying the crisis managers can opt to excuse their actions. This second strategy questions the firms moral responsibility and can be used to repair moral legitimacy, which is about ‘doing the right thing’. However, in practice this strategy often results in blaming employees or external partners (Suchman, 1995). Additionally it suggests that there is insufficient managerial control which could in turn reduce moral legitimacy rather than repairing it.

The third strategy is to justify the crisis, redefining the events in such a way that they appear to align with the moral and cognitive beliefs of the firm.

The fourth strategy could help managers repair cognitive legitimacy by simply explaining the crisis in a way that the general perception of the firm remains positive (Suchman, 1995).

2.2.1.2 Restructure

Beyond these four strategies of normalizing accounts firms can also opt to repair through strategic restructuring, the second broad strategy (Pfeffer, 1981, as cited by Suchman, 1995, p. 598). This entails minor and well guided changes that converge with evenly well guided normalizing accounts. This can effectively contain damage by selectively stating that aspects of firm operations were defective or erroneous, and instantly move to a position in which the firm evidently attempts to overcome the problems. In time of crisis or controversy, such as an

(17)

17

environmental disaster, corporations use communication to restore organizational legitimacy by persuading involved actors that the structures and procedures of the firm align with social norms and rules (Elsbach, 2001). There are two types of restructuring, namely to create monitors and watchdogs, and disassociation (Suchman, 1995).

The first strategy create monitors and watchdogs entitles the firm to “post a bond” (Pfeffer, 1981, as cited by Suchman, 1995) which functions as a symbolic contribution and is done in order to convince key actors that it is safe to reengage doing business with the firm that initially has lost legitimacy. This is again a strategy which can repair pragmatic legitimacy. An example of this can be a firm inviting government regulation or implementing strict regulation.

The second restructuring strategy is disassociation. This strategy uses structural change to distance themselves, often symbolically, from bad influences. The most common form of this strategy is the replacement of executives, as it embodies the willingness of an organization to change (Gephart, 1978). Additionally firms can replace procedures, structures and even geographical locations of operations in order to create this symbolical distance to their past (Suchman, 1995).

In a study focusing on the role of restructuring to repair damaged legitimacy after financial fraud, disassociating from illegitimate business operations is proved to be effective in restoring legitimacy for firms punished for fraudulent reporting (Wang, 2010). Wang (2010) additionally states that investors value the creation of monitors more than the disassociation from organizational structures, whilst the value of disassociation from business operations has the same perceived value of the creation of valuable business operations .

2.2.1.3 Don’t panic

The third strategy is one that managers should adopt: to avoid panic (Suchman, 1995).Whilst this sounds obvious, research indicates that crisis lead to an impairment in decision making

(18)

18

processes which ultimately stimulates organizational failure (Staw, Sandeland, & Dutton, 1981). Whilst it is difficult to perceive it as a strategy rather than a state of mind of those involved, it is a necessity in order to successfully repair legitimacy.

2.3 Legitimacy and Controversial Industries

In the previous section the importance of the concepts of legitimacy and legitimacy reparation have been presented. In our society, issues related to production, distribution and consumption of products and services increasingly challenge firms’ legitimacy (Scherer, Palazzo, & Seidl, 2013). As mentioned in the introduction, especially firms in controversial industries often find their legitimacy challenged (Du & Vieira Jr., 2012; Reast, Maon, Lindgreen, & Vanhamme, 2013) resulting in the need to find new ways and new strategies to gain, maintain and repair legitimacy.

The specific industries to be classified as being controversial changes over time (Campbell, 2007) and depends upon culture (Fam, Waller, & Erdogan, 2004). However, as mentioned in the introduction, several industries, such as alcohol and tobacco, gambling, weapons and firearms and pornography (Reast et al., 2013) can generally be classified as controversial due to their unethical or offensive nature.

A major implication for firms in industries that can be classified as controversial is that the standards of scrutiny are far higher compared firms from other industries (Reast et al., 2013). This results in a restriction in the range of actions and activities such firms can adopt, as they are constrained by their bad reputation (Yoon, Gürhan-Canli, & Schwartz, 2006). For example, corporate social responsibility, or CSR, is a central concept in the field of legitimacy. When CSR activities are perceived as being sincere they will improve a firm’s image, but when the sincerity of the motives are questionable or ambiguous these activities will not only have no positive effect, they will have a negative effect on company image

(19)

19

(Yoon et al., 2006). The tool most firms use to gain, maintain or repair legitimacy, CSR, can thus be useless or harmful for firms in controversial industries. For these firms, that have a dubious reputation to begin with, this forms an extra challenge in their attempts to achieve the required levels of legitimacy. Defining CSR proves to be difficult as it is not only about what firms should do in society, but also about what they should be responsible for (Marens, 2004). Here, the formal definition of CSR used is “context-specific organizational actions and policies that take into account stakeholder’s expectations and the triple bottom line of economic, social and environmental performance” (Aguinis & Glavas, 2012, p. 933).

2.4 Legitimacy and MNEs

As mentioned in the introduction, one of the key issues faced by MNEs operating in multiple host countries is the establishment and maintenance of organizational legitimacy (Kostova & Zaheer, 1999). The ability of MNEs to address the expectations upon which their legitimacy is based is more complex due to two additional legitimacy challenges, which do not apply to firms operating domestically (Kostova & Zaheer, 1999). One of these challenges if the institutional complexity, which is derived from the institutional distance between the MNEs home country and the host country in which the MNE operates. Institutional distance can be perceived as the difference in regulative, cognitive, and normative institutions embedded in social actors in institutional environments, who participate in either economic, political or social exchanges (Kostova, 1999). The other challenge is based upon a MNEs multinationality, as it determines the institutional plurality and the quantity of institutional environments where the MNE and its subsidiaries must achieve organizational legitimacy (Kostova & Zaheer, 1999).

It should be noted that legitimacy is a continuous process rather than something one can obtain once and simply maintain. Also, it does not require unanimous agreement by all involved actors to exist; a firm can gain or repair its legitimacy whilst several groups or

(20)

20

individuals do not perceive it to be a legitimate business (Kumar & Das, 2007). This indicates that MNEs can be legitimate without having to convince all opposing parties into supporting them. Consequently, it simplifies the process for MNEs to become and remain legitimate in host countries. However, simply adapting to a particular set of social demands will not results in social acceptance of the MNE, but rather results in incongruity with other societal expectations (Palazzo & Scherer, 2006; Scherer & Palazzo, 2007; Scherer et al., 2013; Joutsenvirta & Vaara, 2015).

MNEs are generally known to different significantly from domestic firms. The main difference is derived from the consequences of multidimensionality and heterogeneity of the regions in which the firm operates (Doz & Pralahad, 1991). Kostova et al. (2008) state that whilst institutional theorists have identified multiple determinants of organizational legitimacy and characteristics of the legitimation process, examining the MNE can extend theories of organizational legitimacy. This is due to the fact that MNEs challenge some of the assumptions behind these theories (Kostova et al., 2008). The three factors shaping organizational legitimacy that are potentially challenged by the nature of the MNE are: “(1) the environment’s institutional characteristics, (2) the organization’s characteristics, and (3) the legitimation process through which the firm builds its perceptions of the organization” (Hybels, 1995, as cited by Kostova et al., 2008, p. 64). The general underlying mechanism challenging these factors is the aforementioned element of complexity brought forth by MNEs (Kostova et al., 2008). The pressure faced by MNEs to be legitimate for stakeholders across foreign locations lead to the necessity for MNE subsidiaries to translate multiple local regulative, cognitive and normative institutions and attempt to meet the legitimacy demands they raise (Marano & Tashman, 2012).

Whilst one can examine legitimacy of a subunit of an MNE per country or region it operates in, it is also possible to examine firm legitimacy as a whole: a global

(21)

meta-21

environment (Zaheer, 1995). This worldwide legitimation environment incorporates both host and home country influences, global media, and global activist groups (Zaheer, 1995; Kostova et al., 2008), and is central in this thesis. Firms can be considered legitimate when their organizational practices are perceived to satisfy social norms and expectations of their environment (Scherer et al., 2013), but for MNEs this environment can incorporate millions of individuals globally.

2.5 Institutional environments of developed and developing countries .

An important aspect of the IB literature are the characteristics of institutional environments. Patterns of institutional environments heavily influence economic behavior (North, 1990; Peng, Pinkham, & Chen, 2009) and organizational behavior (Biloslavo & Lynn, 2007). It was recently shown that institutions are more pervious to corporate strategic action than was previously assumed by institutional theory (Child & Tsai, 2005). Moreover, when operating in emerging economies, firms have the potential to implement their own preferred environmental policies if they offer significant stimulants to the local economy through investment and technology (Child & Tsai, 2005). These firms have bargaining power due to the benefits they bring, such as the creation of local employment which provides more favorable accommodation of institutional regulations to exploit legal loopholes or negotiate better terms (Leonard, 1988; as cited by Child & Tsai, 2005, 99). However, firms should not underestimate the importance of environmental policies, as securing legitimacy to remain operative may depend upon the ability to conform to environmental standards set by other influential, leading edge firms (Nehrt, 1998).

Organizational legitimacy lies at the core of institutional theory, which states that forces of organizational environment are often the result of culture, norms, values, and beliefs (Powell & Dimaggio, 1991). The large diversity between firms originating in developed and

(22)

22

developing countries is due to a variety of internal and external forces organizations have to deal with when forming and executing their strategies and actions (Peng, 2002). This can be explained using the institution-based view of business strategy, which states that firm strategy is formed due to pressures of the environment, more specifically by the institutional framework in which a firm is embedded (Peng, 2002).

According to Scott (1995) there are three dimensions of a firm’s institutional environment: the regulatory dimension, the normative dimension, and the cognitive dimension. The regulative dimension refers to the formally codified, enacted and enforced laws within a community or nation, and promotes certain types of behavior whilst restricting others (Scott, 1995; North, 1990). The regulatory dimension of the institutional environment focuses on the requirements of governments to comply with (environmental) laws and the enforcement to engage in environmental management programs, and additionally incorporates reliability and effectiveness of such enforcement mechanisms (Albareda, Lozano, & Ysa, 2007). The normative dimension refers to social norms, values beliefs and assumptions shared through a social obligations and expectations (Scott, 1995; Kostova, 1997). The cognitive dimension refers to a framework of meaning or beliefs which is socially constructed and provides the templates for action (Scott, 1995).

A careful examination of normative and regulative dimensions can present a notion of a country’s institutional environment that reflects incentives, guidelines, and restraints that influence how actors behave in the environment they are embedded in (Scott, 1995). Institutional theory states that in order to survive organizations have to conform to rules and social expectations of their environment (Dimaggio & Powell, 1983). Thus, firms can be seen as a system of activities embedded in complex social networks and environments (Meyer & Rowan, 1977), and their behavior is therefore influenced by the guidelines and restraints of the environment.

(23)

23

A grand distinction can be made between two types of institutions; formal rules and informal constraints (North, 1990). Formal institutions are formal legislative constraints, including political rules, judicial decisions and economic contracts. Informal institutions in turn are informal rules, including socially acceptable behavioral norms, which are embedded in the culture of the environment (Scott, 1995). Generally these two types are balanced in such a way that in countries with a strong formal institutional environment the informal institutions play a minor role, and vice versa.

However, as previously mentioned, MNEs are operating in a global meta-environment (Zaheer, 1995). This indicates that western MNEs operating in developing countries are both influenced by the strict formal institutional environment of the home country, as well as the informal institutional environment of the host. The strategies employed by these MNEs can thus be expected to differ from MNEs operating in developed countries as informal institutions only play a minor role for these firms. In western countries firms arise in highly institutionalized contexts. When looking within these western MNEs, many of the appointed positions, procedures enacted, programs created and strategies employed in organizations are due to public opinions, opinions of important actors (e.g. the board of directors, former CEO, etc.), knowledge shared in educational systems, laws, courts, and social status (Meyer & Rowan, 1997). Contradictory, in countries with less formal institutional environments, in regions such as Asia, and Latin America, a majority of publicly traded firms are family owned and controlled (Peng, 2010; Ahlstrom, Young, Chan, Bruton, 2004). The only reason this distinction is possible is due to significant differences in institutional environment.

In the article by Gaur et al. (2007) the effect of host country institutional environment on staffing strategy and subsidiary performance was examined. The results indicated that the institutional distance between the parent and the host country of the MNE significantly influenced MNE managerial strategy (Gaur et al., 2007). As the focus of the current study is

(24)

24

upon western MNEs operating in developed countries, the institutional distance between home and host is large and therefore has a significant effect on strategies adopted.

However there is no answer to the question what the differences in institutional environment between developed and developing countries are, as such differences are dependent upon the specific countries being compared. A possibility is to focus on developed western countries as having strict formal institutional environments, and developing countries as having informal institutional environments. By doing so the characteristics of strict and informal institutional environments can be used to create propositions and expectations. When these characteristics are linked to the concept of legitimacy, and more specifically legitimacy reparation, an opportunity for research is created.

2.6 Research question and working propositions

The importance of repairing organizational legitimacy has been presented in the previous sections. Especially in controversial industries MNEs appear to suffer from legitimacy issues, and face multiple issues when attempting to repair damaged legitimacy after critical incidents. Whilst literature linking strategies of legitimation to MNEs is limited (Kostova & Zaheer, 1999; Kostova et al., 2008; Scherer et al., 2013), literature on specifically examining legitimacy reparation strategies for MNEs is lacking altogether. This study adds to the literature by examining MNE legitimacy reparation strategies after critical incidents. Additionally the IB perspective of western MNEs attempting to repair damaged legitimacy in both developing and developed countries is a contribution to the literature. These different institutional environments combined with the examination of legitimacy reparation strategies adopted by western MNEs can provide insights for four main topics; legitimacy, MNEs, institutional environments, and controversial industries. Whilst there are numerous studies on

(25)

25

each of these topics there are none that combine them. Therefore, combing these aspects, the research question of this study will be the following:

How do MNEs in a controversial industries attempt to repair organizational legitimacy after suffering damage due to a major crisis, and how does the location of the crisis (developed country versus developing country) affect strategies employed?

A total of five propositions have been developed about the legitimacy reparation strategies MNEs are more likely to adopt when operating in either developed or developing countries, due the differences in institutional context. Of these five, four propositions have been developed on the formulation of a normalizing account which separates the threatening exposure from the firm.

First, denying can be expected to be more present in developing countries than in developed countries. As previously stated, a firm adopts the normative deny strategy to restore pragmatic legitimacy, based on self-interested calculations in an attempt to satisfy the firm’s most direct actors (Suchman, 1995). The major downside of this strategy is that if the denial is not genuine, additional damage will be done to a firm’s already impaired organizational legitimacy.

The preference for the deny strategy in developing countries can explained by the informal institutional environment. Due to the lack of strong political rules and judicial decisions firms have opportunities to implement their own environmental policy (Child & Tsai, 2005). In such cases firms can thus opt to deny responsibility for issues as they may declare them not to be important issues to begin with. Also, firms operating in developed countries will have to abide by strict environmental rules and legislation, and will sooner be held responsible for any inflicted damage. Additionally reporters, officials, and other involved actors can more easily be bribed into reporting facts or situations more positively

(26)

26

than is actually the case in developing countries (Heston & Kumar, 2008). Therefore the first proposition will be:

P1: MNEs attempting to repair their legitimacy following a legitimacy crisis in developing

countries will more often deny responsibility compared to MNEs facing a legitimacy crisis in developed countries

Second, excusing actions through questioning the moral responsibility of the MNE can be also expected to be more present in developing countries compared to developed ones. This is due to the fact that this strategy often ends up in blaming individual employees or external authorities (Suchman, 1995), which suggests a lack of managerial control which could damage moral legitimacy rather than repair it. Moral legitimacy, focused on ‘doing the right thing’ is central when dealing with social norms and expectations (Scott, 1995). Firms operating in strict formal institutional environments have to abide by numerous environmental policies and engage in environmental managerial programs (Albareda et al., 2007). After incidents related to the environment these firms can thus be expected not to blame individual employees but rather live up to social environmental expectations. Additionally, their managerial control is bound not only by norms and traditions but by strict (environmental) rules and judicial decisions. Putting the blame on individuals will therefore not suffice for MNEs operating in developed countries. However, firms operating in developing countries are not bound by such strict managerial control. It is the lack of managerial control which is more present in informal institutional environments that leads to the second proposition:

P2: MNEs attempting to repair their legitimacy following a legitimacy crisis in developing

countries will more often excuse their actions compared to MNEs facing a legitimacy crisis in developed countries.

(27)

27

Third, justifying can be expected to be regularly present in both developing countries and developed ones. When justifying, firms attempt to make events appear as aligned with moral and cognitive believes (Suchman, 1995), whilst it might not be the case. However as it can prove difficult to check the validity of these messages it can be used in different types of institutional environments. Both the managerial control found in strict formal institutional environments as well as the important role of social norms in informal institutional environments will encourage firms to align past events to positively reflect the firm. Therefore the third proposition will be:

P3: There is no strong tendency for the normative justify strategy to be more present for

MNEs attempting to repair their legitimacy following a crisis of legitimacy in either developed or developing countries.

Fourth, explaining can unlike the first three strategies be expected to be more often found in strategies employed in formal institutional environments. Explaining is the principle behind the concept of the discrete accident (Perrow, 1984). Accidents happen and are bound to happen again, but it is generally accepted that accidents can be prevented through better training, better quality of materials used, or improved design (Perrow, 1981). The concept of a discrete accident states that whilst there was an error or equipment failure, it can and it will be corrected and it will not happen again (Perrow, 1981). In a transparent and truthful firm, which are more present in strict formal institutional environments due to national transparency rules and policies (Donaldson & Kingsbury, 2013), admitting mistakes can be desirable as the truth will often come to light. Once that is done, firms can promise to make fewer or no more similar mistakes in the future, possibly restoring organizational legitimacy. Therefore the fourth proposition will be:

(28)

28

P4: MNEs attempting to repair their legitimacy following a legitimacy crisis in developed

countries will explain their actions more often than MNEs facing a legitimacy crisis in developing countries.

Besides the four normalizing strategies, a proposition is also be formulated regarding the two restructuring strategies. Both creating monitors and disassociation are expected to be more predisposed to MNEs operating in developed countries. Similar to the logic behind explaining, firms can selectively take responsibility for events by stating that some or several aspects of their operations were imperfect (Perrow, 1981), and can afterwards take active stand in curing those defects and eliminating the flaws from their operations (Suchman, 1995). The actions by which these flaws will be taken care of are the creation of monitors and disassociation. As previously mentioned, MNEs in strict formal institutional environments can be expected to be more likely to admitting flaws publicly. The obligation to adopt strict environmental policies and engage in in managerial programs in countries with strict formal institutions (Albareda et al., 2007) additionally promote the invitation of watchdogs to ensure correct adaptation of environmental actions. This is especially the case after major environmental incidents, such as oil spills, that affect many communities and has a large environmental impact. Therefore the fifth and final proposition is:

P5: MNEs attempting to repair their legitimacy following a legitimacy crisis in developed

countries will more often restructure, either through creating monitors or disassociation, compared to MNEs in developing countries.

It should be noted that the last strategy ‘don’t panic’ is not included in the propositions as it is more general, and is less likely to be used due to its relatively obvious character. Additionally not panicking is something that crucial actors are likely to do internally, but will not report to the media.

(29)

29 3.

Methodology

3.1 Research design

In order to answer the research question the oil industry is adopted as the representative controversial industry. This is because the oil industry provides an opportunity to study MNEs suffering from damaged legitimacy due to the legitimacy threatening events: the occurrences of large oil spills. Oil MNEs are responsible for damages caused by their drilling activities, and these spills offer a major source of damage to organizational legitimacy. There have been several large oil spills in both developed and developing countries, these environmental catastrophes provide an opportunity for empirical research on legitimacy reparation strategies. As previously noted, the quantity of research focused on repairing legitimacy is still very limited today. Additionally, using the oil industry enables making a comparison between strategies employed in crisis occurring in developed countries to crisis occurring in developing ones. This international aspect provides insights on how the institutional environment can affect legitimacy reparation strategies employed by western MNEs in the event of a major crisis.

In order to reach a thorough understanding of the relationship between crisis of legitimacy in the oil industry and strategies executed in order to repair damaged legitimacy, a qualitative research will be conducted. The rationale of choosing a qualitative over a quantitative methodology lies in the social aspect of qualitative research design, whilst quantitative research is rooted in statistics and mathematics (Gephart, 2004). The social aspect is important as legitimacy is a social phenomenon, and therefore requires a research design capable of including that aspect. Additionally the study adopts a multiple embedded case study research design, using four cases as units of analysis, as presented in figure 1. In the left column the two cases in the developing country are presented, in the right column the

(30)

30

two cases in the developing country. Important to note is that Exxon is the responsible firm for the one of the cases in both columns. The previously presented strategies of legitimation reparation (Suchman, 1995) will be used in order to find patterns among four cases. Observations will be conducted systematically, after which final conclusions can be drawn concerning the patterns found within and between the cases.

A key reason for choosing a multiple-case study design is that literal replication is desired, as multiple-cases are used to confirm similar results for the two cases within each country (Heiman, 2002). Additionally theoretical replication supports the aspect of the distinction between developed and developing country as it promotes contrasting results between these cases.

The current research examines firms’ responses to unexpected crisis. Additionally, as legitimacy is an abstract concept and difficult to measure (Zimmerman & Zeitz, 2002), being able to quantify it by using a quantitative research design could prove to be problematic. Finally, according to Pratt (2009), qualitative research is particularly valuable for addressing “how” or “why” questions, which is congruent with the research question of this thesis.

(31)

31

A case study is “an in-depth exploration from multiple perspectives of the complexity and uniqueness of a particular project, policy, institution, program or system in a “real life” context” (Simmons, 2009, p. 21, as cited by Thomas, 2011, p. 512). Because the information required to answer the research question is extensive and qualitative, similar to for example the research conducted by Zimmerman & Zeitz (2002), a case study methodology can be useful and effective. Several factors support the argument that it is appropriate in this research to conduct a case study method. Yin (2003) provides three situations when a case study has a distinct advantage. First of all, there is an advantage when a “how” or “why” question is asked. A second advantage can be perceived when a contemporary set of events is examined in the study, which in this case refer to the incidents that cause the oil spills. The third advantage of case studies is that they can be conducted when the researcher has little to no control over said set of events (Yin, 2003), as is clearly also the case in these incidents. When aiming to examine how oil firms attempt to repair damaged legitimacy after a crisis, these three conditions are met. Furthermore, a deductive research design will be used. The are several reasons for the use of deductive research. First of all deductive reasoning is “a theory testing process which commences with an established theory or generalisation, and seeks to see if the theory applies to specific instances” (Hyde, 2000, p. 83). Similarly, in this study theory has been developed prior to the research, and the propositions are made based upon the aforementioned theory.

3.2 Case Selection

In this research the oil industry is chosen to represent the controversial industries. As discussed in the literature review, the oil industry provides ideal conditions to examine legitimacy reparation strategies in a controversial industry. There are several reasons why the oil industry is particularly relevant to the study of restoring organizational legitimacy. First of all due to the location boundedness of resource extraction industries; oil companies often

(32)

32

have to conduct their business in highly different cultural, administrational, geographical and economical locations (Ghemawat, 2001). Whereas other controversial industries such as the tobacco or alcohol industry can operate globally, oil MNEs are bound to locations with abundant natural resources. This location boundedness create relevant challenges for oil MNEs and makes it particularly interesting to explore and compare responses to legitimacy crisis in developed and developing countries, as oil MNEs do not have the opportunity to operate regardless of the region. Additionally, legitimacy-threatening incidents have to be present in order to be able to examine how developed country MNEs attempt to restore lost or damaged legitimacy. In the oil industry, oil spills represent such incidents. Examining oil MNEs after a major crisis, in this case represented by oil spills, therefore provides a perfect platform to examine the legitimation restoring strategies employed by western MNEs in controversial industries. This is due to the fact that oil spills are major legitimacy threatening events as the MNE gains major local and international negative attention. Involved actors, local authorities and international environmental agencies for example, can try to shut down or restrict MNE activities after critical incidents. If such attempts are successful, they have catastrophic financial implications for the MNE. As MNEs require organizational legitimacy to operate, the ability to recover from these crisis is of the highest of priorities.

As mentioned before, this study adopts an embedded multiple-case study design. Four cases were selected based on several criteria. The first criteria consists in the size of the oil spilled. All the selected cases have a high quantity of barrels of oil spilled: the smallest spill examined in term of quantity spilled is estimated at nearly 100.000 tons of oil (see table 2 for detail), the largest is over 600.000 tons. The amount of oil spill signals the degree of damage to the environment and indirectly the threat the incident represents to the firms’ legitimacy. The larger the damage done to the environment, the more actors will be involved in the aftermath of the incident, and the more damage to the local economy. Larger environmental

(33)

33

damage also attracts attention from international environmental agencies such as Greenpeace who are able to inflict reputational damage on an international scale.

The second criteria, consistent with the research question, is represented by the location of the spill: spills in developed countries vs. spills in developing countries. In keeping with this criteria, two of the cases concern oil spills in the United States while the other two cases are oil spills in Nigeria. The responsible western MNEs are ExxonMobil in both a developed and a developing country, BP in a developed country, and Shell in a developing country. The distinctive focus on western MNEs operating in developed and developing countries is part of the research question and of the contribution of this study.

The third criteria concerns the country of origin of the firms responsible for the incident. The companies examined in the cases originate in either the Netherlands, the U.K. or the U.S.A. This enables me to be able to read most of the relevant information of the cases, as it will be in English or Dutch. Furthermore, it should provide more transparency of the strategies and operations of the MNE. This can be dubious for firms originating in developing countries due to differences in institutional environment; the strict formal institutional character of MNEs originating in developed countries ensure openness and transparency (Millar, Eldomiaty, Choi, & Hilton, 2005).

In the next sections key information on the cases will be presented and reasons behind their selection explained. In table 2 background information on the cases is presented, with detail on the location, date, size, and timeframe used per case.

3.2.1 The Exxon Valdez spill – 1989

The first case selected is the Exxon Valdez spill, which occurred in 1989 in Alaska, United States of America. There are several reasons to select this case, though the spill occurred over 26 years ago today. First of all, it is still considered to be one of the most devastating environmental disasters caused by humanity (“Frequently asked questions about the spill,”

(34)

34

n.d.) and the consequences are still noticeable today (Graham, 2003). Secondly, it is a very well-publicized environmental tragedy, providing sufficient data availability to be able to study it. Third, due to the massive amount of the spill the consequences were not only significant for the environment, but also for the American firm ExxonMobil. They suffered enormous financial and legitimacy damages, harming their market position and organizational legitimacy.

There are several arguments supporting the decision of using the Exxon Valdez case. First of all, it is a major spill in a developed country, of which not a lot are present in history. Moreover, spills with similar magnitudes in terms of damage and quantity of oil spilled in developed countries are in fact rare. Whilst there are some alternatives, they do not fit the current research criteria due to several reasons. For example, the Montara Oil Spill of 2009 is more recent and also occurred in a developed country (off the coast of Western Australia), but the spill also affected several developing countries such as Indonesia making the distinction between the developing and developed countries unclear (ABC news, 2009). Furthermore, the responsible firm PTT Public Company Limited originates in Thailand, which makes many reports and articles impossible to read due to language barriers. Additionally the country of origin is likely to make this emerging market multinational enterprise (EMNE) less transparent reducing validity and increasing difficulty. A second reason for the selection of the case is that the firm responsible for the incident is also responsible for a recent spill in a developing country, allowing for a comparison to be made based on differences in institutional environment.

The main focus of the data collected for the spill is on the two years following the initial disaster in 1989, as the other cases have a similar timeframe. However, some additional recent data providing insights on the aftermath of the incident are also included in the analysis, stretching the official timeframe to 25 years.

(35)

35 3.2.2 The Deepwater Horizon spill – 2010

The second case is the recent and famous Deepwater Horizon spill, otherwise known as the BP oil spill. The spill was located in the Gulf of Mexico and began on April 20th, 2010. One of the main reasons for the selection of this case is that is by far the world’s largest accidental oil spill in history (Robertson & Krauss, 2010). Similar to the Exxon Valdez spill, the consequences for operator BP were significant on many aspects. The reputational damage BP had to deal with provides a perfect opportunity to examine strategies employed in their attempt to repair their legitimacy. A second reason, again similar to the first case, is that due to the size and the fame of the spill, data is freely available and accessible. Third, due to the fact that the spill originated in water of the United States, the two cases in the developed country occur in the same country. This promotes literal replication as the cases are likely to be more similar than cases occurring in different countries. Additionally, the two cases in the developing country also occur in one country.

3.2.3 ExxonMobil spill Niger Delta– 2010

The ExxonMobil spill of 2010 in the Niger Delta is interesting to examine first of all because ExxonMobil is, as in the Exxon Valdez spill, the responsible operating firm. Due to the fact that it is the same firm responsible for two spills in different institutional environments, the validity of the results based on differences in institutional environment is increased. The spill was the result of a ruptured pipeline, resulting in both large scale environmental damage as well as economic damage for the local population (Vidal, 2010). Secondly, as mentioned before, the size of the spill was significant and similar to the size of the other spills (Vidal, 2010). This enhances the external validity of the cases as they were comparable in the amount of pollution to the environment.

(36)

36 3.2.4 Shell Bodo spills Niger Delta – 2008 2009

The fourth case consists of two large oil spills in Nigeria in 2008 and 2009, for which Shell was responsible. The first reason for choosing this case is that it occurred in the Niger delta, similar to the second ExxonMobil spill. This ensures that again similar to the first two cases the cases in developing country both occurred in the same country. Second, although exact numbers are not available, experts who studied the oil spills state that the spills could be as large as the Exxon Valdez spill in 1989 (BBC News, 2011). This makes this case similar to the other spills in case of tons of oil spilled and damage caused to local economy and the environment.

Table 2 Background information on the cases

Oil Firm/Vessel Location of Spill Date of spill Max Spilled

oil (in tons)

Economic state of the Host Country

Timeframe

Exxon Valdez United States, Prince William Sound, Alaska

24 March 1989

104,000 Developed 25 Years, with central focus on the first 2

Deepwater Horizon (BP) United States, Gulf of Mexico

20 April 2010 – 15 July 2010

627,000 Developed 2 Years

ExxonMobil Nigeria, Niger Delta 1 May 2010 95,500 Developing 2 Years

Shell Bodo Nigeria, Niger Delta 2008-2009 Unknown Developing 2 Years

3.3 Data collection

It is pertinent to collect secondary data in order to be able to provide an answer to the research question. By using secondary data, the actions taken by the MNE can be identified in a reliable manner. The data will be gathered from both national and international

(37)

37

newspapers, NGO press releases, firm statements and reports, and annual sustainability reports published by the responsible oil firms. In table 3 the sources used are presented.

It is important to notice that across the cases the data was gathered from a variety of sources. In order to determine the legitimacy strategies adopted by the MNEs first the annual reports were thoroughly examined and coded. The issue that arose however, is that in some cases the data was more limited concerning the adaptation of legitimacy strategies. In those cases it was a necessity to gather the required data from other sources, such as press releases and newspapers. As is clearly depicted in table 3, in the Deepwater Horizon case the quantity of the newspapers used is significantly lower compared to the other three cases. This is due to the fact that in their annual reports and press releases BP largely covered the events that occurred following the initial incidents in the Gulf of Mexico, reducing the importance of gathering additional data from newspapers. For other cases, such as the Shell Bodo spill, the annual reports did not provide sufficient information regarding the spills. In those cases newspapers were required to provide the missing data. A final note regarding the use of annual reports is that for the Exxon Valdez case the annual reports were not available for the two years following the incident (1989 and 1990), which is why they are not included in table 3.

The annual reports published by the MNEs were all gathered from official corporate websites. The database LexisNexis was used to collect the data from newspapers. MNE Press releases and sustainability reports were freely available due to the nature of the information. Sustainability (or annual) reports are specifically useful from the perspective of repairing legitimacy due to risk events, as it is a potential vessel for communicating these strategies to firm stakeholders and can be perceived as a way of managing reputational damage (Linsley & Kajüter, 2008). All data gathered in LexisNexis are gathered within a 2 year timeframe. First an overall search was conducted within the entire database of English newspapers, newswires

Repairing legitimacy after crisis in a controversial industry : comparison after oil spills in developed and developing countries