The impact of national governments and multinational corporations on indigenous community conflict and the moderating effect of the firm’s home country

(1)

MASTER THESIS

The impact of national governments and multinational corporations on indigenous community conflict and the moderating effect of the firm’s home country.

Karin van Wassenaar - 10978224

MSc. in Business Administration – International Management Track Faculty of Economic and Business, University of Amsterdam

First supervisor: dhr. dr. I. (Ilir) Haxhi Second reader: dhr. drs. E. (Erik) Dirksen

(2)

Statement of originality

This document is written by Karin van Wassenaar 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 Economic and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

Table of Figures and Tables

Figure 1. Conceptual model ... 28

Table 1. Variable Description... 30

Table 2. Formula regression Analysis ... 35

Table 3. Summary regression analysis ... 35

Table 4. Means, Standard Deviations, Correlations ... 36

Table 5. Regression check of High level violence ... 39

Table 6. Results with High Level Violence (HLV) as dependent variable ... 40

Table 7. Home country moderate effect ... 44

Table 8. Home country moderate effect ... 44

Abbreviations

CSR Corporate Social Responsibility DJSI Dow Jones Sustainability Index HDI Human Development Index ISDS Investor-State Dispute Settlement MNC Multinational corporation

CSP Corporate Social Performance CFP Corporate Financial Performance

(5)

Abstract

This research indicates the importance of Multi-National Corporation (MNC) policy and the potential impact corporations can have in reducing indigenous community conflict they are involved in. Despite the increasing awareness towards the protection of rights of indigenous communities, there is limited quantitative research that studies the influence of MNCs and the government on indigenous community conflict.

This study indicates, based on an analysis of 115 conflict cases in the Asia-Pacific region, that MNCs with a policy indicating no consent, or an ambiguous statement towards the indigenous community, were more likely to be involved in high-level violence. Companies with its headquarters located in a Western country, were less involved in high-level violence compared to companies from non-Western countries. This is in line with previous literature arguing that countries located in Western countries are more likely to take the environment and society into account. The influence of the home country can be explained by the effect of the country’s institutional framework, as it influences the firm’s behavior by signaling which choices are accepted and supported.

This research extends from the existing literature regarding conflict between indigenous communities and MNCs operating on their land. The inclusion of multiple factors contributes to a better understanding of the complex relationship between MNCs and indigenous communities and the impact the national government has on this relationship. Moreover, the results of this study can be of potential value for management, as it can contribute towards reviewing their existing policies for indigenous communities. The significant result of the moderating influence of a MNC’s home country can be of value for national governments. Governments can protect indigenous communities by setting up legal frameworks to ensure that companies are responsible and decrease conflict with indigenous communities.

KEYWORDS: Asia-Pacific region, Corporate Social Responsibility (CSR), governments, indigenous community conflict, Multinational Corporations (MNCs), MNC policies

(6)

1. Introduction

What constitutes an indigenous community has been subject of ongoing debate among both scholars and practitioners (Murphy and Arenas, 2010). Examples of indigenous communities are the Maori in New Zealand, the Aboriginal in Australia, and the Inuit in Canada and the United States (Cunningham, 2003).

Many countries experience conflict between the indigenous communities and companies operating in their land. These can range from peaceful negotiations to high-level violence (Calvano, 2008). An example of conflict can be found between the Aboriginals in Australia and mining corporations. Mining operations in Australia operate in remote and rural areas, traditionally used and inhabited by diverse aboriginal

communities. Crawley and Sinclair (2003) highlight the enormous gap regarding welfare and employment between Australia’s white population and the Aboriginals and indicate that some are experiencing living in conditions similar to third-world countries.

Garvey and Newell (2005) address that the protection of indigenous communities can be ensured by the government through policies and legal frameworks. This can protect and promote the relative rights of communities and ensure that companies take responsibility for their impact on society.

However, trade agreements have given MNCs the ability to influence regulations that used to be a privilege to national governments (Newell, 2002; VPROTegenlicht, 2015). These trade agreements are

developed to protect corporate rights and not community rights (Banerjee, 2008). The increased power MNCs have nowadays, has created a complex situation. On the one hand, the indigenous community is promised protection, however, this protection cannot be realized on account that corporations can influence the policies in ways that benefits themselves.

The existing problem may seem as if indigenous communities are in a hopeless situation. However, as the intangible assets of firms gain increased importance, not taking the communities into account can damage the reputation of the firm (Porter and Kramer, 2006). This is one of the reasons why companies take more responsibility towards communities in which they operate. In order to avoid, solve, and minimize conflict, it is suggested that awareness of community perspectives need to be raised and that more negotiation power

(7)

should be given to indigenous communities (Bond, 2014; Calvano, 2008; Waddock, 2008). The assumption is that sharing authority and decision-making will enhance the process of resource management (McCay and Jentoft, 1998; Calvano, 2008). The attention to indigenous rights and development of effective processes of governance, autonomy and responsibility, is fundamental towards sustainable futures (Howitt, 2012). This means that MNC policies towards the affected communities are important to minimize conflict.

Recent discussions on how to solve conflict highlights the positive impact MNCs can have, and the mediating influence of governments by providing protection. Therefore, the first research question in this current study is formulated as followed:

RQ1: What is the impact of MNCs characteristics and the national government on the length and degree of indigenous community conflict MNCs are involved in?

This research will study the impact governments and MNCs have on existing conflict between MNCs and indigenous communities. MNCs will be evaluated based on three characteristics. First, the policy of MNCs will be examined, and it will be evaluated whether MNCs take indigenous communities into account by mentioning these communities in their policies. The second characteristic selected is the inclusion of MNCs in the Dow Jones Sustainability Indices (RobecoSAM and S&P Dow Jones Indices (1999). The Dow Jones

Sustainability Indices (DJSI) serves as a benchmark for investors and evaluates in terms of economic,

environmental and social criteria. This variable serves as a compliment to MNC policy as the information MNCs provide may not always be current. The evaluation by a third-party based on the actual performances of MNCs may allow a better understanding of the behavior of the firm. The third characteristic is the overall ownership structure of the firm. This characteristic has been selected as previous research indicates that the ownership structure influences how a firm engages with the community (Hart and Moore, 1990; Oh, Chang and Martynov, 2011).

Influence from the government will be evaluated by the degree of recognition the indigenous community receives. Relations between the national government and the community is important to

(8)

determine the degree of protection the community receives (Garvey and Newell, 2005). It is expected that a community receiving full recognition and autonomy will be involved in a less violent type of conflict than communities receiving limited or no recognition.

The dependent variable in this research is community conflict. This variable is measured by considering the total duration of conflict, and the type of conflict. Type of violence is the degree of violence of conflict and ranges from low to high-level violence (Calvano, 2008). The length of violence has been included, as this may be taken into consideration by companies to calculate the necessity to intervene (Getz and Oetzel, 2009). Including two factors of conflict will allow a better understanding of the actual kind of conflict.

The moderator in this research is the country where the MNC headquarters is situated. This is due to the assumption that the institutional framework influences MNCs strategies and policies as formulated by the firm (Peng, 2002). Several studies highlight how the institutional framework influences MNC policies and strategies (Peng, 2002; Zhao, Tan and Spark, 2014). These institutional frameworks affect the behavior of corporations by signaling which choices are acceptable and supported (Peng, 2002).

As previous literature indicates the important influence of the institutional framework on the

formulation of policies (Peng, 2002; Zhao et al., 2014), it is expected that the country of the MNC headquarters influences how strategies and policies are formulated. Companies with its headquarters located in Western countries are expected to have policies that address social issues, and are more likely to take the livelihood of the affected stakeholders, such as indigenous communities, into consideration (Crane et al., 2013). The expected moderating effects lead to the second research question in this study:

RQ2: How does a MNC’s home country influence the relationship between MNC characteristics and the length and degree of indigenous community conflicts that MNCs are involved in?

To answer these formulated research questions, this study has conducted a quantitative analysis including 115 conflict cases located in the Asia-Pacific region. The countries included in this research are: Australia, Cambodia, China, Indonesia, Lao People’s Democratic Republic, New Zealand, Papua New Guinea,

(9)

Philippines, Thailand, and Vietnam. Evaluating the Asia-Pacific region with a wide variety of countries can extend the current discussion to more countries. The inclusion of a wide variety of MNCs from different countries, allows a better understanding of the influence from institutional frameworks. Previous research is often of qualitative nature, and provides insight into one county, and one single industry. This research will include several countries, and evaluates the indigenous community in a broader context, by including multiple factors that may influence conflict. Moreover, this research will provide practical contributions to MNCs and governments. Increased understanding about the factors that influence conflict, will help (re)formulate policies and strategies for both MNCs and governments in order to reduce conflict. In addition, as this study confirms the importance of the behavior of MNCs towards indigenous communities, it can stimulate MNCs as well as governments to increase awareness of communities and increase active management.

This paper is structured as followed: Chapter two will give an overview of previous literature regarding the power of MNCs, and how it can influence communities and policies. The factors that influences how MNCs behave will also be discussed. Chapter three will present the theoretical framework and the proposed

hypothesis, and is followed by the data and method in chapter four. Chapter five presents the results, and is followed by the sixth chapter which discusses the findings, contributions, limitations, and the direction for possible future research. Chapter seven is the final chapter of this research and presents the conclusion.

(10)

2. Literature review

This chapter will give an overview of previous research. This chapter starts with a brief explanation of who indigenous people are, and where they are situated. In section two, the influence of the government is discussed, and explains how communities can be protected by government policies and regulations. Section three discusses the increasing power MNCs have gained in recent global trends. The power they possess can seriously affect society and influence government regulations. The fourth section of this chapter discusses how the formulation of policies are being influenced by the institutional framework, and explains the differences that can be found among MNCs from different countries. The fifth section discusses how the ownership structure of the firm influences company policies and strategies. The last two sections of this chapter, section 6 and 7, discusses how CSR policies and MNC behavior towards communities can influence conflict.

2.1 Indigenous people

United Nations’ Economic and Social Council’s report defines ‘indigenous communities’ as “peoples and nations are those which, having a historical continuity with pre-invasion and pre-colonial societies that

developed on their territories, consider themselves distinct from other sectors of the societies now prevailing on those territories, or part of them. They form at present non-dominant sectors of society and are determined to preserve, develop and transmit to future generations their ancestral territories, and their ethnic identity, as the basis of their continued existence as peoples, in accordance with their own cultural patterns, social institutions and legal system.” (Martinez Cobo, 1983). Around the world there are approximately 390 million indigenous people spread over 70 countries (UN, 2009). There is no accepted definition of the term ‘indigenous’, but the United Nations has developed a number of characteristics to describe indigenous people.

The characteristics are defined as: 1) self-identification as indigenous peoples at the individual level and accepted by the community as their member, 2) historical continuity with pre-colonial and/or pre-settler societies, 3) strong link to territories and surrounding natural resources, 4) distinct social, economic or political systems, 5) distinct language, culture and beliefs, 6) from non-dominant groups of society, 7) resolve to

(11)

2009). To recognize, promote, and protect the rights and the freedoms of the indigenous people around the world, the United Nations has developed a declaration on the right of indigenous people (UN, 2009). This declaration sets a number of standards for all the member states of the United Nations for the survival, dignity, well-being and rights of the indigenous people of the world. Moreover, this declaration encourages

cooperative relations between States and indigenous people in a harmonious way.

2.2 Government and community relation

The declaration on the right of indigenous people, is to protect the indigenous people of our world and ensure their individual and collective rights (UN, 2008). This declaration sets the standard for states, to which they can formulate their policy towards their indigenous community. This means that the government plays a vital role in the protection of indigenous people. The government can ensure the protection of indigenous communities by setting up policies and legal frameworks to protect and promote the relative rights and ensure that companies take their responsibilities for their impact to the society and environment (Garvey and Newell, 2005). Therefore, with the level of state protection for community interests, the national government is an external stakeholder that mediates MNC and community conflict.

Moreover, as Garvey and Newell (2005) also explain, the relation between a state and community is important to determine the level of protection a community receives. One of the examples they mention is the poor relationship between a community and a government in West Papua, Indonesia. The government

protected the rights of corporations over the local communities. Freeport, a mining enterprise was given a contract which gave the company powers over local communities and resources. They could resettle indigenous inhabitants without compensation for the loss of hunting, fishing grounds, water supplies, or damage to livelihoods. When the community made attempts to seek justice, the government intervened to prevent this from happening. This example points out that the government can also play a negative role in mediating conflict between MNCs and communities.

Poor relations between the government and indigenous communities is not only found in developing countries. Threbeck (2007) addresses this issue by discussing the disadvantages indigenous communities in

(12)

Australia face. The indigenous people in Australia are politically, socially, and economically disadvantaged compared to the rest of the population (Altman, 2004). Crawley and Sinclair (2003) have also highlighted the enormous gap regarding welfare and employment between Australia’s white population and the Aboriginals. And indicate that some are experiencing living conditions similar to third world countries.

2.3 MNC power

The government is responsible for setting up policies and legal frameworks. This assumes that the government has the most influential power in a country. However, recent global trends have increased the power of global corporations, extending corporate influence to activities which used to be a privilege of the state (Newell, 2002). Newell (2002) explains that MNCs gain more power, which are declared in international agreements such as the TRIPS Accord and the GATS. TRIPS has been developed by the WTO to protect corporate rights, not community rights (Banerjee, 2008). These agreements decrease the powers of national and local authority, and increase the power of firms. Indigenous communities all over the world notice that this has adverse impact on their livelihood by restricting community access to natural resources (Banerjee, 2008). An example is the Investor-State Dispute Settlement (ISDS) as part of the NAFTA agreement, which allows companies to sue local authorities setting environmental standards that they claim discriminate against foreign companies (Newell, 2002).

A recent documentary of the Dutch television program VPROtegenlicht (2015) mentions that ISDS enables companies to file billion dollar claims against states when they find that their revenues are in danger due to the formulation of new regulations and laws. Countries that have signed trade agreements including ISDS, are at risk that foreign investors will make claims when regulations are expected to reduce profits.

One of the examples mentioned in the documentary is the French company: Veolia, suing Egypt for raising the minimum wage. Another example is Pacific Rim, now OceanaGold, taking El Salvador for not granting an environmental permit for a goldmine.

These examples point out that companies can influence governments in the decision-making of the political framework. In addition to the power to sue national government, Parkinson (2005) explains that

(13)

companies possess political power, as bargaining power of business can influence government policy. In his article he explains that democratic government is dependent on the overall performance of the economy. Companies operating in a country, contribute to the overall performance of the economy. Governments will have to maintain conditions so that these companies can stay profitable. Governments must keep in mind how their policies will be perceived by business interests. When companies disagree to the policy, they can decide to relocate to countries with better business environments. This will negatively affect the economy, as it leads to capital outflows.

2.4 MNC characteristics

Recent global trends have increased the power of corporations. This has changed how governments operate, as foreign investors are given much more attention than the interests of others in the country. Corporation have the political power to access government decision-making, while the community does not receive political power (Garvey and Newell, 2007).

There are several factors that influence the way MNCs formulate their strategies and policies. In the following section an overview will be given about internal, and external factors that influences MNC behavior. First, the institutional framework of the home country will be discussed, which is followed by the impact of ownership structure.

2.4.1 Institutional influence

MNCs from different countries differ in MNC policies and strategies, which can be explained by the influence of the institutional framework. Institutions are the formal rules and informal constraints that structure political, social and economic relationships - also referred to as the rules of the game in society (North, 1991). Aguines and Glavas (2013) concluded, based on a CSR literature review that firms engage in CSR due to institutional pressures. These pressures particularly come from stakeholders, but also from institutional forces such as regulations, standards, and certifications. Peng (2002) argues that strategies of firms from different countries do not only differ because of influences from the industry and the firm resources, but also because of the institutional differences. The institutional framework influences the behavior of the firm, by

(14)

signaling which choices are acceptable and supported. This means that any strategic choice a firm makes, is affected by the formal and informal constraints of a given institutional framework (Peng, 2002).

Similar arguments have been made by Zhao et al (2014), as they explain that the formulation and enforcement of the formal rules of the game govern the relationship between MNCs and stakeholders. The way Corporate Social Responsibility (CSR) issues are addressed, also depends on the home countries’

institutional framework. Crane et al. (2013) indicates the differences that can be found between CSR policies from different countries. Crane et al. (2013) points out that US companies tend to have social issues at the core of their CSR policy, which is contrasts to many other countries.

2.4.2 Ownership structure

The previous section addressed how the institutional environment, which consists of the informal and formal rules of the game, influences how a firm behaves and formulates its policies. Besides the institutional environment, the ownership structure of the firm also has great influences on how a firm formulates its policies. A research conducted by Panwar, Paul, Nybakk, Hansen & Thompson (2014) found that there is a lower legitimacy for CSR actions of publically traded companies than for family-owned companies.

Hart and Moore (1990) have studied how changes in ownership affect the incentives of employees as well as those of owner-managers. They have found that ownership has an influence on organization decision-making, motivation, and power. Oh et al. (2011) has found that ownership structure has significant effect on the firm’s CSR engagement. They conclude that ownership structure affects strategic decision-making of the firm, by showing that investors have different attitude towards CSR engagement. In this study, owners were separated in three groups: institutional (banks, pension funds, insurance companies, and securities firms), managerial (top management team and outside directors), and foreign ownership. Differences were found between financial institutions with governmental and non-governmental influences.

The study shows that financial institutions that were influenced by the government, showed a higher motivation to encourage CSR. In addition, the study also concludes that foreign investors play a significant role

(15)

in management practices. In a sample of large Korean firms, the study found that foreign investors played a significant role in the adoption of Western-style management practices such as CSR engagement.

Finkelstein (1992) supports the influence of managerial ownership, and addresses the key role of managers’ power in strategic decision-making. He explains that a firm’s CEO is usually the most powerful member of the firm, but this is not always the case, as managers with large shareholding may be more powerful than a CEO. Therefore, it is concluded that CEOs share power with other senior executives in many firms to influence the organizational direction.

2.4.3 CSR policy

The institutional framework and the ownership structure of the firm are of great influence in regards to MNCs strategies and policies. One of the recent trends is the integration of Corporate Social Responsibility (CSR) policies. CSR is a concept whereby companies integrate social and environmental concerns in their business operations and in their interaction with their stakeholders on a voluntary basis and in a context specific way (Aguinis and Glavas, 2013).

According to Porter and Kramer (2006) there are four reasons for having a CSR policy. The first reason is the moral obligation, as companies have the duty to be good citizens and to do the right things. The second reason is sustainability, meaning that companies should manage the firm in such a way that its activities meet the needs of the present, without compromising the ability of future generations to meet their needs. The third reason is the license to operate, as companies sometimes need to fulfill the requirements of

governments, communities and other stakeholders. The fourth reason is reputation, as a bad reputation can cause harm to the business.

The fourth reason has also been discussed by Parkinson (1995). Parkinson (1995) argues that as long as the company’s objective is maximum profits, a concern with the welfare of third parties is necessarily to protect profitability in the long term. Not taking third parties into consideration can seriously damage the reputation of the company. Because more than half of corporation’s assets today are intangible, companies should undertake initiatives that directly benefit society (Waddock, 2008).

(16)

Many studies have been conducted to find a relationship between corporate social performance (CSP) and corporate financial performance (CFP). Margolis, Elfenbein, & Walsh (2009) has found a small, but positive relationship between CSP and CFP. However, they also found that having a bad reputation has a more

pronounced effect on financial performance. A possible explanation for this outcome is that negative events are weighed more heavily than positive events, and that actors are more likely to be blamed for negative consequences than they are to be rewarded for comparable positive outcomes (Margolis et al., 2009).

However, taking the environment and society into account does not guarantee a better CFP. Another research that has been conducted by Barnett and Salomon (2012) conclude that companies with low CSP have higher CFP than firms with moderate CSP. Firms with high CSP have the highest CFP. This reflects a U-shaped relationship between CSP and CFP. An explanation for this result is from the degree of stakeholder influence. Firms with low stakeholder influence capacity (SIC) are less able to transform socially responsible activities into tangible returns because stakeholders are less likely to view their social actions as credible. However, a firm with high stakeholder influence capacity may be rewarded, as it may ease the firm’s ability to contract with key stakeholders, and decrease transaction costs (Barnett and Salomon (2012).

CSR is a strategic initiative, and just like any other strategy, there is always a risk of failure. It can lead to benefits for the firm, but it is not guaranteed. The contrasting results of these studies does not answer whether a firm should conduct a CSR policy. However, regardless if CSR is worthwhile, society expects management to be green (Marcus and Fremeth, 2009).

(17)

2.5 Behavior towards community conflict

“Sacrificing the viability of indigenous cultures for industrial resource extraction to maintain consumer society is unethical. It is ethical to engage with indigenous people in a manner consistent with their wishes, their cultures and means for material survival” (Lerzman and Vredenburg, 2005). This points out the moral reason why it is important to engage in dialogue with indigenous cultures. In addition, a number of reasons for companies to join dialogue or engagement have been pointed out by Waddock (2008). Waddock (2008) explains that companies engage in dialogue to build trust with their stakeholders, which is important because of the intangible assets of the firm. Another reason is to forestall legislation by acting proactively. And the final reason is to have some say over what kinds of expectations stakeholders can reasonably put forward by engaging more directly.

However, as a firm has many stakeholders, some are given more priority than others. With the help of the stakeholder identification theory of Mitchel, Angle and Wood (1997) an explanation can be found on who and what truly counts. Mitchel et al. (1997) explain that stakeholders are identified based on three relationship attributes: power, legitimacy, and urgency. A stakeholder possessing all three attributes are considered as a definitive stakeholder. An example of a definitive stakeholder are the stockholders of the firm. The indigenous community can be defined as a discretionary stakeholder, which is a stakeholder holding the legitimacy attribute. Discretionary stakeholders do not cause pressure on managers to engage in an active relationship.

Not having an active relationship with the indigenous community can be part of the existing conflict. Cavano (2008) suggest that the awareness of community perspectives need to be raised and that more

negotiation power should be given in order to minimize conflict. Historically these communities have possessed little or no power relative to other stakeholders. With no power, the community is not able to influence MNC decision-making in their favor. Calvano (2008) has created a framework that points out the three causes of conflict between Multinational Corporations (MNC) and the local community. The three causes of conflict are: stakeholder power inequality, stakeholder perception gaps and cultural context.

(18)

Bond (2014) distinguishes two types of methods to deal with conflict. The first is conflict management, and the second is conflict transformation. Conflict management often seeks to suppress, contain or otherwise avoid the consequences of conflict. Conflict transformation, on the other hand is more about the willingness for parties to honestly enter and engage a relationship to find an outcome that both is beneficial to the long-term relationship, as well as to find an agreeable solution to the conflict.

In order to resolve the conflict between MNCs and local communities, awareness of community perspectives need to be publicly raised and should be given more negotiation power. One of the factors that is important in determining the effectiveness of community-based strategy is the corporation’s approach to citizen participation (Garvey and Newell, 2007).

A way to create more negotiation power for indigenous communities is through the establishment of co-management in which collaborative institutional arrangements are made among diverse stakeholders for managing or using a natural resource (Castro and Nielsen, 2011). The fundamental assumption is that sharing authority and decision-making will enhance the process of resource management, making it more responsive to a range of needs (McCay and Jentoft, 1998).

McCay and Jentoft (1998) have expressed the potential of co-management as a way of re-embedding management responsibilities within the local communities. However, care must be given on how power is distributed as Castro and Nielsen (2001) discuss that differences in negotiation power in co-management regimes to deal with natural resource conflict can set into new conflict or cause old ones to escalate. This is due to the problem that co-management can lead to strengthening of the state’s control over resource policy, management, and allocation leading to opposing outcomes. Instead of contributing to local empowerment, co-management may further marginalize indigenous communities.

Watson (2013) points out that even with the increased research and policy-making collaborations with indigenous peoples, including official co-management regimes, continuing dissatisfaction still exists. Better communication and mutual learning is needed to create more adaptive co-management regimes. Howitt (2012) highlights the importance of equality and involvement of everyone, and argues that attention to

(19)

indigenous rights and development of good relationships and good processes of governance, autonomy and responsibility is fundamental to sustainable futures.

Lin and Chang (2011) address in their article the potential impacts of local participation on the natural environment based on a case study regarding the Meqmegi, an indigenous community in Taiwan. They conclude that it is critical to be concerned about environment and natural resource conditions to ensure that local participation leads to positive outcomes and can lead to a sustainable future. Since the late 1990s, a number of actions began with the aim of natural resource conservation. The participation of the local community for natural resource conservation and sustainable development is a good strategy and has a positive impact of resource management.

Lin and Chang (2011) explain that local conservation issues are more relevant in areas with higher concentration of indigenous populations, because indigenous people have a higher dependency on natural resources and living traditionally in natural sites. Due to the daily interaction with the environment and local knowledge, MNCs are starting to recognize indigenous communities as stakeholders of resource management, rather than a threat to natural resources (Carter, 2010).

Despite the recognition of importance to engage with indigenous people, and the implementation of co-management regimes, conflict with indigenous communities still exist. Crawley and Sinclair (2003) suggest that only initiatives that strive towards power-sharing with indigenous groups will contribute to more ethical practices such as in the mining industry in Australia. Lerzman and Vredenburg (2005) have extended this research and found similar arguments for forestry in Canada.

Bond (2014) explains that mining companies, and other large-scale industries, can create positive social value and increase conditions of peace in their operating environments. These companies have the potential to support the creation conditions of peace as part of their sustainable development agenda as well as their community relation strategies when it comes to company-community conflict (Bond, 2014).

(20)

3. Theoretical Framework

Trade agreements, such as the TRIPS, have increased the power of corporations (Banerjee, 2008; Newell, 2002; VPROTegenlicht, 2015). This gives MNCs the ability to influence policies that used to be a privilege for national governments. There is a lot of resentment from indigenous communities all over the world towards the TRIPS agreement. Indigenous communities are against these trade agreements because they are developed to ensure protection of corporate rights and not community rights (Banerjee, 2008).

The Declaration on the Right of Indigenous People (UN, 2009) is supposed to protect the rights and freedoms of indigenous people around the world. A set of standards to encourage cooperative relations between states and indigenous people. The protection of indigenous communities can be ensured by setting up policies and legal frameworks and ensure that companies take their responsibility for their impact on the environment and society (Garvey and Newell, 2005).

The increased power of corporations, and the standards as set up by the UN creates a peculiar

situation. On the one hand, the indigenous community is promised protection, however, this protection cannot be realized, on account that corporations can influence the policies in a way that benefits itself. This damages the relations between indigenous communities and the government.

Poor relations between the government and indigenous communities can be found all over the world. The indigenous communities in Australia have been highlighted multiple times, and are described as being disadvantaged and face enormous gaps regarding welfare and employment (Altman, 2004; Crawley and Sinclair, 2003; Threbeck, 2007). Governments are forced to meet the needs of corporations because the economy of the country depends on it (Parkinson, 2005).

This complex situation indigenous communities face may seem unsolvable. However, the recent trend to keep the environment and society in mind has increased the number of MNCs having a CSR policy. There are multiple reasons to operate more responsible, and one is to maintain a good reputation, as a bad reputation can cause harm to a business (Porter and Kramer, 2006). The assets of MNC nowadays consist for more than half of intangible assets, which makes it beneficial to undertake social initiatives (Waddock, 2008).

(21)

Previous research suggests that MNC can have significant positive influence on the conflict they are in with indigenous communities. The literature suggests that in order to minimize conflict, MNCs should give indigenous communities more negotiation power (Calvano, 2008). Involving indigenous communities and engage with them is beneficial for both the community and the corporation. Not only is it important for the intangible assets of the firm (Waddock, 2008), but engaging in a dialogue can diminish conflict (Calvano, 2008) and lead to a sustainable future (Lin and Chang, 2011).

The influence of both the national government and MNC on the indigenous community conflict the MNC is involved in leads to the first research question:

RQ1: What is the impact of MNCs characteristics and the national government on the length and degree of indigenous community conflict MNCs are involved in?

In addition, difference in policies and strategies that can be seen among MNCs can be explained by the influence of the institutional framework (Peng, 2002; Zhao et al., 2014). The institutional framework influences the behavior of the firm, by signaling which choices are acceptable and supported (Peng, 2002). The way companies formulate their CSR policies is also dependent on the institutional framework. Crane et al. (2013) has indicated the difference between CSR policies from different countries, and found that US companies tend to have social issues at the core of their CSR policy.

As previous literature indicates the importance of the institutional framework, it is expected that the country in which MNCs headquarters are located influences how strategies and policies are formulated.

Companies with its headquarters located in Western countries are expected to have policies that address social issues, and are more likely to take the livelihood of the affected stakeholders such as indigenous communities into consideration. The expected moderating effect lead to the second research question in this study:

RQ2: How does MNC’s home country influence the relationship between MNC characteristics and the length and degree of indigenous community conflict MNCs are involved in?

(22)

3.1 Research Gap

The situations indigenous communities face all over the world is a topic that has been addressed many times. Articles have been written regarding the bad relationships they have with corporations, and the lack of protection they receive from governments. A recurring solution for conflict with indigenous communities is the concept of co-management. McCay and Jentoft (1998) have explained the positive effect of sharing authority and decision making on resource management. However, other scholars address that co-management can cause new conflicts or cause old ones to escalate, due to the differences in negotiation power in co-management regimes (Castro and Nielsen, 2011; Howitt, 2012; Watson, 2013).

Trying to grasp the complex situation, and the search for conflict resolution can only be done when the situation is understood in full. However, what is seen in most studies is that there is only one focus point. Lin and Chang (2011) argue the importance of local participation based on a case study regarding the indigenous community in Taiwan. Other examples are the suggestions of power-sharing with indigenous communities based on the mining industry and Australia (Crawley and Sinclair, 2003). Lerzman and Vredenburg (2005) found similar arguments for forestry in Canada.

Previous research are often of qualitative nature, and the arguments are mostly based on one specific country and one specific industry. These studies certainly provide important and useful insights, but do not reflect the overall problem. Therefore, this study will have a different approach and combine several conflict factors into one study. Moreover, this study will be, unlike other studies, a quantitative study by including 115 existing conflict cases from ten countries located in the Asia-Pacific region.

The MNC will be evaluated based on three characteristics, and it will be analyzed how these

characteristics influence the conflict they have with indigenous communities. In addition, the influence of the relation between indigenous communities and the government will also be measured. The concept conflict will be measured based on total length of conflict, and the degree of violence of conflict. Moreover, the home country of MNCs will also be taken into account, as research suggests the influence of the institutional framework of formulated strategies and policies (Crane et al., 2013; Peng, 2002; Zhao et al., 2014).

(23)

All in all, this study will include several perspectives regarding MNC and indigenous community conflict. Not only will the influence of MNCs be taken into account, but also the moderating effect of a MNC’s home country. In addition, the indigenous community’s relation with the government will also be taken into account, as previous studies suggest that the government can have a mediating role in protecting indigenous

communities through policies and legal frameworks (Garvel and Newell, 2005).

3.2 Hypotheses

This section will present the suggested hypotheses in this study. As explained in the previous chapter, MNCs as well as governments are expected to influence community conflict. MNCs is a very broad concept, and to understand the behavior of MNCs three factors will be examined.

The first factor is the stated MNC policy towards the indigenous community, as it is expected that MNC policy will influence community conflict. Previous research suggests that by recognizing indigenous

communities and giving them more autonomy, will lead to less violent conflicts (Calvano, 2008; Waddock, 2008). The second factor, as a complement to MNCs policy is whether or not the MNC is included in the Dow Jones Sustainability Indices (DJSI). The critical evaluation in order to be included in this index, allows a better understanding how MNCs values the environment and the society. The third factor is the overall ownership structure of the firm, since previous research suggests that this influences the attitude of the firm towards its stakeholders, and how it engages with the community (Hart and Moore, 1990; Oh et al., 2011). Community conflict will be measured in two ways. First, the type of violence will be measured to indicate how serious the conflict is. This can vary from low level violence such as peaceful negotiations to high level violence including deaths (Calvano, 2008). To gain better insight on the extent of the conflict, the second measurement will be the length of conflict as measured in months. This variable has been included as companies may calculate the necessity to intervene based on the duration of conflict. The firm may be less motivated to intervene if it appears that the conflict is likely to be short-lived or non-recurring (Getz and Oetzel, 2009). The opposite is likely to happen during a long term conflict. A long term conflict is associated with weak states, and it is unlikely that the government will intervene effectively. This increases the likelihood for MNCs to intervene

(24)

(Hironaka, 2005).These variables together will allow for a more holistic view on the kind of conflict the MNC is involved in.

3.2.1 MNC policy

Previous studies highlight the importance of indigenous community recognition and the equal distribution of power among stakeholders (Calvano, 2008; Crawley and Sinclair, 2003). The underlying assumption is that recognition and involvement of the indigenous community will lead to more peaceful negotiations. These reflections lead to the assumption that the degree of MNC community recognition, by which the community receives negotiation power will significantly influences community conflict.

There are two different approaches in dealing with conflict. The first is conflict management, and the second is conflict transformation (Bond, 2014). The strategy in how a MNCs handles its conflict influences the relationship with the indigenous community. Bond (2014) explains that conflict management is more effective, as it is more about the attitude and the willingness for parties to honestly enter and engage a relationship to find an outcome that both is beneficial to the long-term relationship, as well as to find an agreeable solution to the conflict.

The recent trend of CSR policies of MNC show that MNCs take more responsibility for their business activities. However, due to the lack of knowledge on what the actual motivations of CSR are, and the difficulties caused by a lack of a universal definition for CSR, complicates the ability to compare the effect of a company’s overall CSR policy on the type of conflicts. To avoid the negative effect of the validity of evaluating the entire CSR policy, the specific MNC relation with indigenous community will be evaluated. It is expected that conflict is more peaceful and less violent when indigenous communities are given reasonable/equal negotiation power as state-agencies, and other stakeholders.

To evaluate the relationship between MNCs and communities, the MNC’s policy will be evaluated. It is expected that MNC policies that have mentioned the indigenous community specifically will have a significantly positive effect on the type of conflict. This proposition is based on the underlying assumption that recognition

(25)

and involvement of the indigenous community will lead to more peaceful negotiations (Calvano, 2008). Thus, the following two hypotheses are proposed:

H1: MNC that have specifically mentioned the indigenous community in their policy, are expected to be involved in a more peaceful type of conflict

H2: MNC that have specifically mentioned the indigenous community in their policy, are expected to be involved in conflict with a shorter duration

3.2.2 Dow Jones Sustainability Indices

MNC policies will be evaluated based on information that can be found on company websites. As this may cause a bias in favor of MNCs, the Dow Jones Sustainability Indices (DJSI) will complement MNC policy. The DJSI consists of listed companies that have been chosen based on the Corporate Sustainability Assessment by RobecoSAM. This assessment is conducted each year by an independent third-party and provides insights into business practices of the world’s largest companies (RobecoSAM, 2015). The DJSI is globally recognized, and analyzes 2,900 listed companies each year, based on environmental, social, and governance factors.

The expectation from society to conduct business in a responsible and sustainable way (Marcus and Fremeth, 2009), has increased the number of companies with CSR policies. Having a CSR policy does not necessarily guarantee financial benefits. However, not operating in a responsible way may cause harm to the company’s reputation and financial performance (Margolis et al, 2009; Waddock, 2008).

Formulated MNC policies can be found on many company websites and describe how they conduct their business operations. However, as convincing as these policies may sound, an actual assessment is often missing. Therefore, an assessment conducted by a third-party allows an evaluation of the actual corporate social responsibility performance. This assessment allows more insight and a better understanding regarding MNC’s behavior towards the environment and society. Thus the following two hypotheses are proposed:

H3: MNCs that are included in the DJSI are expected to be involved in a more peaceful type of conflict

(26)

3.2.3 MNC ownership structure

The ownership structure of a firm influences MNC behavior (Hart and Moore, 1990). The ownership structure can be divided into three categories: state-owned, publically traded, and privately owned. Research has found that there is a lower legitimacy for CSR action of publically traded companies than for family-owned companies (Panway et al., 2014). In addition, financial institutions with governmental influences show a higher motivation to CSR, than institutions that are not influenced by the government (Oh et al., 2011).

The differences that can be seen among companies with different ownership structure can be explained based on the motivation of the stakeholders. A publically traded company has the social responsibility to increase its profits (Friedman, 1970). The increased costs that may arrive for conducting business more responsibly may lead to the decision not to conduct with CSR. There is also no guarantee that CSR can lead to profit (Porter and Linde, 1995).

A higher motivation for CSR in privately owned companies can be explained by personal values and individual concern for the issues (Aguinis and Glavas, 2013). A higher motivation for CSR due to governmental influences can be explained by the policies and standards that exist in the specific country. The increased awareness for indigenous people is assured through the Declaration on the Right of Indigenous People, as formulated by the UN. Member states can formulate their policies based on this declaration and ensure the protection of the indigenous communities situated in their countries. Through this, governments can maintain the protection of indigenous communities and ensure that companies take their responsibilities for their impact on society and environment (Garvey and Newell, 2005). The expected influence of MNC ownership structure leads to the following two hypotheses:

H5: The ownership structure of MNCs will influence the degree of violence of conflict H6: The ownership structure of MNCs will influence the duration of conflict

(27)

3.2.4 Indigenous community autonomy

Government policies and legal frameworks protect and promote the relative rights and responsibilities of companies and communities. The relationship between state and community is important to determine the level of protection a community receives. Garvey and Newell (2005) explain that a government can play a positive or a negative role.

Previous studies highlight the importance of indigenous community recognition. The underlying assumption is that recognition and involvement of indigenous communities will lead to more peaceful negotiations (Calvano, 2008). Therefore, the extent to which a community receives autonomy from the national government is an important factor that reflects recognition and protection. It is expected that conflict will be more peaceful and less violent when indigenous communities are given reasonable/equal negotiation power as state-agencies, and other stakeholders.

However, the opposite is expected to happen when the relationship between government and

indigenous community is weak. The government in West Papua, Indonesia, protected the rights of corporations over local communities, and gave mining enterprises a contract giving power over local communities and resources (Garvey and Newell, 2005). Local communities were relocated without compensation for the damages they suffered from. When the communities made attempts to seek justice, the government intervened to prevent this from happening. This example shows that the national government can play a positive or negative role in mediating conflict between MNCs and communities. Therefore, it is proposed that:

H7: A high extent of received community autonomy from the national government will positively influence type of community conflict.

3.2.5 Moderating effect: Home country

Differences among strategies and policies as formulated by MNCs can be explained by the influences of their institutional framework (Peng, 2002; Zhao et al., 2014). This means that the formulated CSR policy is also influenced by the institutional framework. Institutional forces, such as regulations, standards, and

certifications, as well as pressures from stakeholders, influence firms to engage in CSR (Aguines and Glavas, 2013). These pressures are different among countries, which can explain the differences between companies.

(28)

Crane et al. (2013) addresses these differences and points out an important difference that can be found in the way companies address social issues in different countries. An example is how in the United States many social issues, such as education, healthcare, or community investment have traditionally been at the core of CSR. This is different compared to many other countries; as social issues have traditionally been seen as a task for the government. However, the growing concern regarding sustainability issues, especially in developed countries, is expected to influence MNCs policy in addressing social issues in regards to their stakeholders, including the indigenous communities.

As previous literature indicates the importance of the institutional framework, it is expected that the country in which MNCs headquarters is located will influence how strategies and policies are formulated. This expectation is based on the differences that can be found among institutional frameworks of countries. The growing concern for sustainability, which is especially seen in developed countries, has increased the number of companies engaging in CSR. As mentioned previously, an increased engagement with the community will lead to a more peaceful type of conflict (Calvano, 2008).

CSR engagement is more often seen in developed western countries. Therefore, it is expected that MNCs based in western countries will have a more specific policy towards addressing social issues in regards to indigenous communities (Crane et al. 2013). This is expected to positively influence the conflict experienced between MNCs and indigenous communities. Therefore, the final hypothesis of this study proposes that:

H8: MNCs with their headquarters based in a Western-country will positively influence MNCs policy towards indigenous community conflict.

Figure 1 shows the conceptual model used in this research. This model visualizes all the proposed hypotheses. The model reflects how MNCs and the national government is expected to directly affect indigenous community conflict. MNCs will be measured based on three characteristics that represent three independent variables. These are MNCs policies, the inclusion in the DJSI, and MNCs ownership structure. The indigenous community conflict will be measured by the types of violence of conflict, and the duration of the

(29)

conflict as expressed in total number of months. This reflects a total of 8 hypotheses, as can be seen in the conceptual model.

Figure 1. Conceptual model

MNC characteristics

IV1: MNC policy

IV2: DJSI

IV3: Ownership structure

Government relation/ Community autonomy DV1: Type of violence of conflict DV2: Length of conflict H1, H2, H3, H4, H5, H6

Independent variables Moderating variables Dependent variable

Control variables:  HDI score  Corruption control MNC home country H8 H7 Community conflict

(30)

4. Data and Method

This research will be an exploratory quantitative research and will contribute to the existing dialogue about conflict with indigenous people and conflict resolution. This research will investigate whether

community autonomy, as recognized by the national government and MNCs characteristic, affects conflict with indigenous communities. Countries that are inhabited by indigenous communities will be the unit of analysis in this research.

In order to run a quantitative analysis, a data set has been built together with seven other students. Each student has collected data according to a prior agreed set of variables. Everyone has collected 50 cases in total, resulting to a total dataset of 400 cases. These 400 cases represent indigenous community conflicts from around the world. This research will focus on indigenous communities located in the Asia-Pacific region. Out of the entire data set, a total of 115 cases have been selected. These cases consist of the following countries; Australia, Cambodia, China, Indonesia, Lao People’s Democratic Republic, Malaysia, Papua New Guinea, and the Philippines.

By conducting a country-level analysis in which will be examined how MNCs’ home country influences the relations between MNC characteristics and community conflict, more insight will be gained in how the institutional framework influences the type of conflict. The extent to which the indigenous communities are recognized differs per country. It is expected that indigenous communities receiving a high degree of autonomy will be respected by MNCs, and thus encounter a lower degree of conflict than communities receiving a low degree of autonomy. Moreover, MNCs with its headquarters located in Western countries are expected to take more responsibility in regard to their impact on the environment and society, leading to the involvement in a less violent type of conflict than MNCs from non-Western countries.

Indigenous community conflict, as indicated by third-party organizations is the initial point of reference throughout the data collection process. Once the conflict is indicated, further research has been conducted by investigating all the parties involved in conflict. This data comes from sources such as provided information by

(31)

non-governmental organizations (NGO), the United Nations, and company websites. Table 1 provides a brief overview of the variables used in this research together with the respective descriptions.

Table 1. Variable Description

Variable Description

Policy towards community

1. No mention of community consent or no mention of specific policy 2. Ambiguous/vague statement towards indigenous community 3. Strong statement towards indigenous community

Dow Jones

Sustainability Index

0. MNC not included in the DJSI 1. MNC is included in the DJSI Ownership structure 1. State-owned

2. Publically traded 3. Privately owned Length of conflict 1. 0 - 72 months

2. 72 - 132 months 3. 133 – onwards Types violence of conflict 1. Peaceful negotiations 2. Peaceful protests 3. Court action

4. Low-level violence (from either side) and intimidation tactics

5. High-level violence (from either side), including physical damage but no kidnapping or deaths

6. High-level violence (from either side), including kidnappings but no deaths 7. High-level violence (from either side), including deaths. In comment column, briefly describe evolution over time

High level violence 0 = Type of violence 1-4 1 = Type of violence 5-7

Home country Country in which the HQ is located; 0 = Non-Western Country

1 = Western Country Community

autonomy

Degree of community autonomy as received from the government 1. No autonomy at all

2. Cultural recognition

3. Limited recognition of rules

4. Partial recognition of rules or right of consultation

5. Almost full recognition of community rules by government with autonomy or right of consent

(32)

4.1 Dependent variables

There are two dependent variables used in this research. These two variables both measure indigenous community conflict. First, the degree of violence of conflict was indicated, ranging from low to high level violence (Calvano, 2008). Each conflict has been evaluated on a 7-point continuum, in which 1 reflects peaceful negotiations and 7 reflects high-level violence (from either side), including deaths.

In addition to the types of violence, the total length of conflict has also taken into account. This variable has been included as companies may calculate the necessity to intervene based on the length of the conflict. The firm may be less motivated to intervene if it appears that the conflict is likely to be short-lived or non-recurring (Getz and Oetzel, 2009). The opposite is likely to happen during a long term conflict. A long term conflict is associated with weak states, and it is unlikely that the government will intervene effectively. This increases the likelihood for MNCs to intervene (Hironaka, 2005).

The duration of the conflict in the Asia-Pacific area have been going on for many years, which made it difficult to indicate whether the duration of the conflict was short or long-term. Therefore, a new variable has been created in which three intervals were made to categorize the length of conflict. The first interval was 0 – 72 months, the second interval was 72 -132 months, and the third interval was 132 months and onward.

4.2 Independent variables

There are two categories of independent variables used in this research. The first category consists of the characteristics of MNCs and includes three independent variables. These three variables are

complementary to each other to provide a better understanding of MNCs. The characteristics consist of MNCs policy, the inclusion in the DJSI, and the ownership structure of the firm. The second category is the

government relations towards the community and consist of one independent variable.

All three MNC characteristics are expected to influence the behavior of MNCs. The ownership structure influences decision-making, motivation, and power (Hart and Moore, 1990). This can be explained by the different motivations of the investors of MNCs (Oh et al., 2011). The company policy will be evaluated based on

(33)

information that is provided on company websites. This criteria will measure the degree of engagement with indigenous communities, as increased community awareness and engagement is beneficial to find solutions to conflict (Calvano, 2008; Bond, 2014). The DJSI is included as an addition to the company policy towards indigenous communities. The DJSI is based on an evaluation in terms of economic, environmental, and social criteria (RobecoSAM and S&P Dow Jones Indices (1999).

The government and community relation is the fourth variable that is expected to influence indigenous community conflict. Setting up policies and legal frameworks by governments can ensure indigenous

community protection (Garvey and Newell, 2005).

4.2.1 MNC characteristics IV 1: MNC Policy

MNCs policy indicates the firm’s behavior towards the indigenous community. The policy has been coded from 1 to 3, in which 1 reflects no mention of community consent or no mention of specific policy, 2 reflects a vague statement towards indigenous community, and 3 a strong statement towards indigenous community. Caution must be given to codifying the policy, as the policy has been indicated based on the information found on a company’s website, and can lead to bias due to personal judgment and interpretation.

IV2: Dow Jones Sustainability Index

The indication whether the company is included in the DJSI is a complementary variable on top of the indicated MNC policy. This sustainability index provides additional information in regards to the responsibility the firm takes towards people and the environment. All firms that are included in this index have been indicated with a 1 and all that are not included with a 0.

IV3: Ownership Structure

The ownership structure of the firm has been indicated based on three types of structure. Each firm has been coded as either state-owned, publically traded, or privately owned.

(34)

4.2.2 Government Relation

IV4: Community Autonomy

The community autonomy of the indigenous community is the degree of community autonomy as received from the government. This variable has been coded from 1 to 5 in which 1 reflects no autonomy at all, and 5 almost full almost full recognition of community rules by government with autonomy or right of consent.

4.3 Moderating variable

The moderating variable in this research is the home country of the MNC. The institutional framework of the home country is expected to influence formulated polices and strategies of the firm (Crane et al., 2013; Peng, 2002; Zhao et al., 2014). Differences in institutional framework, especially between Western- and non-Western countries (Crane et al., 2013) may cause a moderating effect and influence the formulated policies towards indigenous communities.

This variable has been coded by indicating in which country MNCs headquarters is located. This was followed by the creation of a new variable. All non-Western countries were indicated by a 0, and all Western countries were indicated by a 1. An overview of all the countries and the indication can be found in the appendix.

4.4 Control variables

The development degree and the control of corruption is expected to influence the direct effect. The government can play a vital role in the protection of indigenous people, through policies and legal frameworks (Garvey and Newell, 2005). However, these policies may not always be respected. An example given by Garvey and Newell (2005) about a situation in West Papua, Indonesia point out that governments can favor

corporations, and prevent communities from seeking legal justice. Another example can be found in the Philippines. To defend mining companies against community protest, most of the country has now been militarized, which has led to a lot of deaths (Vidal, 2015).

(35)

The Corruption Perceptions Index (Transparency International, 2015) measures the perceived levels of public sector corruption worldwide. This index shows that corruption takes place in many countries, but is especially high in developing countries. Therefore, the variable Human Development Index (HDI) and control of corruption have been included as control variables.

4.5 Method

Hierarchical multiple regression analysis has been conducted to estimate the relationships among variables. The only correlations found in this research involved the categorical dependent variable of ‘type of conflict’. In order to run a regression analysis, it is necessary to transform this variable into a numerical

variable. Therefore, a new variable has been created in which the distinction has been made between low-level violence and high-level violence. Types of violence of 1 to 4 have been selected as low-level violence, and type 5-7 have been selected has high-level violence.

In order to run a regression analysis, the categorical variables have been translated into dummy variables. In this analysis the variables ownership structure, MNC policy, and community autonomy have been transformed into dummy variables.

For each variable for which the creation of dummy variables was necessary, the dummy variable with the smallest value has been left out as a reference. The formula used in this regression analysis is as followed:

Yi= α + β1X1i + β2X2i + ... + βKXK, Policy1 + ε, ε ~ iin(0,σ)

Table 2 shows an overview of all the formulas used for the regression analysis. Eight hierarchical multiple regression analysis have been performed to investigate the independent variables to predict high level violence. The control variables ‘Control of Corruption’ and the ‘Human Development Index (HDI) have been included as a second step in the multiple regression analysis. This control can be recognized by the even numbers in the overview below. All the uneven numbers represent the formulas of the regression analysis without this control.

(36)

Table 2. Formula regression Analysis Model Formula

1 High level violence = α + β1DPublic + β2DState-owned + ε, ε ~ iin(0,σ)

2 High level violence = α + β1DPublic + β2DState-owned + β3 HDI + β4 Corruption control ε, ε ~ iin(0,σ)

3 High level violence = α + β1Policy1 + β2Policy2 + β3DJSI + ε, ε ~ iin(0,σ)

4 High level violence = α + β1Policy1 + β2Policy2 + β3DJSI + β4 HDI + β5 Corruption control ε, ε ~ iin(0,σ)

5 High level violence = α + β1DAutonomy1 + β2Dautonomy3 + β3DAutomomy5 + ε, ε ~ iin(0,σ) 6 High level violence = α + β1DAutonomy1 + β2Dautonomy3 + β3DAutomomy5 + β5 HDI + β6

Corruption control ε, ε ~ iin(0,σ)

7 High level violence = α + β1DPublic + β2DState-owned + β3Policy1 + β4Policy2 + β5DJSI + β6DAutonomy1 + β7Dautonomy3 + β8DAutonomy4 + β9DAutomomy5 + ε, ε ~ iin(0,σ) 8 High level violence = α + β1DPublic + β2DState-owned + β3Policy1 + β4Policy2 + β5DJSI +

β6DAutonomy1 + β7Dautonomy3 + β8DAutonomy4 + β9DAutomomy5 + β10 HDI + β11 Corruption control ε, ε ~ iin(0,σ)

To test the moderation of the home country, the PROCESS macro, as created by Professor A.F. Hayes (2013), has been used, which allowed a moderated regression analysis. To format the variable home country into a numerical variable, a distinction has been made between Western-and non-Western countries by creating a dummy variable. All Western countries were numbered with 1, and all non-Western countries with 0. Table 3 provides a summary of the hierarchical regression analysis and the moderated regression analysis.

Table 3. Summary regression analysis

Model Control variable Independent variable Dependent

variable Moderator Corruption Control HDI Ownership Structure DJSI Community autonomy MNC Policy High level violence Home-country

Hie

ra

rch

ical

re

gressi

on analysis

1. x x 2. x x x x 3. x x x 4. x x x x x 5. x x x x 6. x x x x x x 7. x x x x x 8. x x x x x x x Moderated regression analysis 1. x x x