International Health and Safety within the multinational corporation : the effect of subsidiary autonomy, embeddedness, and internal knowledge transfer on the progression of Health and Safety initiatives within the MNC

(1)

CONFIDENTIAL

MASTER THESIS

International Health & Safety within the

Multinational Corporation

”The effect of subsidiary autonomy, embeddedness,

and internal knowledge transfer on the progression

of Health & Safety initiatives within the MNC”

Marijke van der Vliet University of Amsterdam

Student nr 5830044 Supervisor Dr. C. Williams Second Reviewer Dr. J. Pinkse

August 14, 2009

Abstract

This paper examines the influence of various organizational theories in ex-plaining the progression of CSR initiatives within the multinational. The value of CSR is increasingly being recognized in the last decade. In this pa-per a specific part of CSR will be highlighted namely Health & Safety (H&S). To explain the progression of these H&S initiatives this paper draws on three organizational theories; control management theory, resource depen-dency theory and the knowledge based view. From these theories, a model is derived in which three variables influence the progress of H&S initiatives, namely autonomy, embeddedness and knowledge transfer. A qualitative re-search is conducted covering 7 multinationals in which 16 international H&S initiatives were identified. Both qualitative and quantitative analysis show that embeddedness and knowledge transfer were related to the progress of an initiative. For embeddedness this relation seems to be negative, whereas for knowledge transfer a positive influence on progression exists. The relation-ship between autonomy and progress was less straightforward. The results have valuable implications for theory but also for H&S managers and MNC managers in general.

(2)

Preface

During my Business studies the complex relation between business and their social responsibilities concerning the environment they operate in, always intrigued me. With this paper I got the chance to explore CSR and in specific H&S within the MNC. The unique possibility to get access to 7 multinationals and be able to talk with the specialists who were in the middle of H&S sharing their knowledge and ideas made this research a very joyful and interesting journey.

I would like to use this opportunity to express my gratitude to all the people who were willing to make time in their busy agendas to participate in this research. I learned a lot from the conversations and I really appreciated that each of them made time available for me. Without your contributions, this Master Thesis would not have been possible.

Furthermore, I would like to thank my supervisor, Christopher Williams for his dedication and his enthusiasm during the development of this master thesis. It was very much appreciated.

With kind regards to every reader,

(3)

List of Figures

1 Conceptual Model . . . 21

List of Tables

1 Overview of interviewees . . . 26

2 Examples of Interview questions . . . 26

3 Evaluation of Dependent Variable . . . 28

4 Overview of participating MNC’s including A,B,C coding . . 31

5 Overview of key quotes per Research Variable . . . 55

6 Results per initiative for Dependent Variable (Progress) . . . 56

7 Results per initiative for Independent Variable Autonomy . . 56

8 Results per initiative for Independent Variable Embeddedness 57 9 Results per initiative for Independent Variable Knowledge Transfer . . . 57

10 Exemplary Contingency Table showing amount of occurrences per combined state . . . 58

11 Actual data for Autonomy and Progress . . . 59

12 Expected data for Autonomy and Progress . . . 59

13 Actual data for Embeddedness and Progress . . . 60

14 Expected data for Embeddedness and Progress . . . 60

15 Actual data for Knowledge Transfer and Progress . . . 61

16 Expected data for Knowledge Transfer and Progress . . . 61

(6)

1 Introduction

1.1 Background

This paper aims to shed light on the dynamics inside the MNC concerning CSR initiatives. By combining CSR theories and organizational theories, this paper wants to contribute in bringing greater understanding about or-ganizational factors inside headquarter and subsidiary relationships and the implications they have on the implementation of CSR initiatives.

In the last two decades, the concept of CSR has been more and more ra-tionalized and became associated with broader organizational goals such as reputation and stakeholder management (Lee, 2008). It has developed from a purely economic point of view (CSR as maximizing returns to sharehold-ers; Friedman 1972) to, a comprehensive ”proactive social responsiveness view” that shows the long-term role in a dynamic social system (McGee, 1998). The CSR theories gave a comprehensive inside in how and why the motivations for CSR activities within the MNC shifted but it did not fully explain the process of CSR implementation (Husted & Allen, 2006). As such, and because CSR had become more strategic, scholars began to use organizational theories to examine why and how MNC’s implemented CSR (McWilliams, et al., 2002; Waldman, 2004; Husted and Allen, 2006).

This theory development has created an organizational view of CSR. However, there are still numerous unresolved theoretical and empirical issues regarding the strategic implementation of CSR that could not be resolved by the theories mentioned before (McWilliams, et al., 2006). For example, the inability of describing CSR strategies and modeling or assessing the effects of CSR on the firm and stakeholder groups (McWilliams, et al., 2006). This illustrates that there is still little research about the effectiveness and practi-cality of CSR activities. Husted & Allen (2006) were one of the first scholars who examined how MNC structure influenced the role of the subsidiary in CSR practices. However, the exact role of the subsidiary in relation to CSR implementation was not explained. Therefore, this research will focus on the influencing factors on the development of MNC’s CSR initiatives and what role the relationship between headquarters and subsidiaries play in this process.

One way of examining the influencing factors, regarding the involve-ment of subsidiaries in CSR issues, is studying research done by organiza-tional scholars. There is extensive research about parent-subsidiary rela-tionships and the central issue has always been control of headquarters, and the amount of autonomy the subsidiary is given (Birkinshaw & Morrison, 1995). Moreover, in most corporations, there is an internal competition for control and as a result, headquarters need to balance the distribution of

(7)

their power (Birkinshaw & Hood, 1998). Different scholars acknowledge the possible value creating capabilities of the subsidiary. However, first the orga-nization has to create an environment that enables the subsidiary to initiate unique and independent initiatives (Birkinshaw, 1997). Furthermore, it is argued that subsidiaries, because they are often located in another country, have a unique business network that can give them power within the MNC (Andersson, Forsgren & Holm, 2007). This presence of dense exchange re-lationships with local partners can be referred to as network embeddedness (Forsgren & Holm, 2007).

Research has shown that network embeddedness has a positive influence on the performance of subsidiaries. This can influence, in turn, the nego-tiating power of the subsidiary concerning their MNC’s future investments (Andersson et al., 2001, 2002). Because CSR nowadays is more being seen as a strategic factor within the MNC, the before mentioned influencing factors, autonomy and embeddedness may not only be influential in core business activities, but in CSR development as well. Therefore, in this research, an attempt is made to examine the influence of autonomy and embeddedness in CSR practices within the parent-subsidiary relationship. It is important to consider what affects this influence. In other words, how can the effect that autonomy and embeddedness have on CSR practices be explained.

In this research, the Knowledge Based View (KBV) is used to explain the possible influence of autonomy and embeddedness on CSR. The main reason for this is that there is a growing consensus about the importance of the transfer and utilization of knowledge and the relation with autonomy and embeddedness influence (Yamin, 2005). Moreover, the way knowledge flows and the different types of knowledge can have influence on the manner in which an MNC initiates certain activities and can even influence how suc-cessful they are (Grant, 1996; Mudambi, 2002). Because CSR is nowadays becoming more a strategic asset, it is plausible to consider a link between CSR and the KBV.

One main problem in CSR research however, concerns defining the effec-tiveness in terms of results and progression of CSR initiatives. Without a good definition, it is difficult to compare results across studies, constraining the ability to understand the implications of CSR activity (McWilliams et al., 2006). To overcome this difficulty, this research will focus on a sub-part of CSR namely, Health & Safety (H&S).

H&S within the company is not new; it exists much longer than the discussion about CSR. However, like the sudden interest for CSR, H&S has gained more attention and focus as well. Moreover, societal and economic changes, like globalisation, have increased the need for global integration with regional and global actors ( e.g. WTO, VN, ILO). This has triggered

(8)

a need for H&S standards across countries. Besides institutional factors, there are also firm factors that have influenced H&S practices.

For example, competitive pressures on the MNC are ever more intense and to keep a good company name an effective international H&S policy is vital. The MNC’s are responsible for their actions and they are obliged to do everything in their power to keep all their stakeholders safe and healthy. MNC’s can create bad publicity by neglecting their H&S responsibilities, which can eventually have a negative effect on the company value. Shell Nigeria is an example of the effect publicity regarding H&S policies inside the company can have. Because of these changes and pressures, H&S has gotten more attention in the last couple of years and therefore it is very interesting to look at the implementation of today’s international H&S initiatives.

As mentioned, the influence of autonomy, embeddedness and knowledge transfer on CSR (H&S) initiative is going to be examined. These variables are important for H&S for the following reasons; (1) the amount of au-tonomy that the subsidiaries receive to develop and implement their own programmes could be very influential for the progression of the H&S ini-tiatives. Moreover, when subsidiaries are given a lot of freedom this could also affect the way in which they internalize initiatives developed elsewhere (corporate or subsidiary). In other words, by increasing autonomy the MNC could increase the empowerment of the subsidiaries, which could have a ma-jor impact on the progression of international initiatives. (2) How much the subsidiary is embedded could also be important for H&S initiatives be-cause subsidiaries, when they are embedded, may be very much influenced by their local environment. For example, local culture and the local insti-tutional quality, like local laws, can affect H&S activities. Furthermore, the function or role of the subsidiary is very important, because this could influ-ence how embedded the subsidiary is and eventually what the impact is of local/national institutional factors. (3) knowledge transfer could also be im-portant for international H&S initiatives. The headquarter understanding of the local environment, subsidiaries understanding of headquarter initia-tives and the status of reporting are all examples of knowledge based factors that could influence for H&S initiatives.

1.2 Research Question

Drawing on a sample of Dutch, Dutch/English, Dutch/French and German multinationals, this paper looks at the role of organizational theories in the progress of international H&S initiatives. As such, the central research question in this paper is: ”How do subsidiary autonomy, embeddedness and knowledge transfer affect the progress of international H&S initiatives within the MNC”. Because of the explorative nature of this research, a qualitative research approach was chosen. By conducting semi-structured interviews

(9)

with key persons in the H&S area, some unique information was generated. The results show that autonomy, embeddedness and knowledge transfer all contributed in explaining the progress of international H&S initiatives. In addition, other influencing variables emerged from the interviews, which also will be discussed. This study contributes to both organizational theory and CSR theory, by addressing how elements of the organizational theory can explain CSR practices. Furthermore, this study gives some practical implications for managers about how H&S initiatives can be enhanced.

1.3 Structure

The structure of this research will be as follows; firstly, the theoretical back-ground will focus on the literature on CSR and discuss the key findings. Furthermore, three leading theories in organizational research are elabo-rated upon from which three variables are derived. Secondly, the method-ology will be discussed, followed by the presentation of the results. Finally, the discussion, conclusion, and suggestions for future research will be given.

(10)

2 Theoretical Background

The theoretical background will be structured, as follows; this Section will shed light on the relevant academic literature for this research. Section 2.1 will give an overview of the literature on CSR with a focus on the historical background of CSR, the definition of CSR, global versus local CSR, CSR and the MNC and lastly the subpart of CSR, Health & Safety is explained. Sec-tion 2.6 will present three competing organizaSec-tional theories, management control theory, resource dependency theory and knowledge based view.

2.1 Historical background of CSR

The development of corporate social responsibility has a long and eventful history. Especially in the last fifty years there has been a lot of scholarly research on social responsibility (Carroll, 1990). The concept of CSR, and especially how it relates to organizational goals, has been steadily evolv-ing ever since the concept was introduced half a century ago (Lee, 2008). The earlier views concerning the responsibilities of the firm used the term shareholder approach. The shareholder approach implies that ’the social responsibility of business is to increase its profits’ (Friedman, 1962). Public intellectuals like Milton Friedman strongly opposed the idea of CSR for the specific reason that it imposed an unfair and costly burden on shareholders (Friedman, 1962). He also believed that corporate managers did not have the right skills and expertise to deal effectively with social problems (Fried-man, 1962). Although, this view has had strong support, the stakeholder approach was given more and more attention as well (Marrewijk, 2003).

In the last two decades, the concept of CSR has been more and more rationalized and became associated with broader organizational goals such as reputation and stakeholder management (Lee, 2008). As such, organizations were not only accountable to its shareholders but should also balance the various interests of the stakeholders that can influence or are influenced by the activities of organizations (Freeman, 1984). These philanthropic views might be the foundations of CSR, but the variety of approaches to corporate responsibility clearly show that CSR is new and a separate phenomenon (Marrewijk, 2003).

Organizations are forced to fundamentally rethink their position and have to incorporate the complex societal context they are part of. Although this obligation might be viewed as an ’unfair and costly burden to sharehold-ers’ (Friedman, 1962) the vast majority of studies and popular literature on CSR state that CSR positively affects the bottom-line performance of a cor-poration (Margolis and Walsh, 2003; Orlitzky et al, 2003). Moreover, prior research also found that negative CSR can have a damaging effect on over-all product evaluations, and positive CSR associations can increase product

(11)

evaluations (Brown & Dacin, 1997). Consequently, even institutional in-vestors have given their support to the environmental aspect of CSR (Lee, 2008). The rationalization of the CSR concept caused a growing number of corporate managers that are convinced that CSR can contribute to the financial performance of corporations (Lee, 2008). As a result companies are increasingly adopting CSR activities and put them high on their business agenda.

In sum, alternative perspectives on the role and place of companies in the broader social environment have created multiple conceptualizations of CSR, ranging from a purely economic point of view (CSR as maximizing returns to shareholders; Friedman 1972) to a comprehensive ”proactive so-cial responsiveness view” that shows the long term role in a dynamic soso-cial system (McGee, 1998). Sen and Bhattacharya (2001) find the societal view of CSR adopted by Brown and Dacin (1997) to be the most comprehensive, namely ’as the company’s status and activities with respect (to responsive-ness) to its perceived societal obligations’. This indicates that, CSR, being a societal concept, can no longer be ignored by the MNC. However, with-out objective and behavioural indicators of CSR, research will not be able to predict the direction of socially responsible business practices of a cor-poration (Lee, 2008). Therefore, it is paramount to develop objective and behavioural indicators that could verify if corporations are doing what they say they are doing (Lee, 2008). To make CSR more objective it is impor-tant to be more specific about what CSR is. In the following section, the definition of CSR is being elaborated.

2.2 CSR - definition

In the last paragraph the evolution of CSR towards a rationalized concept was discussed, however to be able to examine CSR, a clear definition concern-ing this concept is needed. Unfortunately, there is no universal agreement on how to define CSR. In fact, CSR has suffered various and even contradictory characterizations (Garriga & Mele, 2004). Marrewijk (2003) argues that in general, CSR refers to company activities - voluntary by definition- showing the inclusion of social and environmental interests in business operations and in interactions with stakeholders.

This should be viewed as the broad definition of CSR (Marrewijk, 2003). Foran (2001) argues that CSR can be defined as the set of activities and behaviours that firms develop and adopt towards their labour force, towards the environment in which their operations are embedded, towards authority, and towards civil society (Foran, 2001). A similar definition of CSR, which is used by Husted and Allen (2006), is the firm’s obligation to react to the externalities created by market action (Sethi, 1990). Externalities are

(12)

positive or negative impacts of the production of the firm on the utility or production of a third party.

2.3 CSR - Local & Global

Furthermore, within the area of CSR a distinction can be made between global and local CSR. Several scholars examined the difference between lo-cal norms and universal norms. They agree with each other that a distinc-tion between the universal and the particular responsibilities of corporadistinc-tions is paramount (Spicer, 2004; Reed, 2002). According to Husted and Allen (2006) ”local’ CSR deals with the firm’s obligations based on the standards of the local community, whereas ’global’ CSR deals with the firm’s obliga-tions based on those ’standards to which all societies can be held” (Husted & Allen, 2006, p. 840).

The question whether subsidiaries tend to adopt CSR practices of the home country of their parent firm, or develop their CSR activities to the host-country context in which they are located, indicates another impor-tant distinction between local and global CSR (Muller, 2006). If companies choose to follow a global CSR strategy and effectively involve the subsidiaries to implement them, MNC’s have the possibility to function as mechanisms for ’upward harmonization’ of CSR standards internationally (cf. Tsai & Child, 1997; OECD, 1999). On the other hand, if companies emphasize the dialogue between local stakeholders and responsiveness to local institutions and acknowledge the importance of endogenous CSR development by the subsidiary, the possibility exists for a more pure ’responsive’ CSR strategy. However, one that could also be classified as ’reactive’ and potentially aimed at the minimum level required by law (cf. Meyer, 2004; OECD, 1999).

2.4 CSR - MNC

In the last decades, the MNC is more seen as a federative institution (Ghoshal and Bartlett, 1990). This may have implications on how CSR policies are being implemented within the company. However, research concerning the organizational issues of CSR within the MNC has just recently begun to de-velop (Christmann, 2004; Watson and Weaver, 2003). For instance, Husted and Allen (2006) examined the relationship of global and local (country-specific) CSR to international organizational strategy using the Bartlett and Ghoshal typology. In this research they explored the way in which the three product- market organizational strategies (multidomestic, transna-tional, global) influenced the integration and responsiveness of the MNC regarding CSR activities. The findings suggest that all MNC types place similar importance on global CSR issues. However, the multidomestic and transnational MNC’s place greater importance on country-specific CSR than do global MNC’s (Husted & Allen, 2006). According to this study, ”CSR

(13)

seems to conform to the MNC organization strategy established for product-market activities”(Husted & Allen, 2006, p.846). As such, Husted and Allen (2006) concentrated on the relationship between the product- market orga-nizational strategies and the implementation of local or global CSR.

The role of the subsidiary in the host country is of crucial importance, being in the triad of the nation-state, the MNC and the subsidiary (Amba-Rao, 1993). Being in the position of close contact with the host government, the subsidiary can act as a mediator between the parent corporation and the government in finding a solution for situations where specific differences are present (Trotter, 1989). Concerning CSR activities, subsidiaries could pro-vide input to influence centralized decisions, for example, by identifying local conditions and local needs environmental impact and with that decreasing government intervention (Naor, 1982). However, the power and impact of the subsidiary depends on how the MNC is organized. Using the typologies by Harzing (2000), multidomestic firm, the transnational firm and the global firm, this would mean that the subsidiary has operations that are indepen-dent, interdependent and intermediate dependant, respectively (Amba-Rao, 1993). To conclude, Amba-Rao (1993) stress the importance of site-specific knowledge and the experience of the subsidiary and that ’it is a significant contribution to the ethical and social considerations of the organization’.

The latter research has contributed to the literature by making a dis-tinction between local and global CSR and by showing the usefulness of institutional theory in explaining CSR policies by firms. However, the way in which CSR activities evolve and what makes them successfully progressed throughout the multinational is not specifically addressed. In a pilot study by Muller (2006) the role of subsidiaries and how they can contribute to CSR is more explicitly examined. The results of this preliminary exploratory re-search show that the driver behind autonomous subsidiaries proactive stance seems not to come from being local responsive but that subsidiaries draw from the parent companies overall CSR vision. In other words, subsidiaries with greater autonomy have proactive CSR strategies that are strongly in-fluenced by the home country and international policies, instead by the host country context (Muller, 2006). Both the pilot study by Muller (2006) and the study by Husted & Allen (2006) argue that CSR policies within the MNC are more likely to be global strategies rather than local. Furthermore, these global strategies tend to be better implemented when the subsidiaries are given autonomy (Muller, 2006).

This result differs from the management control theory, where it is stated that higher autonomy results in more innovative activities but less adapta-tion of activities that come from headquarters (Birkinshaw & Hood, 1998; Young and Tavares, 2004; Anderson, Forsgren & Holm, 2007). CSR is a

(14)

very broad topic and therefore difficult to compare between companies. In this paper, a specific part of CSR was chosen in order to make a comparison more reliable, namely H&S.

2.5 CSR - Health & Safety

Companies have seen an increase in the amount of stakeholders they are accountable for, it evolved from only the shareholders towards a whole va-riety of different stakeholders (employees, local communities, environment, NGO’s, governments). These different stakeholder groups have different in-terests and therefore often require different types of CSR activities to look after the various needs. CSR practices vary from companies reducing the pollution of the air, employment of minority groups, developing educational programmes for the community the companies are located in or, improving industrial H&S - these and other programs designed to improve quality of life are all part of social responsibility (Backman, 1975).

In this research, the focus will be on the latter program - Health & Safety (H&S). The International Labour Organization (ILO) and the World Health Organization (WHO) decided to share a common definition of occupational health since 1950. The definition was revised in 1995. The definition adopted by the Joint ILO/WHO Committee on Occupational Health states: ”Occu-pational health should aim at: the promotion and maintenance of the highest degree of physical, mental and social well-being of workers in all occupations; the prevention amongst workers of departures from health caused by their working conditions; the protection of workers in their employment from risks resulting from factors adverse from health; the placing and maintenance of the worker in an occupational environment adapted to his physiological and psychological capabilities; and, to summarize, the adaptation of work to man and of each man to his job.” *(internet site, WHO)

Unlike for example environmental issues, every company has to deal with H&S. As such, by implementing profound and effective H&S policies, corpo-rations have the ability to create competitive advantage. The ’Enterprise for Health’ Corporation , who argue that the increasing globalisation and im-proved technology are coming with great challenges for European companies, explains this more clearly. In order to have competitive advantage, it is no longer enough to base corporate decisions on shareholder value alone. They also see that the important factor for the survival capability of companies is the competence of management to communicate with other stakeholders, namely, the employees and society. Furthermore, they stress the importance of a corporate culture based on partnership and an exemplary health policy that creates the foundation for innovative potential and flexibility whilst safeguarding long-term competitiveness.

(15)

Moreover, it is said that the main objectives for managing H&S risks is to meet legal requirements, to minimise impact of potential accidents, to prevent accidents and work related illnesses and to meet other stakeholder expectations (H&S Executive UK, 1993). Throughout the MNC’s accidents, work related illnesses happen but are often preventable. The primary cause of these H&S issues are often human or technical failure but have often have their root cause in organizational and management failures (H&S Executive UK, 1993). Therefore, it is paramount to have management processes in place to counterattack these issues. Managing the risks regarding H&S requires the identification of hazards, events and circumstances and assess how these can affect the outcome towards the H&S situation (Hardjono & Bakker, 2002). As such, in this research the focus will be on one specific part of CSR, international H&S initiatives.

Scholars have used organizational theories, to be able to explain certain business situations inside the MNC. In this paper, organizational theories are used to see if they can explain the dynamics of non-financial activities within the MNC as well.

2.6 Competing Organizational Theories

In this paper the three variables autonomy, embeddedness and knowledge transfer and their influence on the progression of H&S initiatives throughout the organization are being explored. Although these variables all come from different schools of thought, these three variables have shown their influence on various kinds of organizational processes. The variables derived from the competing theories, management control theory, resource dependency theory and the knowledge based view and these theories made their contributions in explaining the dynamics between headquarters and subsidiaries. In the following section these theories are being elaborated upon and it is explained how they all share the same goal; explaining headquarter and subsidiary relationships and the effect on MNC performance.

2.6.1 Management control theory - Subsidiary Autonomy

Since 1990, a rapid increase in the development of large global corporations has taken place (Bartlett & Ghoshal, 1993). Due to new environmental demands, such as the globalisation of competition, markets, and technol-ogy and the related economic and social consequences, changes in strategy, structure and management have occurred (Bartlett & Ghoshal, 1993). With these changes the existing paradigms of strategy, organization and decision-making, that explained an earlier form of the corporate model, seemed not as useful as before (Hamel & Prahalad, 1993). Therefore, Ghoshal and Bartlett (1990) propose that instead of the classic M-form organization (Williamson, 1975), the MNC should nowadays be viewed as an inter-organizational

(16)

net-work, also called the network view of the MNC. This means that the MNC must be seen as federative rather than a unitary organization. Because the MNC is embedded in various economic and institutional domains, it simply cannot be governed as a unitary organization anymore. In other words, there has been a shift from a centralized view towards a more decentralized view of the MNC in which subsidiaries are given more power (Hedlund, 1993).

Besides that, subsidiaries are given more decision-making rights; the subsidiaries themselves are striving for more autonomy and charter as well, to be able to fully grasp the opportunities the environment and their specific knowledge give them (Birkinshaw & Hood, 1998). MNC’s are said to be dispersed structures of power and the authority of top management does not result in hierarchical power automatically being the control mechanism (Anderson, Forsgren & Holm, 2007). MNC’s can be seen as much more loosely coupled than the traditional view of the MNC’s where they are being displayed as extremely hierarchical (Anderson, Forsgren & Holm, 2007). When the MNC is looked at as being a federation, the headquarter is seen as just one part among all the other parts existing within the organization. The networks, which both headquarters and the subsidiaries are part of, are bases that are paramount for intra-organizational power (Provan, 1983).

As such, transnational management is seen as critical for maintaining innovation and growth for the MNC (Ghoshal & Bartlett, 1990). Seen in this light, the hand of the nature should determine the most appropriate control mechanism and the role of the subsidiary, especially the complexity in its local environment connected with the level of its resources and capabilities (Ghoshal & Nohria, 1997).

The consequences of subsidiary autonomy were highlighted by Young and Tavares (2004) in which the innovative potential of subsidiaries, lat-eral communication and networked structures and changes in mandate were recognized as autonomy outcomes. The choice of the amount of autonomy that is given to a subsidiary is an important factor within the management control theory (Thorne & Hilton, 2006). Although decentralisation gives the subsidiary more power and decision rights it can at the same time erode the control of the headquarters, which makes it more difficult to, for example, subvert rent-seeking behaviour of subsidiaries (Birkinshaw & Hood, 1998).

To conclude, the degree of decentralisation can have a positive but also a negative influence on the progression of initiatives. It can be expected that decentralisation is going to have a positive effect on the innovativeness of the subsidiaries, translated into unique H&S activities. However, because of the lower control of headquarters, the progression of one initiative throughout the MNC could be more difficult. This leads to the following proposition:

(17)

Relevance to progress of the H&S initiative: ’The greater the local Autonomy of the subsidiary the less H&S initiatives will progress throughout the organization’

2.6.2 Resource Dependency - Subsidiary Embeddedness

The second variable that is being examined is embeddedness, which derives from the resource dependency theory. In the following section the back-ground of this theory and the influence of embeddedness is elaborated upon. A consequence of the changing roles inside the MNC is that the subsidiaries gain a power base in the host government (Ghosal and Bartlett, 1990). This closeness towards the local host network is called subsidiary embeddedness. Which can be described as, ’mutual adaptations in developing processes and products between a focal subsidiary and a small number of host coun-try counterparts (e.g. customers, suppliers, research centres and universi-ties) with whom the subsidiary has developed lasting business relationships’ (Yamin and Forsgren, 2006, p.173).

Many scholars emphasize the important role that external linkages play in the development of the capabilities of subsidiaries (Birkinshaw and Rid-derstrale, 1999; Forsgren et al., 1999; Anderson and Forsgren, 2000). Be-cause these external linkages can be helpful in facilitating knowledge flows from a variety of exchange partners in the local business and other networks to the MNC subsidiary (Yamin and Otto, 2004).

It seems that the subsidiaries are of great importance to the MNC be-cause they can contribute to the innovative capabilities of the firm. It is however paramount to create an environment within the MNC where sub-sidiaries are given the opportunity to initiate such entrepreneurial initiatives. This ability to value, assimilate and apply new knowledge, also known as ’absorptive capacity’ is also being put forward by the research of Pinkse, Kuss and Hoffmann (2009) in which it is argued that it is important for the MNC to develop a shared sustainability - specific absorptive capacity by education and to source necessary information. Ghoshal and Bartlett (1988) examined the factors that influence the ability of subsidiaries to carry out different innovative tasks.

These innovative tasks can be categorized in creation, adaptation, and diffusion. In the literature, four important processes can influence these dif-ferent innovative tasks. These processes are normative integration, internal communication, local autonomy, and the local slack of resources.

According to Ghoshal and Bartlett (1988), normative integration be-tween headquarters and the subsidiary facilitates creation, adoption and diffusion of innovations by the subsidiaries. Moreover, high levels of intra-subsidiary communication stimulated the creation by the intra-subsidiary. In addition, both adaptation and diffusion is stimulated through high levels

(18)

of headquarters-subsidiary and inter-subsidiary communication. However, both local autonomy and the extent of local slack that is given to the re-sources do not seem to have any influence (Ghoshal & Bartlett, 1988). How-ever, local autonomy and the extent of local slack may not have any influence on innovation, the role of the subsidiary ’embeddedness’ seems to do have influence on the strategic power (Yamin & Forsgren, 2006).

Strategic embeddedness has two related consequences with important im-plications for the MNC in retaining power to exclusive control over strategy (Yamin & Forsgren, 2006). Firstly, the subsidiary embeddedness generates knowledge-based resources that come from the relations with external part-ners (Andersson et al. 2002; Forsgren, Pedersen & Foss, 1999). This will increase the power of the subsidiary because such resources are typically outside the control of the MNC headquarters (Mudambi & Navarra, 2004). This context-specificity is also found to be very important in sustainabil-ity related knowledge, and it may increase the costs when the MNC’s are trying to have one global strategy regarding sustainability (Pinkse, Kuss, Hoffmann, 2009). This implies that local variation is needed within the strategy and it allows for a certain amount of individual absorptive capacity to be developed within the subsidiary (Pinkse, Kuss, Hoffmann, 2009). The second consequence of embeddedness is the invisibility of the subsidiary’s network (Yamin & Forsgren, 2006). Therefore, the upper level of the MNC tends to have at most a vague understanding of the subsidiary’s network and this may limit their ability to control the subsidiary (Yamin & Fors-gren, 2006).

To conclude, it seems that the strategic domain and the operational do-main are not as divided as before. Although the headquarters still have the most power, this power is in fact distributed and not residing only at headquarters (Yamin & Forsgren, 2006). Also Robock, Simmons and Zwick, (1977) argue that the MNC is the typical case of an organization facing het-erogeneous task environments and that its national subsidiaries are often embedded in very different environmental conditions. This external embed-dedness of a subsidiary can positively contribute towards the development of products and processes in the MNC (Andersson et al., 2002).

Research has shown that local embeddedness of MNC units (Anderson and Forsgren, 2000) is very influential in MNC performance. This can be explained by the unique knowledge flows that the local embedded subsidiary has access to. Namely, cross-border or transnational knowledge creation is seen as the mechanism for integrating the differentiated competencies of the MNC (Subramanian and Venkatraman, 2001) through for example global innovation strategy (Pearce, 1999).

(19)

sec-tion, it can be stated that embeddedness creates a unique subsidiary envi-ronment in which it is difficult for headquarters to maintain full control. On one hand, the embeddedness could give important local knowledge and ex-pertise to increase the innovativeness of subsidiaries, but on the other hand, it also decreases the strategic power of headquarters. Seen in this light, it could be interesting to examine the influence of embeddedness and how this is handled by the headquarters on the progression of H&S initiatives. This leads to the following proposition:

Relevance to process of the H&S initiative: ’The higher the local subsidiary Embeddedness the less H&S initiatives will progress throughout the organization’

2.6.3 Knowledge Based View (KBV) - Subsidiary Knowledge Transfer

Researchers in organization theory (Levit and March, 1988), economics (Nel-son and Winter, 1982) and strategic management (Prahalad and Hamel, 1994; Schendel, 1996) have identified organizational learning as one of the most important subjects for scholarly inquiry. A key topic in this broad area of research is the flow of organizational knowledge (Gupta and Govin-darajan, 2000). Knowledge perhaps has the greatest ability to become a source of sustainable differentiation and hence competitive advantage (Dier-icks and Cool, 1989; Lippman and Rumelt, 1982; Drucker, 1969). Moreover, in the last few years the KBV has developed as a credible and legitimate theoretical lens to aid our understandings of how firms survive and compete in today’s economy (Eisenhardt & Santos; Choi & Lee, 1997).

As stated, knowledge management is increasingly viewed as a source of competitive advantage. Managing knowledge well is seen as a crucial part of a firm’s success (Mudambi, 2002). Or, as Riesenberger (1998) has stated, ’as MNCs aim to replicate their success across borders they will need to focus not just on ’what’ they know, but ’how’ they gain that knowledge and diffuse it throughout the organization’. In order to manage knowledge well, a thorough insight in this topic is therefore crucial. Davenport and Prusak (1998) see knowledge as a fluid mix of framed experience, values, contextual information, and expert insight that provides a framework for evaluating and incorporating new experiences and information. It originates and is applied in minds of knower’s. In organizations, it often becomes embedded not only in documents or repositories but also in organizational routines, processes, practices and norms (Davenport & Prusak, 1998).

When looking at knowledge, Mudambi (2002) distinguishes the micro perspective. In this perspective, there is a distinction between intra- and

(20)

inter-firm knowledge flows. The micro perspective can be best described as a source-target characterization of knowledge transfer (Mudambi, 2002). Each knowledge flow occurs between a source and a target along a certain channel (Gupta & Govindarajan, 2000). There are four flows identified: (1) From subsidiary to parent, (2) From location to subsidiary, (3) From subsidiary to location (4) From parent to subsidiary. All of the knowledge flows mentioned above are subjected to transmission losses (Shannon and Weaver, 1998). This means that, in general, the flow of knowledge received by the target will be smaller than the flow transmitted by the source. In this research, mainly the subsidiary-parent and parent-subsidiary flow is examined.

Different ways of transmitting knowledge can be chosen (for example, e-mail, face to face, phone), this choice is influenced by the nature of the knowledge that is being transferred (Mudambi, 2002). Moreover, knowledge can be tacit or explicit, Szulanski (1996) points out that tacit, context-specific, and ambiguous knowledge is most difficult to transfer. Explicit knowledge is more easily articulated, coded and transferred (Nonaka, 1994). Tacit knowledge is far more difficult to articulate and is derived from indi-vidual experiences (Matusik and Hill, 1998). Both types are valuable to an organization, however tacit knowledge is more difficult to capture since it resides within an individual. The tacitness of knowledge is one of the most widely recognized barriers to transfer and replicate knowledge (Lippman & Rumelt, 1982; Polanyi, 1996; Zander & Kogut, 1995).

All these different flows and types of knowledge have influence on the manner in which a MNC initiates certain activities and even how successful they are. Many scholars have researched the contributions of knowledge-based theory (Lippman & Rumelt, 1982; Grant, 1996; Mudambi, 2002) in relation to the performance of the MNC. Both the influencing factors in knowledge sharing as the consequences of knowledge sharing are frequent researched topics (Persson, 2006; Yamin and Otto, 2004). A growing number of scholars argue that the increasing capabilities in subsidiaries, challenge the ’from above’ perspective (Yamin and Forsgren, 2006). Birkinshaw and Hood (1998) confirm this by saying that in most organizations there is an internal competition for power, that illustrates the changing roles of subsidiaries, partly influenced by the subsidiaries own choices and initiatives.

Operational structure is an important determinant of knowledge sharing by subsidiaries. MNC headquarters can influence knowledge flow transfer within the corporation by creating the criteria used to evaluate subsidiary performance. Furthermore, outward knowledge transfer had a positive re-lation with the use of corporate socialisation practices. In other words, the development of, for example international training programmes, stimulating visits across MNC units, and international task forces, can increase the

(21)

like-lihood for knowledge sharing (Bjorkman, et al., 2004). Williams (2009) also stresses the importance using knowledge based logic in analyzing initiatives. It is argued that barriers to effective knowledge transfer will prevent very important information about opportunities and problems, and solu-tions and ideas, to be distributed throughout the MNC. Seen in this light it becomes clear that the role of subsidiary learning and shared strategic goals are important factors in promoting global initiatives. Moreover, variables like inter-unit networking, subsidiary learning and shared strategic goals can all be influenced by managers to activate a knowledge network and create an environment where global opportunities can be discovered and progressed (Williams, 2009). This leads to the following proposition:

Relevance to process of the H&S initiative: ’The greater the Knowledge Transfer within headquarters and subsidiary the more the H&S initiatives will progress throughout the organization’

2.7 Theoretical Framework

The theoretical framework for this research can be summarized as follows:

Figure 1: Conceptual Model

2.8 Summary

Management control theory, resource dependency theory and knowledge based view have contributed in describing the dynamics inside the multi-national and core business activities like; technology, research and develop-ment, innovation and so forth. The three theories and the variables that derived from them are not the only variables that emerged from these three theories, however this research will focus on these theories since these seem very relevant and applicable.

From the management control theory results show that increased auton-omy that is given to the subsidiary gives them more room for innovation and building their own network to increase their efficiency. However, it is

(22)

argued that this increased power also diminishes the power of headquarters and therefore the influence they have on the subsidiaries activities. From the resource dependency theory, the second variable embeddedness derived.

From the literature it can be concluded that embeddedness can influence the relationship between headquarter and subsidiary. The more the sub-sidiary is embedded in his local environment the more power it seems to have over its own activities. The transfer of knowledge is the third vari-able that is examined in this research. Throughout the years, scholars have found strong support of knowledge being an important source of sustain-able differentiation and hence competitive advantage (Diericks and Cool, 1989; Lippman and Rumelt, 1982; Drucker, 1969). Furthermore, in the ever-changing environment knowledge transfer can be identified as the glue that holds the MNC together.

All these variables have proven their influence on the core activities of the business, in this paper a less core business activity is under investiga-tion namely, H&S. However, H&S becomes more and more strategic and therefore, it will be interesting to see if the proposed relationships are also applicable for non-financial activities within the MNC. In figure 1 the con-ceptual model of this research is presented.

(23)

3 Methodology

3.1 Research Design

There is only a very limited amount of existing academic research on the influence of embeddedness, autonomy and knowledge sharing on the progress of international H&S initiatives. Hence, first exploratory work is required before a quantitative study can be performed. Exploratory work makes it possible to detect the influence of the research variables on the progress of international H&S initiatives. Therefore, a qualitative research design is most appropriate (Robson, 2002). The proposed model (figure 1) was tested with an explorative qualitative research design. The qualitative data was collected by multiple methods, namely interviews and the collection of company documents.

This research has a partly inductive and partly deductive approach. The inductive approach of using explorative and qualitative methods creates the opportunity for theory development. The deductive approach, of using exist-ing organizational variables creates the opportunity to test the applicability of organizational theories in non-financial areas within the MNC.

3.2 Validity and Reliability

Construct validity, external validity, internal validity and reliability are paramount when talking about qualitative research design. Construct va-lidity is defined by Yin (2003) as ”establishing correct measures for the concepts being studied”. By using the general interview guide approach, construct validity will be enhanced. This is because the interviews were conducted in a focused manner, which makes the data collection systematic for each respondent (Patton, 1990).

External validity is defined by Johnson and Christensen (2000) as ”the extent to which the results of a study can be generalized to and across populations, settings, and times”. The external validity of this study will be moderate because several companies will be included and per company, one interview is conducted. For example, survey research relies on statistical generalization whereas case studies rely on analytical generalization; the researcher strives to generalize a particular set of results to some broader theory (Yin, 2003). Despite the fact that a case study holds a moderate external validity Yin (2003) argues that this limitation only pertains to statistical generalizability.

Internal validity is defined by Yin (2003) as ”establishing a causal re-lationship, whereby certain conditions are shown to lead to other conditions, as distinguished from spurious relationships”. This definition is applicable

(24)

for causal studies but not for case study research because these are not con-cerned with making causal claims (Yin, 2003). For case studies, the internal validity may be extended to the broader problem of making inferences. A case study involves an inference every time an event cannot directly be ob-served. In this study, the internal validity was increased by taping and transcribing, which made it possible to reanalyze the interviews.

Analyzing and interpretation of data unavoidably bring with them a cer-tain amount of bias of the interviewer. To eliminate this bias, a systematic content analysis was conducted in which all the proposed themes were noted down as well as the emerging themes that were not expected before hand. Furthermore, besides the researcher, peer researchers checked the interview results in order to check the consistency of the executive researcher. More-over, besides the interview guide it will be possible for the interviewees to add extra information. This will allow insights to the complexity of the case to be greater, thereby increasing the internal validity (Patton, 1990).

Reliability is defined by Yin (2003) as ”demonstrating that the opera-tions of a study can be repeated, with the same results”. To overcome the reliability problem, a logbook was made to monitor every step of this re-search. This was done to make sure that another researcher could replicate this research and hopefully arrives at the same results.

3.3 Level of analysis

The unit of analysis in this research is the international H&S initiative. This unit of analysis is chosen because it creates a good starting point to investigate what the influencing factors are in relation to these initiatives. Moreover, within one interview several initiatives could be identified, capi-talizing upon the opportunity of creating a relatively large sample size.

The unit of analysis is important because it gives the right focus during the interviews and the research in general. During this research, an attempt is made to identify as many different types of initiatives and to examine how these specific initiatives have been influenced. By integrating the level of analysis in the interview questions, it is endeavoured to generate different in-ternational H&S initiatives and eventually compare these different initiative examples.

3.4 Data collection

As indicated before, a qualitative research was conducted within seven MNC’s. To make it possible to generate general ideas about H&S initiatives across countries and industries, the research was conducted with seven MNC’s with

(25)

different orientations like, chemicals, food and beverages, airline industry, banking and electronics.

In this research most of the MNC’s were Dutch or partly Dutch only one MNC did not have Dutch nationality but was German. The international scope and size were taken into account when choosing MNC’s for this re-search, these elements made it possible to generate international initiatives. Furthermore, because of the unique size and scope, Health & Safety is a ma-jor subject within the MNC which enables the generation of rich information about this subject.

3.4.1 Interview Participants

Data collection was done via interviews that were conducted between Novem-ber 2008 and February 2009. Altogether, one interview per MNC was con-ducted, which made 7 in-depth interviews in total. When selecting the interviewees, it was ensured that they had a good overview of what kind of activities regarding H&S were initiated both at corporate as at subsidiary level.

Within the cooperating MNC’s data was collected from several H&S rep-resentatives located at corporate, safety specialists, safety managers, safety directors, and occupational physician. Only corporate representatives were chosen because they are expected to have the best overview and informa-tion about the progress of H&S initiatives on both corporate and subsidiary level. This approach enables a better perspective but a certain amount of bias due to this selection has to be taken into account.

Because of the fact that these ’experts’ often have high positions, they are extremely occupied, under high time pressure and often protected by secretary, which makes it difficult to approach them. To overcome this issue, the researchers own network was used to be able to have access to these key informants. In table 1, a summary is given of the interviewees.

(26)

Table 1: Overview of interviewees

3.4.2 Operationalisation

The interviews were conducted following a general interview guide approach (Patton, 2002). An interview guide was prepared in order to make sure that the same information was obtained from a number of people from head-quarters by covering the same material (Patton, 2002). This means that the interviews were semi structured, with predetermined questions. However, the order could be changed based upon the perception of the interviewer of what seemed most appropriate (Robson, 2002).

In addition, questions could be modified and questions that did not seem appropriate with a particular interviewee could be omitted or extra questions included (Robson, 2002). By adding open-ended questions, it allows to go into more depth, encourage co-operation and rapport and can produce unanticipated or unexpected answers (Robson, 2002). Some examples of formulated questions are presented in table 2.

(27)

During the interviews, it is very important that the interviewees can talk freely and openly. Because the interviewer has a major influence on this, a couple of important interview techniques were considered during the in-terview. Namely, more emphasize was put on listening, the questions were put in a straightforward, clear and non- threatening way to overcome that people feel confused or defensive. Furthermore, cues which could influence the interviewees responds were eliminated, and at last an active and inter-ested posture by the interviewer was adopted by varying voice and facial expression (Robson, 2002).

During the interview, additional space was given to give room for emerg-ing variables not discussed in the conceptual model. By sometimes leavemerg-ing the structure of the interview, a lot of extra interesting information was received due to this free and open approach. The interviews lasted between 45 minutes to 60 minutes, were digitally recorded and were located at the interviewees’ offices or took place via teleconference. After the interviews, the records were transcribed immediately and were send to the participants for comments and feedback. In table 1, the participants and their charac-teristics were presented.

Uncertainties were examined by triangulating the documentation on these initiatives, annual reports, sustainability reports and the MNC’s web-sites. When uncertainties could not be clarified the Interviewees gave the opportunity to be contacted by phone or email.

3.5 Research Variables

3.5.1 Dependent variable - Progress International H&S initiative In this research, the dependant variable is the progress of an international H&S initiative throughout the multinational. As such in this research, the dependant variable will be measured by how well the initiative has pro-gressed throughout the multinational, as assessed by the degree of internal-ization by both corporate and the various subsidiaries. In table 3 an overview is given to illustrate the way the dependent variable was measured.

(28)

Table 3: Evaluation of Dependent Variable

Given the explorative nature of this qualitative research and the em-ployed methodology of semi-structured interviews, emphasis is laid on scop-ing the relationships as identified in the theoretical framework, rather than on using hard criteria and quantitative measurements. In analyzing the interviews, a project was classified as an initiative when this project was. The purpose of identifying different initiatives is to distinguish initiatives that have progressed well in either way between corporate and subsidiaries (SP = Smooth Progress) and initiatives that had more difficulties progress-ing throughout the multinational (DP= Difficult Progress). The cases that were less clear and were more in between SP and DP were classified as being mixed (MP = Mixed Progress) These qualifications are based upon the qualitative information that was extracted during the semi-structured interviews on these initiatives.

3.5.2 Independent Variables - autonomy, embeddedness, knowl-edge transfer

The objective of this research is to examine how autonomy, embeddedness and internal knowledge transfer influence the progress of the implementa-tion of internaimplementa-tional H&S initiatives. These variables have all proved their influence on initiatives that involve core business practices. In this research, it will be investigated how these factors influence initiatives that have no direct effect on revenue, like H&S practices. In the following section, the three independent variables will be discussed.

3.5.3 Research Variables - autonomy

Between MNC’s there are differences regarding the governance model they use. In some companies, decision making authority is highly centralized at the headquarters and the national subsidiaries are not involved in any de-velopment of new programs or even in adjusting any products or processes

(29)

developed by the parent company (Ghoshal & Bartlett, 1988). In this re-search autonomy is measured by the degree of decentralisation. In other words, when a subsidiary is given a lot of power, decision rights, freedom to initiate own ideas, then the subsidiary will be identified as being highly autonomous.

When, on the other hand, the organization is more centralized and the subsidiary is given no power, no decision rights, and no freedom to initiate own ideas then the subsidiary will be addressed as being low in autonomy. However, there will also be cases where it is less straight forward and au-tonomy is in the middle of those two extremes. Therefore, in this research there is also a category, which will be used for the cases where the subsidiary is given significant autonomy but is still very dependent on headquarters (mixed autonomy).

3.5.4 Research Variables - embeddedness

Subsidiaries are a part of an organisation that is dispersed internationally. The geographical distance is an important factor. Subsidiaries are not only separated by distance though, because they operate within distinct legal, political, cultural and economic areas (Yamin, 2002). This may enable the subsidiaries to absorb new knowledge from the unique environment in which they are in. This strategic embeddedness may be very influential on strate-gic power (Yamin & Forsgren, 2006). In this research, embeddedness will be measured by the amount of cultural, political, legal and economic ties the subsidiary has with the community it is in. Themes that will be used to identify embedded subsidiaries are; the amount of local culture within the subsidiary, the power of local laws, the amount of ties with external institu-tions (e.g. universities). In this research the subsidiary can be identified as being low embedded, mixed embedded and highly embedded.

3.5.5 Research Variables - knowledge transfer

It is stated that organizational replication is very difficult because the repli-cation will always be partial. Organizational routines are not transferred easily which means that knowledge and more general routines to the sub-sidiary by the parent lose some of their effectiveness and value because they are very context dependant (Madhok, 1997, Yamin). In this research knowl-edge, transfer is measured by the amount and the frequency of information sharing is taking place inside the multinational. Themes like cross divisional knowledge sharing, team meetings, sharing best practices, training, and me-dia use (internet, email etc.) are examined to determine whether within the initiative knowledge sharing can be identified as high, mixed or low.

(30)

3.6 Data Analysis

A thematic content analysis was used to identify the themes autonomy, em-beddedness and knowledge transfer and during the interview it was allowed for emerging theories to arise. Those who were most prominently put for-ward during the interviews were also identified and retrieved. Phrases and words that were typical for the themes were retrieved and mapped in a sys-tematic overview. This made it possible to analyse the different themes that were detected and how they influenced the results of each initiative (Miles & Huberman, 1984).

(31)

4 Results

4.1 Health & Safety initiatives analysis

For each multinational, several initiatives were identified. In total 16 exam-ples of H&S initiatives were discussed during interviews with key informants. These 16 initiatives are mapped and presented in table 4.

Table 4: Overview of participating MNC’s including A,B,C coding In line with the framework as developed in Section 2, the results are based on how well the initiative has progressed throughout the multinational, as assessed by the degree of internalization by both corporate and the various subsidiaries. The results have been rated to have Smooth Progress, Mixed Progress, or Difficult Progress. The initiatives that are identified as being mixed or difficult, can still be a success in one of the multinational’s sub-sidiaries or in corporate, but for these initiatives the organization has been unsuccessful in progressing it throughout the entire multinational. The ini-tiatives that are identified as being well progressed throughout the MNC may have been generated either at corporate or in one of the subsidiaries, but have been successfully spread and internalized throughout the multina-tional.

(32)

4.2 Results initiatives within the multinational

4.2.1 Initiative 1: Avian Flu Program - DSM

The first initiative is the Avian Flu Program, which was coordinated by corporate and had a global orientation. The project started in 2005 because of; 1) the turmoil that was caused by the media and 2) the request of the executive board and the signals that were coming from the ARBO Service. These factors played an important role for corporate to think about what should be done about the Avian flu. This quickly led to the formulation of a concrete goal by the executive board to start a worldwide approach. Now, the ’alarm phase level’ of the World Health Organisation (WHO) is stable. Therefore, the company decided that the project is going into a kind of hibernation state. The company indicates that concerning the Avian Flu Program they achieved everything they wanted and that the program has become a structural part of the other scenarios within all the subsidiaries. The results were checked by a central team and the eventual result of the initiative within the subsidiaries was ascertained by means of finished plans and staying up to date for the plans in the coming years. Furthermore, the CAO has included the requirements and this means that with every audit of a subsidiary the state of the avian flu scenario within the whole calamity scenario planning system is being evaluated.

The initiative was a corporate one and only emphasized the basic princi-ples. The choice concerning the continuation of divisions during an outbreak, are left with the subsidiaries, corporate does not interfere with those things. The only interference from corporate level is about determining how serious an outbreak is. Therefore, it is clear that the autonomy of subsidiary was high in relation to the Avian flu initiative.

The local legislation and governmental factors seem to play an important role within the Avian flu program. Because of the fact that every country has its own rules, the company has decided that in principle the local leg-islations and activities of local government are leading. This shows that the subsidiaries are highly embedded in their local environment, which can greatly influence the initiative.

From the interview, it became clear that there was a strong communica-tion regarding the Avian flu initiative. Multidisciplinary teams guided the project and this was identified to be an important factor in the success of this project. Furthermore, regular communication took place between the local project group and the corporate project group. The latter distributed templates in order to get one standardized approach throughout all the sub-sidiaries. The local project group would then modify the templates towards a more localized realization. This illustrates that DSM clearly identifies

(33)

internal knowledge transfer as important and that all knowledge in the or-ganization was used and distributed regularly throughout the oror-ganization in order to succeed.

Results - Smooth progress Autonomy - High

Embeddedness - High Knowledge transfer - High

4.2.2 Initiative 2: Vitality Program - DSM

The second initiative is the Vitality Program initiative, which started in 2006 as a result of the identification of demographic developments and in-ternal client questions. The company recognized that in certain parts of the world it was necessary to also have health promotion besides primal care and preventive activities to control the risks in the job. Because of these findings, the company decided to look for an instrument that can measure the health of the employees. This instrument is called the ’Health Risk As-sessment’, which is an electronic instrument that the company called the ’Vitality Checkpoint’. With this instrument, they started to do measure-ments in a couple of subsidiaries. After the measuremeasure-ments were done, a plan was made for interventions. One important problem, which was stressed by the interviewee, was that they found that there were few health stimulating interventions for sale.

Many companies are busy with measuring and some with personal coach-ing but it is not really structured yet. The initiative was implemented in the Netherlands and in only one other location (in the US). They are talk-ing about maktalk-ing the program also available for ex-pats. The company indicates that, the fact that this initiative is only being implemented in so few countries has to do with the type of impact of this program. In many countries where the subsidiaries are located there are different norms re-garding the privacy considerations and interference of the employers. In the Netherlands, the acceptance for interference of the employer is much higher on average than in the rest of the world. Furthermore, the interviewee says that they have a rather complicated organisation and that the business ori-ented and subsidiary coordinated organizations are more difficult to roll out programs when the direct effect of it is unknown.

Furthermore, the interviewee believes that the most important thing is to have local commitment and that local support is a prerequisite for a successful initiative. The fact that local commitment is not a natural state

(34)

within the subsidiaries is due to their autonomous nature, indicating high autonomy.

The emphasis was on measuring and finding counter actions for the dis-covered health issues but the set-up of internal knowledge transfer was lim-ited and hardly mentioned during the interview. Overall, it can therefore be stated that in comparison with the other initiative within this company, the knowledge transfer within this initiative was low.

Results - Difficult progress Autonomy - High

Embeddedness - High Knowledge transfer - Low

International Health and Safety within the multinational corporation : the effect of subsidiary autonomy, embeddedness, and internal knowledge transfer on the progression of Health and Safety initiatives within the MNC

CONFIDENTIAL

MASTER THESIS

International Health & Safety within the

Multinational Corporation

”The effect of subsidiary autonomy, embeddedness,

and internal knowledge transfer on the progression

of Health & Safety initiatives within the MNC”

Contents

List of Figures

List of Tables

1

Introduction

2

Theoretical Background

3

Methodology

4

Results