What are the effects of national governance, MNE activity and indigenous community isolation on MNE-indigenous community conflicts in Latin America?

What are the effects of national governance, MNE activity and indigenous

community isolation on MNE-indigenous community conflicts in Latin

America?

Topic: Indigenous, MNEs and National Governance Supervisor: Dr. Ilir Haxhi

Statement of Originality

This document is written by Pieter Hogewoning, who takes full responsibility for its content.

I hereby declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in the creation of it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

Abstract

This thesis investigates conflicts between multinational enterprises (MNEs) and indigenous communities in Latin America. Using data obtained from 277 conflicts, this study offers an explanation of different factors and patterns in a conflict’s degree of violence and will explore aspects of the three main actors of conflicts: indigenous communities, MNEs and national governments. Firstly, this study argues that countries’ with higher political stability and lack of violence will have conflicts with a lower degree of violence. Secondly, it states that MNEs with extractive activities have conflicts with a higher degree of violence. Thirdly, it claims that higher geographic isolation of indigenous communities leads to a higher degree of violence. Moreover, we argue that regulatory quality has a negative moderator effect on the first two relationships and that democratic nature of communities has a positive moderator effect on the third relationship.

The only statistically significant relationship we found was the moderator effect of regulatory quality on the relationship between political stability and lack of violence and the conflicts’ degree of violence.

Past research on this topic provides predominantly qualitative case studies. This study contributes to existing theory by providing a quantitative analysis of variables across a large number of cases, giving a more general view of the dynamics of conflicts. It shows the trend across a region, allowing for comparison and analysis with other regions of the world. Its practical contribution consists of providing MNEs, indigenous communities and governments with valuable knowledge on the prevention and resolution of these conflicts.

Key words: MNE-indigenous community conflicts, Latin America, community isolation, community governance, MNE activity, national governance

Table of content

1.Introduction ... 6

2. Literature Review ... 12

2.1 MNE influence, activities and governance ... 13

2.2 National governance in conflicts ... 15

2.3 Clash of interests and institutions ... 17

2.4 Isolation of indigenous communities ... 19

3. Theoretical Framework ... 20

3.1 Factors affecting MNE-indigenous community conflicts ... 20

3.2 Political stability and lack of violence ... 22

3.3 Extractive industries vs. non-extractive industries ... 24

3.4 Geographic Isolation of Indigenous Communities... 26

3. 5 Regulatory quality ... 29

3. 6 Democratic nature of indigenous community ... 32

3. 7 Conceptual model ... 33

4. Method & Analysis ... 34

4.1 Data collection and sample ... 34

4.2 Variables and Method ... 35

4.2.1. Dependent variable ... 35

4.2.2. Independent variables ... 36

4.2.3. Moderating variables ... 37

4.2.4. Control variables... 38

4.3 Method ... 39

5. Results and Analysis ... 41

5.1 Descriptive statistics ... 41

5.2 Correlation analysis ... 42

5.3. Multicollinearity statistics ... 46

5.4 Regression Analysis ... 47

6.1. Findings ... 51 6.2 Limitations ... 55 6.3 Theoretical Implications ... 56 6.4 Managerial Implications ... 57 Conclusion ... 58 7. Bibliography ... 60

1. Introduction

Since the arrival of European settlers in Latin America, there have been conflicts between indigenous communities and foreign entities entering their territories to engage in value-adding (usually resource-seeking or extracting) activities (Loomba, 1998). Nowadays these foreign entities are mostly multinational enterprises (MNEs) aiming to achieve competitive advantage by gaining access to resources present in these territories (Anderson, 2003; Dunning, 1998). In this age of globalisation, the barriers to internationalise, trade and engage in cross-border activities are getting smaller (Tisdell, 2001; Garrett, 2000). Mining and other resource extraction legislation has become less strict and improved technology allows further development of the means to search, reach and exploit natural resources in the most demanding ecological environments (Ehrenfeld, 2003; Sawyer & Terence Gomez, 2012). This leads MNEs to explore more remote areas of the world that have not been reached by others or have been impossible to exploit until now (Banks & Ballard, 2003). This has made MNEs exploit more territories inhabited by indigenous communities (Murphy and Arenas, 2010; Sawyer & Terence Gomez, 2012).

Stakeholders criticise MNEs for acting in socially and environmentally irresponsible ways and undermining indigenous communities’ habitat, culture, history and identity (Korten, 1995; Pinkey, 2015; Weaver, 2001). As a consequence of the exploitative nature of the MNE activities, conflicts arise because the indigenous communities receive little of the benefit and profits (Anaya, Williams, 2001; Whiteman & Mamen, 2002; Ballard & Banks, 2003). The circumstances and characteristics of these conflicts are diverse and have different outcomes (Gertz & Oetzel, 2007). In many cases, one of the stakeholders resorts to violence (Ballard & Banks, 2003; Gertz & Oetzel, 2007).

Most of the research in this field consists of case studies of specific conflicts, specific indigenous communities, specific countries or specific industries (Fontaine, 2011; Goyes & South, 2016). The past research looks in-depth into conflicts and their actors, and give detailed and descriptive breakdowns of the events (Anaya, Williams, 2001; Jamhali & Mirshak, 2009). However, research on this topic is mainly qualitative analysis and does not show a large overview of possible trends and relationships that may or may not exist between recurring elements of the conflicts. It does not give an explanation for the happenings or characteristics in the conflicts as recurring transnational phenomena or circumstances.

This thesis will look at similarities and differences in certain concepts in order to view a large sample of conflicts as a whole, instead of individually. We will then see which of these characteristics have an influence on the MNE-indigenous community conflicts. Moreover, most of the analyses cover conflicts in the United States of America, Canada and Australia. Research is especially lacking for conflicts in African and Latin American cases. In this paper, Latin America is the area of focus. Many countries in this region have a long history of MNEs exploiting indigenous populations and causing conflicts with them (Galeano, 1972; Manriquez & Sanchez, 2003; Garrido & Renee, 2015).

Governments in Latin America, past and present, have often not operated in the benefit of indigenous communities, so national governance is a concept on which this research will fill some prevailing gaps (Kroger & Lalander, 2016). The first research question of this thesis is:

RQ1: To what extent does national governance affect the degree of violence of MNE-indigenous community conflicts in Latin America and how does regulatory quality moderate this relationship?

National governance concepts are of great importance to MNE-indigenous community conflicts (Sawyer &Terence Gomez, 2012). We investigate the concept “political

stability and lack of violence” which illustrates to what extent a country has a history of being a stable democracy without seizures of power and violent uprisings (Kaufmann et al., 2011). Latin America has a long history of political instability, and of governments violently abusing and marginalising indigenous communities (Rostica, 2015). However, past research focuses on particular governments or countries. No empirical research exists on whether this concept results in greater violence within conflicts. We expect that the higher the score in this concept, the lower the degree of violence within the conflict.

We will also explore the interaction effect of regulatory quality of a country on the above relationship. The legal institutions of a country are of substantial importance in solving conflicts. Past research focuses on the weakness of government institutions to enforce regulations and constitutions in these conflicts (Kroger & Lalander, 2016). The past research mainly looks at breaking laws and the origins of the subsequent conflicts. There are comparisons of countries, but not a widespread consensus to showcase whether this relationship exists across a large number of countries (Kroger & Lalander, 2016). We expect strong legal institutions and competence to have a positive moderation effect on the relationship between political stability and lack of violence of a country and the degree of violence of such conflicts.

Many of the activities of MNEs in conflict with indigenous communities in Latin America are of an extractive nature (Burchardt & Dietz, 2014). These activities go hand in hand with the exploitation of large pieces of land, leading to forced land dispossession, a decrease in livelihood or the natural habitat being affected negatively (Burchardt & Dietz, 2014; Sawyer & Terence Gomez, 2012). Since indigenous communities often depend on the land they live on for their livelihood and as their places to perform spiritual activities or events, this makes these conflicts turn violent frequently (Bruijn & Whiteman, 2010; Sawyer & Terence Gomez, 2012). The research on the effects of other industries on violence is very

limited, and this study will fill the gaps that exist here. We will therefore focus on the difference between extractive industries and non-extractive industries. The second research question is:

RQ2: To what extent do resource extracting MNEs engage in more violent MNE-indigenous community conflicts in Latin America and how does regulatory quality moderate this relationship?

We expect that extractive industries will stir up conflicts that have a greater degree of violence.

We will also test the moderator effect of the regulatory quality of a country on the above relationship. Previous research gives insight on governments failing to implement and enforce their legal institutions on extractive activities (Whiteman & Gayle, 2002; Gudynas, 2011). Gaps persist in whether regulatory strength has also resulted in better resolution of conflicts between extractive MNEs and indigenous communities. We expect that in countries where resource extraction is well regulated, there is less transgression of the law and fewer unsustainable or unethical practices. Countries with better legal institutions also have a large influence on MNEs’ corporate social responsibility policies and promote sustainable private sector development. We expect the regulatory quality to weaken the effect of extractive activities resulting in more violent conflicts.

Although there is widespread recognition of cultural and institutional differences, existing literature fails to cover characteristics of indigenous communities as possible root causes of conflict violence. In this thesis, we will address the concept of geographic isolation of indigenous communities. Geographic isolation looks at the extent to which indigenous community members stay within their community and have limited contact with the outside world. Areas more geographically isolated receive less attention from the media and civil

society organisations such as NGOs (Huertas Castillo, 2004). This means there is less supervision of MNE activities by outside parties. Moreover, geographic isolation implies communities are self-sustaining for their livelihood, meaning they are likely to be largely dependent on resources from the land around them. MNE’s disruption or harm to the community environment endangers the existence of these isolated communities (Larsen, 2015; Huertas Castillo, 2004). We expect this to lead to more severe reactions from the indigenous community than if the MNE activity has less effect on the community’s survival (Larsen, 2015). The third research question is:

RQ3: To what extent does the geographic isolation of an indigenous community increase the degree of violence of MNE-indigenous community conflicts?

We expect a greater degree of geographic isolation to cause a higher degree of violence.

We will also test the moderator effect of the level of democracy of the indigenous community on the geographic isolation’s relationship with the degree of violence. Prior research focuses on institutional differences that exist in indigenous governance (Fontana & Grugel, 2016; Gibson et al., 2000). They point out the challenges to involve and stimulate participation because of these differences (Van Cott, 2002; Gustafson, 2002). There is a gap in researching the level of democracy of indigenous communities and its effect on conflict resolution. The concept level of democracy of an indigenous community insinuates that when there is consent or agreement between the MNE and community, the majority of the community accepts the decision. Although in many conflict cases we cannot speak of consent, democracy facilitates the communication and negotiation between the parties. We thus expect community democracy to make geographic isolation a smaller obstacle to conflict resolution.

This research is an exploratory and quantitative research, looking at 277 conflicts in Latin America. The conflicts include cases from: Argentina, Belize, Bolivia, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, El Salvador, Guatemala, Guyana, Haiti, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, Surinam, Uruguay and Venezuela. We will make a regression analysis to test our hypotheses regarding MNE-indigenous community conflict and shine more light on the dynamics of conflicts.

It will contribute to the existing theory in several ways. Firstly, it will be useful to identify the various conditions and characteristics that can be transferred across different contexts and settings. We expect patterns of recurring phenomena by which we can relate conflicts to each other and create new frameworks for theory.

This thesis has a considerable practical contribution. This regional analysis shows differences between national institutions, their respective governance scores and how these differences affect conflicts. It creates a space for similar research in other regions. Comparisons with other regions can provide further learning on how to provide the best solution for MNE-indigenous community conflict resolution. This research will also help in calling other entities or organisations to monitor and control these conflicts more closely to exert pressure that will eventually lead to better compromise for all parties and less violent escalation of the conflicts. MNEs can take aspects of their foreign direct investment (FDI) or sustainability policies into consideration. Local governments can look to the improvement of their policies to enhance peaceful and constructive negotiations. Governments could also establish themselves more effectively as protectors of the vulnerable indigenous communities, their lands, cultural heritage and identity in order to improve conflict resolution.

and MNEs. A theoretical framework will cover the core argument of this study. The methodology and explanation of the variables will come next. The data analysis, findings and the discussion of results follow. We will end with reflection on this study’s limitations and a conclusion.

2. Literature Review

To answer the research questions and understand properly what is meant by MNE-indigenous community conflicts, we must first define the ambiguous term “MNE-indigenous”. Although there is not a widespread agreement on the definition of “indigenous” (Canessa, 2007), the International Labour Organization (ILO) proposed in the Indigenous and Tribal Peoples Convention No. 169 the following definition of indigenous peoples:

“(a) tribal peoples in independent countries whose social, cultural and economic conditions distinguish them from other sections of the national community, and whose status is regulated wholly or partially by their own customs or traditions or by special laws or regulations;

(b) peoples in independent countries who are regarded as Indigenous on account of their descent from the populations which inhabited the country, or a geographical region to which the country belongs, at the time of conquest or colonisation or the establishment of present state boundaries and who, irrespective of their legal status, retain some or all of their own social, economic, cultural and political institutions (ILO, 1989, Article 1)”.

Regarding the widespread accounts of abuse and misplacement of indigenous communities, the lack of empirical research conducted on the matter leaves a lot of headroom to make scientific contributions (Whiteman & Mamem, 2002; Calvano, 2008). We will analyse the roles and aspects of the MNEs, the indigenous communities and the national governance, as well as the relationships between them.

2.1 MNE influence, activities and governance

Due to the increasing rate of globalisation, MNEs have grown to possess and exercise increasing power over national governments’ decision-making (Dunning, 1998; Garrett, 2000). Due to the tax income they provide, the investments they make in a country, the employment they give to locals and their defence by international trade agreements, they are an important and protected player that can influence national policies (Newell, 2002; Parkinson, 1995; Sawyer &Terence Gomez, 2012). Newell (2002) also sees a trend in national and local authority decreasing at the cost of MNE power increasing. For example, some agreements allow companies to sue local authorities for setting environmental standards which they claim to be forms of discrimination against foreign MNEs, discouraging legal steps against MNEs (Newell, 2002).

The increase in their power has enabled the MNEs to increase their possibilities of investing and gaining access to almost all countries in Latin America (Banerjee, 2003). Their activities have consequences for local communities living in the proximity, and inevitably the indigenous communities (Whiteman, 2002; Galeano, 1972). Most previous literature focuses on the negative externalities of the activities of resource extractive industries on indigenous communities (Larsen, 2015; O’Faircheaellaigh, 2013; Sawyer & Terence Gomez, 2012).

Lertzman and Vredenburg (2005) study the interaction between extraction companies and indigenous communities. They acknowledge the frequency of conflicts and focus on how these conflicts can be managed and solved (Lertzman & Vredenburg, 2005). They state that the only way to find a compromise that is mutually beneficial, is to reach an agreement with the tribes themselves (Lertzman & Vredenburg, 2005). They find that by giving communities more negotiation power and by looking at the conflicts from the community perspective, the negotiations and interaction are more peaceful (Lertzman & Vredenburg, 2005). In this way there is respect and awareness of community interest (Lertzman & Vredenburg, 2005;

Calvano, 2008). Castro and Nielsen (2011) also emphasize that these conflicts need co-management in which stakeholders make the arrangements, implying that the indigenous communities get a say in the management of the resource extraction.

This is challenging to achieve because of the collision of interests that exists between the MNE, national authority and the indigenous community (Ali, 2003; Reed, 2003). For the communities, conservation and protection of the natural ecosystem is of primary interest. For governments and MNEs, economic development and profit have a higher priority (Twyman, 2000; Kroger & Lalander, 2016; Lalander, 2014). The differences in negotiation power between communities and MNEs, due to the possession of capital and government support, cause new conflicts and the escalation of existing ones (Castro & Nielsen, 2011). Calvano (2008) adds to this by concluding that the three causes of conflict are stakeholder power inequality, stakeholder perception gaps and cultural context. We will discuss the last two in more detail later on.

In addition to this, the circumstances in which problematic encounters between MNEs and indigenous communities are born are diverse. For instance, indigenous communities have been removed from their lands in the name of nature conservation (Spence, 1999). Some examples of this happening were the establishment of the world-famous Yosemite and Yellowstone national parks (Spence, 1999). It seems controversial that tribes that have lived in harmony with nature for centuries need to make place for entities that want to conserve this area, but with commercial motives (Spence, 1999). Other conflicts develop due to the building of dams, windmill parks or agricultural land (Rodriguez, 2014; Bratman, 2014). This poses the question whether the degree of violence in conflicts has a relationship with the activities of the MNE.

Bellier and Preaud (2011) state that there has been progress in the recognition of indigenous communities in recent years. The participation of indigenous communities in the

writing of constitutions as well as establishing educational institutions is proof of this (Bellier and Preaud, 2011; Gustafson, 2002). Although the arrival of MNEs has caused suffering to indigenous communities, researchers acknowledge an increase of MNE corporate social responsibility (Fontaine, 2011; Lertzman &Vredenburg, 2005). However their actions should not compromise the local community’s property rights or health and safety matters (Burchardt & Dietz, 2014; Larsen, 2015). Development of good corporate governance is necessary to create a sustainable future for all parties (Howitt, 2012)

The literature in corporate governance and sustainable practices regarding indigenous communities is growing significantly. MNEs cannot afford scandals due to the media attention and the disastrous consequences for their image and value (Calvano, 2008). The reputational damage they suffer, as well as the costs of solving the conflict or even having to leave the country due to unacceptable practices, can be enormous (Oetzel et al., 2007; Calvano, 2008).

However, regarding the multiple examples of violations of indigenous rights in recent years, other researchers question whether there has been an improvement for the indigenous communities (Sawyer & Terence Gomez, 2012). Legislation on FDI and especially resource extraction is still loosening up and facilitating MNEs to exploit territories in Latin America, with Argentina as the latest example (Jamasmie, 2016). A large amount of scepticism exists about whether MNEs are showing progress in sustainability practices and whether governments do enough to control this, blinded by the economic interests of their activities (McNeish, 2013; Lalander, 2014).

2.2 National governance in conflicts

National governments determine for a large part the norms that MNEs and civil society have to comply with. In most of the Latin American countries, whether they have

strong or weak legal institutions, actions by governments have been corrupted or inadequate in protecting human rights of the indigenous communities (Richards, 2010; Reed, 2003; Sawyer, 2004; Kroger & Lalander, 2016). For example, in the case of the Mapuche indigenous tribe in Chile, the state criminalised resistance to MNE land seizure and protests against discrimination. It applied the so-called ‘anti-terrorist law’, which delegitimised the indigenous character and criminalised their mobilisation as attacks against private properties (Richards, 2010). There is an irony about the government’s defence of indigenous people, which really should confront MNEs if any violation of rights or land spoliation takes place (Richards, 2010). Many local governments have not been capable of going against the destruction of the environment through exploitation by MNEs (Whiteman & Gayle, 2002).

Foreign direct investment is an attractive source of income for a government. It comes along with a boost of the local economy and improvement of infrastructure (Lalander, 2014). In many cases, governments see less significance in conserving and upholding the cultural heritage of its indigenous tribes and allowing them the sovereignty and independent ruling of their lands (O’Fairchellaigh, 2013; Montesinos Coleman, 2015). The communities’ activities do not add value for economic development of the nation (Lalander, 2014). Governments prefer to stimulate value-adding economic activity rather than making efforts to preserve these communities (Kroger & Lalander, 2016; Richards, 2010). Kroger & Lalander (2016) question whether laws that guarantee the protection of the local communities against MNEs have any effect. The government grants MNEs permission to operate in areas inhabited by indigenous communities. Thus, the government decides over land that is within its territory, and the interest of the community itself becomes of lesser importance (Ali, 2003; Lalander, 2014). MNEs mostly act with consent or cooperation with governments, so their actions cannot be seen as “illegal” (Whiteman & Mamen, 2002).

Many government factors play a role in these conflicts. It ranges from corruption figures to the formal recognition of indigenous communities. The study will focus on the concepts ‘political stability and lack of violence’ as well as ‘regulatory quality’ of countries.

2.3 Clash of interests and institutions

In general, indigenous customs and traditions are distant from the capitalist thought. They have resisted to institutions of the free-market society, dominant in the world today (Galeano, 1972; Fontana & Grugel, 2015; Tuwihai Smith, 1999; Nietze, 2003). North (1991) states that institutions in indigenous societies are characterised by “tribalism”. This constitutes a “dense social network of informal constraints that facilitates local exchange and where people have an intimate understanding of each other” (North, 1991). The indigenous community sees the threat of foreign entities as a binding sentiment and a mechanism to defend themselves (North, 1991). When foreign entities enter and disrupt their social network, conflicts are bound to occur (North, 1991).

Indigenous tribes have especially strong relationships with the land they live on, in spiritual, cultural and economic sense (Fontaine, 2011; O’Fairchellaigh, 2013; Sawyer & Terence Gomez, 2012). Indigenous communities believe they are the true owners of the land they inhabit, passed on by their ancestors. They view themselves as being part of the land (Weaver, 2001). Land rights form a considerable problem in MNE-indigenous community conflicts. Especially in Latin America there is unclarity about these land rights (Anaya & Williams, 2001). In most countries, the law states that natural resources belong to the government that has sovereignty over its own country’s land and that history of the land’s ownership does not play a role in ownership (Montesinos Coleman, 2015). There are many laws that recognise the sovereignty of indigenous tribes: indigenous titles obtained in their domestic law are often accepted (Anaya & Williams, 2001; Hodgson, 2002; Canessa, 2007).

However different rules of the game exist when it comes to land rights (Montesinos Coleman, 2015). Domestic law is more powerful than the legal action that indigenous communities can take, undermining the indigenous communities’ position in legal sense (Whiteman and Gayle, 2002).

The misunderstanding that occurs due to MNEs and indigenous communities’ lack of communication is considerable. Although Galeano (1972) believes that this problem lies in the hands of greed and arrogance of MNEs, indigenous communities are often a complex partner to come to agreement with (Fontana & Grugel, 2016). They are part of a small community that has resisted the adaptation to modern society despite all the pressures to do so (Larsen, 2015). They are often not familiar with the idea of formal institutions like contracts or what international law, i.e. World Bank mandate, might state about the possession of land (Anaya & Williams, 2001; Reed, 2003). This mandate for example, changed the wording of the Manual of Operation regarding Indigenous Peoples, from “consent” of actions to “consultation” (Anaya & Williams, 2001). This implies that no participatory role has to be given to indigenous communities.

There are cases where firms attempt to make the cooperation work. They do this with the implementation of educative programs, providing jobs or investing in other infrastructure for the community (Assies et al., 2011). Critics however call it ignorant to believe that these communities have the same perception of development as an MNE (Briggs & Sharp, 2004; Tuhivai Smith, 1999). Their interests lie for a large part in “preserving and transmitting their territories, ethnic identities and knowledge to future generations” and not in the creation of new jobs or making their community more profitable (Bartra, 1974; Hodgson, 2002).

What makes the matter even more complex is that many indigenous communities lack formal institutionalised governance (Rodriguez-Garavito, 2011). This makes the decision-making process of indigenous communities dealing with MNEs exploiting their territory

challenging. For the MNE, it is unclear who to negotiate with and who has the legitimacy to speak for the community. MNEs have abused this phenomenon, supposedly acquiring and getting “permission” to use this land. This happened to the traditionally nomadic Machiguenga tribe in Peru. The state gave land titling rights to subgroups to encourage permanent settlement (Sawyer & Terence Gomez, 2012). This divided the community and gas exploitation could take place because of an agreement with these newly created settlements (Sawyer & Terence Gomez, 2012).

In most cases, we cannot speak of ethical dealing with the indigenous community in conflict situations. According to Lertzman and Vredenburg (2005), the only way to avoid acting unethically is to cooperate with indigenous peoples in ways that are at all times agreeable to them. MNEs must satisfy the needs that the indigenous communities put forward themselves.

2.4 Isolation of indigenous communities

The geographic isolation of the indigenous community is another factor to take into account. Indigenous communities often live in relative isolation from other cultures. Although there is obviously a large difference in geographical isolation among indigenous communities, it is essential to identify the consequences. Studies show that geographical isolation of a community leads to further separation from other communities in a cultural sense (Gupta & Ferguson, 1992). There is little interaction with other cultures, which also limits the influence from the outside world (Gupta & Ferguson, 1992).

Moreover, geographically isolated communities are more self-sufficient and rely more on the natural environment for survival. This is especially important for this topic as most MNE-indigenous conflicts arise from land seizure and exploitation (Kroger & Lalander, 2016; O’Faircheaellaigh, 2013). Land dispossession, damage and unsustainable practices are

disastrous for the livelihood of these communities, considering this land could be one of the only sources of income for the community. Geographical isolation also implies other forms of isolation (Huertas Castillo, 2004). For example, very limited media coverage and knowledge about these communities, leading to less social control from other organisations (Huertas Castillo, 2004, Larsen, 2015). We will elaborate on this point in the theoretical framework.

3. Theoretical Framework

3.1 Factors affecting MNE-indigenous community conflicts

There is a large history of problematic encounters between MNEs and indigenous communities. Especially the twentieth century in Latin America has known a vast number of conflicts concerning fossil fuels and mineral ores extraction activities (Galeano, 1972; Garrido & Renee, 2015). It is a well-covered issue in Latin America and the rise of numerous indigenous activist movements in the 1990s has significantly increased the amount of literature on the topic (Sawyer, 2004; Yashar, 2005). Many studies have analysed the conflict processes looking at the MNE characteristics, mostly focusing on the MNE’s sustainability effort and corporate social responsibility (Anderson, 2003; Lertzman & Vredenburg, 2005). Researchers commonly view the MNE as the central actor in these conflicts, whose actions and disregard of the indigenous communities are the main reason for the conflicts to occur in the first place.

The role of national governments in these conflicts has also been the subject of prior research. Corruption plays a central role in conflicts, with the unequal distribution of benefits from extraction activities as a result (Whiteman & Gayle, 2002). Kroger & Lalander (2016) make the claim that governments have increasingly little to say when it comes down to negotiations with MNEs. Their role as enforcers of environmental and human rights and

2016). Other studies claim that the government is the most important body in determining and imposing the ethical norms that civil society and businesses have to comply with (Chabal, 2002). This study will look at the influence of the formation of the government and the implementation of its legal institutions on conflict resolution.

The literature broadly agrees that conflicts take place due to irresponsible behaviour on part of the MNEs or negligence of incompetent government institutions to avoid these conflicts (Whiteman & Mamen, 2002; Garrido & Renee, 2015; Sawyer, 2004). Research moreover agrees on the difference in indigenous culture, values and beliefs from the other actors as being a reason for misunderstanding and disagreement (Anderson, 2003; Sawyer, 2004).

However, there is a large gap the significance of indigenous governance in conflicts. This thesis incorporates some of the relevant community factors. As one of the three actors that are present in conflicts, it is vital to examine the community closer, in relation to national governance and MNE activities. In this thesis, we explore each of the three actors and analyse characteristics relevant to conflict situations. The predominant objective is to observe which of the factors have significant effects on conflict resolution and dynamics. We can then derive which of the actors is most important in their influence on conflicts in Latin America.

According to Oetzel et al. (2007), two main characteristics determine the severity of a conflict: the length of a conflict and the degree of violence of a conflict. Previous studies have focused on each of these aspects for their respective cases. This study will focus on the degree of violence within the conflicts. The more violent a conflict becomes, the more problematic and challenging it becomes to find a solution that both parties can live with.

This thesis will complement previous literature in creating new theoretical foundations on MNE-indigenous conflicts by pointing out the significance and the interdependence of the actors in conflict resolution.

3.2 Political stability and lack of violence

A fundamental part of this research concerns the governments’ roles in the conflicts in Latin America. Disproportionate political turmoil, with many violent uprisings, seizures of power and ousting of presidents characterise Latin America’s history (Santana, 1979; Rostica, 2015). This lack of political stability of a large proportion of the countries we study has been disastrous for ethnic minorities and especially indigenous communities (Yashar, 2005; Rostica, 2015). Yashar (2005) points out the lack of adequate country-level governance as a reason for the suppression of indigenous rights and henceforth an eruption of violence within these groups in the region (Garrido & Renee, 2015; Santana, 1979). Kauffman et al. (2011) tells us that the way a government is formed is one of the pillars of national governance. Illustrated by examples below, we show indigenous communities have suffered due to the lack of national governance.

Gerlach (2003) shows us the case of Ecuador. Here, since the exploration of oil in the Amazonian region in the 1960s, there has been a lot of political chaos (Gerlach, 2003). This chaos reached its peak in the 1996-2000 period in which it had 5 presidents. It had severe impacts on indigenous recognition and rights, leading to the ethnocide of the Shuar indigenous tribe in the Amazon (Gerlach, 2003).

In her research on indigenous recognition and rights in Latin America, Yashar (2005) points out that two main obstacles exist on a political level when it comes down to the protection of indigenous rights and recognition: the lack of political will of individual presidents (and therefore their government) and weak state institutions. Far-right wing movements are responsible for many of the undemocratic seizures of power in Latin America (Yashar, 2005; Mainwaring & Pe_{́rez-Liñán, 2013; Santana, 1979). Their political priority did} not lie in the protection of marginalised minorities such as the indigenous communities (Yashar, 2005). Moreover, governments that have taken power by force and have not been

elected through a democratic process are also more likely to use force or unethical methods to suppress opposition (Rostica, 2015; Mainwaring & Pe_{́rez-Liñán, 2013).}

Most of these studies relate to specific examples of governments or the political evolution of certain countries (Manriquez Sanchez, 2003; Gerlach, 2003) Even though they offer in-depth analysis of each individual case they treat, they do not offer an explanation that can be transferred across geographical settings for the case of indigenous communities. Given the fact that so many similarities exist between the political unrest in countries and their respective marginalisation of indigenous peoples, it is necessary to make a quantitative analysis to clarify and generalise these relationships.

Another example of a country in Latin America where the political disorder had grave consequences for the indigenous population is Guatemala (Rostica, 2015). In 1982, the military junta under command of Efraín Ríos Montt took power. This brutal regime reigned with suppression of the opposition and widespread violation of human rights which the indigenous Mayan tribes protested heavily. Under Montt’s orders, the army started a genocide of the Mayan communities, looting villages, raping women and eventually killing an estimated 100,000 people (Rostica, 2015).

According to Chabal (2002), a higher degree of democracy leads to the stability of a country and eventually to a more developed country with institutions of a higher quality. Thus the democratic foundation of a government is essential for its progress and improvements of institutions. We can relate this back to the second reason Yashar (2005) identifies as an obstruction to indigenous recognition: weak state institutions. A state with stronger institutions will also have a larger influence on the sustainability policies of firms and set the general norm that firms need to abide by (Kolk & Lenfant, 2013).

Another example that illustrates the importance of the role of the state is that of Bolvia’s “gas war” in 2003 (Garrido & Renee, 2015; Terence & Sawyer Gomez, 2012). The

government had suppressed the widespread protests against gas extraction by indigenous groups. More blockades and strikes brought the country to a standstill, and eventually led to the forced resignation of President Gonzalo Sanchez de Lozada. Extensive violence erupted, eventually killing 60 people of the Guarani indigenous community.

At the hand of this example and the existing theory the following hypothesis is formulated:

H1: The higher the degree of “political stability and lack of violence” in a country, the lower the degree of violence in an MNE-indigenous community conflict

Political stability indicates the current political power to be in a position where no threat of undemocratic coup by outside parties exists (Kaufmann et al., 2011). Moreover it incorporates violent uprisings and coups of the past in a country. If there is no existing violent culture within a country, these conflicts are also less likely to turn violent. Vice versa, if the use of violence is a widespread phenomenon in the country, this would mean that conflicts would turn more violent. The large variance in countries’ scores in this concept across Latin America will provide for a meaningful and relevant analysis.

3.3 Extractive industries vs. non-extractive industries

Extractive industries play a role in the majority of the conflicts that have arisen between multinationals and indigenous communities (Sawyer & Terence Gomez, 2012; Kroger & Lalander, 2016; Hilson, 2002). Mineral extraction increased enormously in Latin America since the 1970s, also increasing the frequency of conflicts between MNEs and indigenous communities (Ballard & Banks, 2003). Sawyer (2004) puts forward that indigenous communities bear the burden of the negative externalities of resource extraction

and see little of the benefits of the profits. They are the most harming industries in terms of the ecological impact, transforming land, destroying or disrupting the ecosystem and causing extensive pollution and contamination (Hilson, 2002; Sawyer & Terence Gomez, 2012). This is particularly harmful for indigenous communities because of their dependence on the environment for their livelihood. Most extractive activities cannot ensure a sustainable alternative for indigenous communities in which both the MNE activity proceeds and the community can live in the same continuity prior to the arrival of the MNE. Whiteman and Mamen (2002) identify that not only does the environment face threats due to the arrival of the MNE in an indigenous territory, it also instigates various other negative consequences for socio-economic life of the communities, such as the threat to their spiritual and cultural identity and an increase of corruption among others. In accordance with existing literature, the United Nations put forward the six recurring reasons for conflicts between extractive MNEs and indigenous communities: 1) poor engagement of communities and stakeholders, 2) inadequate benefit sharing, 3) an extreme impact on the economy, society and the environment, 4) mismanagement about funds and financing war, 5) inadequate institutional and legal framework and 6) willingness to address the natural resources question in peace agreements (United Nations, 2012).

It is difficult to involve the communities in cooperation and participation of extractive industries, as Fontana and Grugel (2016) found out in their research in Bolivia. Other industries, such as the exploitation of opportunities for tourism, allow for a better cooperation with the indigenous communities (Buultjes & Fuller, 2007). Some of these industries create jobs or opportunities for development of the communities that do not compromise their livelihood or environment (Buultjes & Fuller, 2007).

It is remarkable that most studies of MNE-indigenous community conflicts have placed their focus on extractive industries and have not succeeded in incorporating other

industries as well. There is an evident gap to investigate whether the dynamics of conflicts of extractive MNEs and indigenous communities differ from the other industries and do in fact cause more violence. It is possible that there is more focus on extractive industries than other industries, possibly causing similar damage. This study will look at whether there are differences on this issue and look into the relationship between an MNE being active in an extractive industry and the degree of violence of the conflict.

One example of violence erupting due to resource extraction is in the case of the U’wa indigenous tribe in Boyaca, Colombia. Occidental Petroleum Company from the United States of America started oil exploration on their territory in 1992, exploiting lands with sacred sites and destroying the natural habitat in order for their activities to take place (Miranda, 2007). Demonstrations against this oil exploration started and in one protest, violence erupted and led to the death of 3 people in 2002 (Miranda, 2007).

Based on past literature and this example, our second hypothesis is:

H2: There is a positive relationship between the MNE activity being of an extractive nature and the degree of violence of an MNE-indigenous community conflict.

We expect a direct positive relationship with a firm operating in an extractive industry and the degree of violence of the conflict.

3.4 Geographic Isolation of Indigenous Communities

Another factor that will be interesting to explore is if the geographical isolation of communities is significant in these conflicts. Two characteristics of indigenous communities that complement each other are its identity and its geography (Calvano, 2008). They feel themselves as being part of the ancestral land they live on, and this natural surrounding

becomes part of their belief and culture (Whiteman & Cooper, 2000). Their identity also has to do with the connection they have to other surrounding communities or the extent to which they are embedded within a broader community (Weaver, 2001). Geographic isolation of a community indicates that they are very much embedded within their own group identity, with little outside influence of other communities (Larsen, 2015). Some geographically isolated communities deliberately try to avoid all sorts of contact with the outside world in order to maintain their cultural and spiritual identity (Huertas Castillo, 2004). The group identity of indigenous communities reinforces once an external group potentially threatens their existence (Bruijn & Whiteman, 2010). It is used as a defence mechanism and might make the indigenous community’s reaction to an outside party arriving stronger, and possibly more hostile and violent.

Past cases show that the more isolated a community, the less media coverage or national attention exists for these conflicts (Huertas Castillo, 2004). This might result in higher degree of escalation of the conflict, as there would be less social control of the conflict. Geographic isolation also means it will be more difficult for MNEs to provide solutions for reallocation of land. There has been considerable abuse on removing communities from their land and redistributing them to other lands, disrupting their whole lifestyle and cultural heritage (Paige, 2005). Moreover, geographic isolation leads to having little connection with the outside world, which in turn might lead to less external pressures on MNEs, by for example NGOs or human rights organisations. These would force them to adjust more to the needs of the indigenous community and allow for more participation from the community (Huertas Castillo, 2004).

According to Huertas Castillo (2004), when indigenous communities are more geographically isolated, this means the MNE most likely has no knowledge on how to deal

with this group. Prior knowledge and past experience cannot be applied as effectively with such communities. This creates more misunderstanding and disagreement between the actors. Cultural distance also comes hand in hand with geographic isolation. This results in further complexity of the conflict, as the differences in views and approaches will cause misalignments in communication and understanding of each other (Huertas Castillo, 2004). The probability that a conflict will then resolve itself in a way that works well for both parties is minimal (Bruijn & Whiteman, 2010). Moreover, these differences also decreases the opportunity for co-management of projects to take place (Fontana & Grugel, 2016). Collaboration and co-management is mostly founded on a basis of trusting each other (Lertzman & Vredenburg, 2005). This degree of trust will be harder to reach with the greater cultural distance that may exist between the parties.

Huertas Castillo (2004) gives the example of the geographically isolated Madre de Dios indigenous communities in the Peruvian Amazon. She tells us that these communities claim that purposeful violence has been used to encroach their land to facilitate resource extraction. They “have been hunted down like animals”, without any outside organisation with the knowledge, ability or will to go against these violations (Huertas Castillo, 2004).

Based on this example and the existing theory on the geographic isolation of indigenous community we formulate the third hypothesis:

H3: The higher the degree of geographical isolation of an indigenous community, the higher the degree of violenceinf an MNE-indigenous community conflict

We expect a positive relationship between the geographical isolation of the indigenous community and the degree of violence in MNE-indigenous community conflicts.

3. 5 Regulatory quality

The country’s ability to apply its laws and ensure that justice is done, is an influencing factor. Sawyer & Terence Gomez (2012) show that a government’s regulatory system that is malfunctioning and leaves space for not abiding, will have subsequent breaking of laws without penalty.

According to Kaufmann et al. (2011), the ability to formulate, implement and ensure regulations that are to the benefit of civil society and the population as a whole characterises good national governance. Its legal system regarding investments should be solid in order to guarantee development and trust (Kaufmann et al., 2011). As mentioned, one of the main factors that endanger implementation of indigenous rights according to Yasher (2005) is the weakness of institutions. The judicial system is the institution that is present to ensure that justice is done and that people have protection from unlawful actions.

In addition to this, Kaufmann et al. (2011) name their third pillar for national governance: the trust civil society has in national institutions and regulations. On many occasions in the past, indigenous communities have not received the regulatory protection that was necessary (Sawyer & Terence Gomez, 2012; Kroger & Lalander, 2016). Developments in Latin America have given rise to the amount of ethno-territorial protective laws (Schilling- Vacaflor, 2012; Van Cott, 2002). However this has coincided with a period in which many new laws promoting FDI and loosening resource extraction legislation took place as well (McNeish, 2013; Grudynas, 2011).

Kroger and Lalander (2016) question the ability of several national governments in Latin America to stand by their commitments to their legislation. They show in their results that: “neither the ratification of ILO-169 on the rights of Indigenous peoples nor progressive constitutional reform guarantees that Indigenous and other ethnic land rights are given priority over mining, logging, pasture-land, industrial plantation expansion, oil extraction, or

similar industrial activities” (Kroger & Lalander, 2016, p. 682). This leads to a substantial decrease in the amount of trust indigenous communities have in their national governments. If they feel they will not be protected by the national governing body, this incentivize them to take justice into their own hands. Subsequently more violent conflicts are likely to occur.

An example of a case is that of the Yatsuni territory, home to several indigenous communities, in Ecuador. In 1995, the Yatsuni initiative planned to stop oil drilling in this protected part of the Amazon. The Yatsuni initiative received global attention and support from media and NGOs such as Oilwatch and Accion Ecologica, and the Ecuadorian government recognised the initiative (Lalander, 2014). They implemented the initiative, together with further protection of the Yatsuni territory in the 2008 constitution. However, the Ecuadorian president Rafael Correa declared the end of the Yatsuni initiative by presidential decree and allowed for oil drilling to go ahead in 2013, with the approval of the national assembly later that year (Lalander, 2014). This is just one of the many examples where a democratically elected government has failed to act in a coherent and accountable manner, resulting in absence of trust from the indigenous community.

Using this example and the literature on this matter we will test the following hypothesis:

H4: The regulatory quality has a positive effect on the relationship between political stability and lack of violence and the degree of violence of MNE-indigenous community conflicts

We therefore expect a high score on regulatory quality to weaken the negative relationship between political stability and lack of violence and the degree of violence in a conflict.

Researchers point out that states have limited power to confront extractive MNEs with human rights violations and environmental damage they cause (Gerlach, Kotger & Lalander, 2016). Extractive MNEs are usually large corporations with a lot of capital power, protected by international trade laws (Sawyer & Terence Gomez, 2012; Lalander, 2014). As mentioned, their activities have disastrous consequences for indigenous communities (Whiteman & Gayle, 2002; Ali, 2003; McNeish, 2013). National governance institutions are present to ensure institutions are to the benefit of civil society and the population as a whole.

We assume that the strength of the regulatory quality of a country reflects in its legislation regarding the exploitation of natural resources. There should be strong control to ensure that exploitation or extraction does not have negative effects on the indigenous community or its environment. If not, appropriate jurisdiction and sanctions by the government is executed effectively and the conflicts become less violent.

The Blackfire mine in Chicomuselo, Mexico provides us with an example of strong regulatory intervention. After the widespread protest against environmental violations of the Canadian mining company Blackfire on territory of the tzeltales indigenous community, an activist was gunned down and killed. The Ministry of Environment of Mexico took action and immediately shut down the mine permanently to avoid further escalation of violence. Using this example and previous literature on this matter, we formulate the fifth hypothesis:

H5: The regulatory quality has a moderator effect in the relationship between the extractive nature of an MNE and the degree of violence of MNE-indigenous community conflicts

3. 6 Democratic nature of indigenous community

The governance of the community itself also plays a large role in how these conflicts evolve. Especially in the process of consultation and negotiating, it can be very challenging to come to agreements with indigenous communities because of their lack of official governance (Assies et al., 2001). In contrast to state institutions or MNEs, many indigenous communities are “intrinsically traditional, pre-capitalist and uniquely in favour of collective rights” (Fontana & Grugel, 2016, p.257). In many encounters between MNEs and indigenous communities there is a lack of clarity about which representative is to speak for who and with what legitimacy (Fontana & Grugel, 2016). Given the different institutions and formal rules these communities have, it is difficult negotiate if there is non-democratic rule or with communities that do not have chosen leadership. There is often a lot of strife and contestation of power within the community (Rodriguez Garavito, 2011; Gustafson, 2002). According to Agrawal and Gibson (2009), the extent to which indigenous communities really have representative practices depends on many factors. Most importantly, they state that the quality and effectiveness of indigenous community institutions are essential to the participation of the community members (Agrawal & Gibson, 2009).

Fontana and Grugel (2016) recognise the problems that exist in Latin America because of a coinciding struggle to uphold indigenous rights for their self-governing systems, whilst in the meantime trying to accelerate the democratisation process. They claim that indigenous communities that possess democratic governance systems have “intrinsically legitimate, inclusive and non-authoritarian rule” (Fontana & Grugel, 2016, p. 257).

Geographically isolated communities are bound to be under decentralised authority (Gustafson, 2002). By ensuring that democracy is present and by institutionalising community governance at local level, this will neutralise the possibilities of anti-systemic mobilisation (Fontana & Grugel, 2016). Therefore, once democratic institutions ensure that

indigenous communities have been heard, there will be a higher degree of consent and participation, which will result in less violent uprisings.

Based on this example and the theory on democracy within indigenous communities, we formulate the following hypothesis:

H6: The democratic nature of the indigenous community has a moderator effect in the relationship between the geographic isolation of an indigenous community and the degree of violence of MNE-indigenous community conflicts

The expectation is that the democratic nature of an indigenous community will reduce the effect of geographical isolation causing a higher degree of violence in conflicts.

3. 7 Conceptual model

Figure 1 below shows the conceptual model that represents the variables and relationships between them. H1 represents hypothesis 1, the negative relationship between “political stability and lack of violence” and the degree of violence of MNE-indigenous community conflict. H2 represents hypothesis 2, the positive relationship we expect between the extractive nature of the MNE and the degree of violence. H3 represents the positive relationship we expect between the geographic isolation of an indigenous community and the degree of violence. H4 and H5 represent the interaction effect, decreasing the degree of violence in both cases, that the moderator variable regulatory quality we expect to have on the relationships established in H1 and H2. H6 represents hypothesis 6, where the moderator “democratic nature of an indigenous community” we expect to decrease the degree of violence of the conflict.

Figure 1. Conceptual Model

4. Method & Analysis

4.1 Data collection and sample

In order to do an empirical study and reach valuable insights on these conflicts, a group of students investigated the same topic and collected data on MNE and indigenous community conflicts across the world, as well as country facts of figures of where the respective conflicts have occurred. As the focus of this paper was to look at the conflicts in Latin America, we filtered for this criteria, resulting in a total of 277 cases in Latin America. We created a database of the conflict where we gathered data about 83 variables per conflict and coded it into a common scheme. These variables then gave us more information on each of the three players: the MNE, the indigenous community and the government, and their

Geographic isolation H1 Regulatory Quality Degree of violence of MNE- Indigenous community conflict Political Stability and Lack of Violence Extractive nature of MNE Control Variables

MNE Stated policy Community Autonomy Governmental effectiveness H2 Democratic nature of community H4 H5 H3 H6

interactions in the conflicts. For the conflict to qualify as an MNE-indigenous community conflict, whatever kind of disagreement between a multinational company and an indigenous community has to be present, no matter the duration or intensity.

We found information on the conflicts using a vast number of resources. These included local newspaper reports, other local or regional media sources, anthropologic studies, activists’ reports, NGO observations, Environmental Justice Atlas, Minewatch, MNE sustainability reports and websites and sources from governmental institutions. We found most country-level variables from the World Bank and the Central Intelligence Agency databases, where we selected the most recent entries, mostly from 2014.

The sample consists of 277 cases from Latin America. The conflicts include cases from: Argentina, Belize, Bolivia, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, El Salvador, Guatemala, Guyana, Haiti, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, Surinam, Uruguay and Venezuela.

4.2 Variables and Method

4.2.1. Dependent variable

We chose the degree or level of violence the conflict had reached at the peak of its intensity as the dependent variable for all hypotheses. It will be of great interest to see if there is a pattern of the independent variables significantly affecting the degree of violence of a conflict. The ‘degree of violence’ variable was coded on a scale from 1 to 5 which is shown below:

1. Peaceful protests by the indigenous community or peaceful negotiations between the community and the MNE.

2. Court action was undertaken by the community in reaction to the MNE or the government’s actions

3. Low-level violence erupted due to confrontations between the community and the MNE, including occupation of property (from either side)

4. High-level violence erupted due to confrontations between the community and the MNE, including serious injuries or damage or kidnappings (from either side)

5. Death was a result of the confrontations between the community and the MNE from either side (Calvano, 2008).

4.2.2. Independent variables

The first independent variable is “political stability and lack of violence”, one of the six dimensions of governance measured by the Worldwide Governance Indicators (WGI) which are “a long-standing research project to develop cross-country indicators of governance” (Kauffman et al., 2011). This particular variable consists of “the likelihood that the government will be destabilized or overthrown by unconstitutional or violent means, including politically-motivated violence and terrorism” (Kauffman et al., 2011). Each country in this sample has a percentage score for this variable. This variable was then coded into six categories according to the percentage range that they country scored: (1) 0-10%, (2) 10-25%, (3) 25-50%, (4) 50-75%, (5) 75-90% and (6) 90-100% (World Bank, 2015). This variable will be used to test the first and fourth hypothesis.

The second independent variable is the MNE’s activity. Here we can look at the environmental category the activity of the firm falls into and what is therefore the source of the conflict. We categorised this variable in the database as follows: (1) nuclear activities, (2) mineral ores and building extraction, (3) waste management, (4) biomass and land conflict,

(5) fossil fuels extraction, (6) water management, (7) infrastructure, (8) tourism creation, (9) biodiversity conservation conflict, (10) industrial and utilities conflict and (11) other (EJAtlas, 2016).

For this variable, the goal was to see whether extraction of natural resources, therefore

all conflicts that fall into category 2 and 5, were to cause a higher degree of violence than

the others. We created a dummy in which 1 corresponded to all conflicts caused by natural resource extraction activities and 0 to all other kinds of conflict. We will test the second and fifth hypothesis with this independent variable.

The final independent variable is the geographic isolation of the indigenous community. This variable will be used to find out the third and sixth hypothesis, to see whether a higher degree of isolation also results in higher degree of violence. For geographical isolation the coding will be:

1. Most community members live within broader communities with mixed populations, i.e. not highly isolated.

2. Highly isolated, i.e., community members live within community with little contact to the outside world.

4.2.3. Moderating variables

The regulatory quality is another one of the WGI scores of countries and is defined as “the ability of the government to formulate and implement sound policies and regulations that permit and promote private sector development” (Kauffman, 2011). The fourth and fifth hypotheses state that the “regulatory quality” of a country moderates the relationship between ‘political stability and lack of violence’ and degree of conflict and the relationship between the MNE activity and the degree of violence respectively. In the same way as with ‘political stability and lack of violence’, this variable is coded into six categories according to the

percentage range that they country scored: (1) 0-10%, (2) 10-25%, (3) 25-50%, (4) 50-75%, (5) 75-90% and (6) 90-100% (World Bank, 2015).

The second moderating variable that we will use is the indigenous governance, and more specifically the degree to which it is a democratic one. This variable will be made up of the concentration of power which is a categorical variable and is coded, using ethnographic studies of the communities as follows: (1) acephalous, (2) democratic, (3) age-group-based, (4) elder council and (5) monarchical (Currie et al., 2004).

The interest of this study was to see if there would be an effect on the relationship between geographic isolation and degree of violence, if we have a democratic community or one with another type of governance. A dummy was therefore created with 1 meaning there was a democratic society and 0 of any other form of governance.

4.2.4. Control variables

There are three control variables we use to check all relationships between the independent and dependent variables and moderator.

The first is the community autonomy. This is the extent to which the government and the constitution of the country recognises the indigenous community as an independent entity, and are respected in having their own form of governance and traditions. This variable was coded from 1 to 5 going as follows: (1) no autonomy at all, (2) there is cultural recognition of the existing tribe, (3) there is limited recognition of rules and own laws (communities can issue some minor rules like marriages recognized as valid by government), (4) partial recognition of rules or right of consultation and (5) almost full recognition of community rules by government with autonomy or right of consent.