The roles of governments and NGOs in the resolution of conflicts between indigenous communities and MNEs

(1)

1

The roles of governments and NGOs in the resolution of

conflicts between indigenous communities and MNEs

Thesis Master of Science Business Administration

Track International Management

Name: Suzan van Eijden

Student number: 11411635

Supervisor: Dr. Ilir Haxhi

Second reader: Drs. Erik Dirksen

23 June 2017

(2)

2

Statement of originality

This document is written by Suzan van Eijden who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

3

Abstract

Multinational enterprises (MNEs) in the extraction industry and indigenous communities are often entangled in violent and enduring conflicts, resulting from their diverging interests. Although these conflicts can have devastating effects, the current literature only addresses the dynamics of the resolution processes to a limited extent. The role of the government in resolving the conflicts receives fragmented attention, and the interaction between the influence of governments and non-governmental organisations (NGOs) remains unstudied. By building on stakeholder theory, in the current study, I argue that, first, the level of autonomy of an indigenous community affects the resolution of conflicts, and second, the involvement of NGOs moderates this relationship. The active involvement of Latin American states in the International Labour Organisation Convention 169 and the United Nations Declaration on the Rights of Indigenous Peoples, and the constitutional changes towards the formation of pluri-national states represent promising developments regarding the autonomy of indigenous communities in Latin America. Therefore, the used sample consists of 221 conflict cases from 21 countries in Latin America. Against my expectations, conflicts between MNEs and autonomous communities tend to be more violent and longer than conflicts that involve communities that enjoy lower levels of autonomy. Furthermore, I do not find any support for the effect of the involvement of NGOs on the severity and the duration of a conflict. Although these findings contradict previous studies, the current study complements existing knowledge in three ways. First, it offers a different angle by showing that the effect of community autonomy is not always unambiguously positive. Second, a stakeholder theory perspective sheds new light on the potential interaction between governments, NGOs, and the other involved actors in the resolution of a conflict. Third, this study advances the understanding of global drivers of conflict resolution by taking a quantitative, rather than a qualitative approach. Finally, this study is of practical relevance, since it brings both governments and managers a step closer to understanding the dynamics of conflict resolution. The negative effects of community autonomy suggest that governments should reconsider the motivations behind the numerous autonomy movements in Latin America. Moreover, the absence of the effect of NGO involvement on the resolution of conflicts helps managers to determine their priorities regarding their efforts to resolve conflicts.

Keywords: indigenous communities; MNEs; resource conflicts; conflict resolution; autonomy; NGO involvement; Latin America

(4)

4

List of tables and figures

Table 1 Summary of the regression models ... 33

Table 2 Descriptive statistics and correlations ... 35

Table 3 Tolerance and Variance Inflation Factor values ... 37

Table 4 Regression results for dependent variable "Degree of violence" ... 38

Table 5 Regression results for dependent variable "Human rights violation" ... 39

Table 6 Regression results for dependent variable "Duration” ... 40

Table 7 Conflict cases in the sample by country ... 53

Figure 1 Hypothesised relations between community autonomy, NGO involvement, conflict severity, and conflict duration ... 27

(6)

6

1. Introduction

Continuously increasing globalisation pressures have intensified the competition between multinational enterprises (MNEs) when it comes to the access to and exploitation of natural resources (Rosenfeld Sweeting & Clark, 2000). Since a large part of the world’s natural resources are located in territories where indigenous communities live, MNEs in general, and particularly in the extraction industry, often encounter indigenous communities when exercising their activities (Lertzman & Vredenburg, 2005). Indigenous communities are communities which are historically tied to their territories, and they often consider themselves distinct from other groups that are now also inhabiting these territories (Martínez Cobo, 1987 in United Nations, 2009). In this context, MNEs have often been put on the spot for not taking into account the environmental and socio-cultural consequences of their actions, because indigenous communities usually lack power, legitimacy, and urgency compared to other stakeholders of the MNE (Calvano, 2008; Foster, 2012; Mitchell, Agle & Wood, 1997). Since protecting the environment is often a vital part of the indigenous identity (Bruijn & Whiteman, 2009), it follows that indigenous communities and MNEs often have significantly different interests in relation to the way in which these natural resources are exploited and managed.

In addition, communicational differences between indigenous communities and MNEs recurrently lead to misunderstandings amongst indigenous communities regarding the good intentions that MNEs have for supporting the community’s development (Calvano, 2008). Therefore, MNEs often encounter resistance within indigenous communities when making investments, and conflicts between MNEs and communities are an important issue (Lertzman & Vredenburg, 2005). Not only do these conflicts have negative effects for the involved community (Foster, 2012); they also cause financial and reputational damage for MNEs (Laplante & Spears, 2008), and have the potential to disrupt national societies (Suliman, 1999). Therefore, it is in the interest of all parties to investigate ways in which they contribute to the factors that fortify or temper those conflicts. Nevertheless, research on the dynamics of conflict resolution processes is limited, and the prevailing focus on a single conflict or determinant restricts the generalisability of the results (Calvano, 2008; Fontana, Sastre-Merino & Baca, 2015; Kolk & Lenfant, 2012). The current study hence aims to address this shortcoming by quantitatively exploring to what extent the autonomy of indigenous communities and the involvement of NGOs influence conflict resolution processes.

One of the parties that can contribute to the recognition of indigenous communities is the national government, by providing them with community autonomy (Foster, 2012). Although the recognition by law of indigenous communities has recently gone through positive developments, enforcement often falls short. Since the entrance of MNEs into a country likely stimulates the economy, governments tend to shut their eyes to violations of these laws by MNEs (Bellier &

(7)

7 Préaud, 2012; Espinoza, 2015; Foster, 2012). From a multi-level institutional perspective, a community’s degree of autonomy determines its level of embeddedness in a society. Considering the various levels of institutions and the interdependencies between them (North, 1991; Williamson, 2000), various degrees of autonomy exist: formal political autonomy, effective political autonomy, and autonomy as a social entity’s freedom to determine its own destiny (Hoekema, 1996 in Assies, 2005). The prevailing explanation for a lack of recognition builds on the concept of effective political autonomy, which addresses the desire of indigenous communities to participate politically and economically (Hoekema, 1996 in Assies, 2005). To realise this level of autonomy, it is necessary to create a situation that stimulates the enforcement of the formal rules and legislation (Foster, 2012; Hoekema, 1996 in Assies, 2005; North, 1991). According to this explanation, conflicts are hence fed by governments’ neglect of indigenous communities in decision-making processes regarding the start of extraction projects or the distribution of their revenues (Helwege, 2015). However, some authors follow a socio-cultural approach by arguing that economic and political participation is not enough to prevent conflicts; they focus on a higher level of autonomy, which addresses a community’s freedom to determine its own destiny (Acuña, 2015; Espinoza, 2015; Hoekema, 1996 in Assies, 2005). This approach argues that all indigenous communities’ claims to land, resources, and autonomy should be related to their desire to preserve their authentic culture (Acuña, 2015; Espinoza, 2015; Wiessner, 2011). According to this explanation, conflicts are hence influenced by a lack of recognition of indigenous’ ontologies and a threat to their identity (Acuña, 2015; Bruijn & Whiteman, 2010).

Existing research offers a limited perspective of this phenomenon for several reasons. First, in terms of conflict resolution, previous literature addresses the dynamics of these conflicts in a fragmented way, either focusing on the severity or the duration of conflicts (Calvano, 2008; Fontana, Sastre-Merino & Baca, 2015), without considering them both as key dimensions of the conflict. Second, the studies that include the effect of community autonomy in their conflict analyses often focus on ethnic conflicts without MNE involvement, thereby offering a partial standpoint regarding this interaction (e.g. Coakley, 1994; Cunningham, Skrede Gleditsch & Salehyan, 2009). Finally, methodologically, previous studies are often qualitative in nature and hence focus on one or few conflict cases in isolation, thereby failing to contribute towards a global perspective (e.g. Acuña, 2015; Helwege, 2015; Jamhali & Mirshak, 2009; Tockman & Cameron, 2014). Therefore, the lack of studies on the effect of community autonomy on the severity and the duration of conflicts between indigenous communities and MNEs constitutes an important gap in the literature, urging more in-depth comparative research in order to fully understand the dynamics of conflicts and the role of the government herein. This study aims to address this gap by answering the following research question:

(8)

8 RQ1: To what extent does community autonomy affect the severity and the duration of conflicts between indigenous communities and MNEs?

By granting indigenous communities autonomy, national governments can empower them in decision-making processes and hence enable them to make decisions about their own futures (Bruijn & Whiteman, 2009; Foster, 2012). Thereby, autonomy increases the community’s power as a stakeholder in decision-making processes related to extraction projects, which reduces perceived unfairness (Mitchell et al., 1997). Subsequently, the resulting balance in power dynamics facilitates communication between indigenous communities and MNEs, and enables proactive engagement and collaboration (Calvano, 2008). Autonomous communities may feel more respected, and more empowered to participate in peaceful negotiations about extraction projects, which contributes to reduced severity and duration of conflicts (Helwege, 2015).

There is a growing belief that the national government is not the only party that is responsible for natural resource management (Castro & Nielsen, 2001). The increased attention for dispersed authority and shared multi-level governance (Hooghe & Marks, 2003; Peters & Pierre, 1998) suggests that third parties, such as nongovernmental organisations (NGOs), are continuously becoming more important for the relationship between states, indigenous peoples, and MNEs (Bellier & Préaud, 2012; Calvano, 2008; Guay, Doh & Sinclair, 2004). This hence suggests a growing involvement of NGOs in conflict situations between indigenous communities and MNEs. In such situations, NGOs dedicate themselves to serving societal interests by focusing on social, political, and economic goals (Teegen, Doh & Vachani, 2004). From a stakeholder theory perspective, the influence NGOs can exert on decision-making depends on their possession of power, legitimacy, and urgency (Mitchell et al., 1997). NGOs exert their power by deploying both outsider and insider strategies (Peterson, 1992 in Teegen et al., 2004) in order to increase their salience and eventually achieve their goals (Eesly & Lenox, 2006; Mitchell et al., 1997). Outsider strategies have a pressuring character and are aimed at mobilising the public opinion. Due to NGOs’ public legitimacy and the sense of urgency they create for the criticised party, they can pressure both governments and MNEs towards increased accountability (Brown & Kalegaonkar, 2002; Guay et al., 2004; Palacios, 2004; Stavenhagen, 2009). Insider strategies have a collaborative character and aim to directly influence decision-makers. In this respect, a NGO can fulfil a bridging role between the different actors involved in an investment project, such as the national government, MNEs, and local communities. This can enhance collaboration, contribute to the resolution of existing conflicts, and increase the long-term sustainability of investment projects (Dahan, Doh, Oetzel & Yaziji, 2010; Doh & Teegen, 2002; KimDung, Bush & Mol, 2016).

Although several studies recognise the growing importance of NGOs in sustainable development (e.g. Dahan et al., 2010; Teegen et al., 2004), existing research offers a limited perspective of this phenomenon on several aspects. First, the general effect of NGOs on the

(9)

9 dynamics of conflicts, such as the severity and the duration, is not studied extensively. Second, even though the tempering effect of NGOs on the power of governments is recognised (Doh & Teegen, 2002; Florini, 2003; Teegen et al., 2004), the joint influence of national governments and NGOs on the resolution of conflicts, and the interaction between them, remains unstudied. Finally, methodologically, previous studies on the role of NGOs in conflicts are often qualitative in nature and hence deal with one or few cases of conflict, thereby lacking to contribute towards a global perspective (e.g. Arenas, Sanchez & Murphy, 2013; KimDung et al., 2016; Murphy & Arenas, 2011). Considering the current relevance of both national governments and NGOs, this study aims to address this gap by answering the following research question:

RQ2: To what extent does NGO involvement moderate the relationship between community autonomy and the severity and duration of conflicts between indigenous communities and MNEs?

Given the rise of shared multi-level governance (Hooghe & Marks, 2003; Peters & Pierre, 1998) and the importance of enforcement to attain effective autonomy (Hoekema, 1996 in Assies, 2005), NGOs can strengthen the effect of community autonomy on conflict resolution by creating an environment that stimulates adherence to the rules. First, they can pressure both governments and MNEs towards respecting autonomy and thereby they can empower indigenous communities, by using their power, legitimacy, and urgency (Mitchell et al., 1997). Second, the collaborative character of NGOs increases peaceful interaction and collaboration between the different actors in a conflict (Castro & Nielsen, 2001), which can contribute to a reduction in both the severity and the duration of conflicts.

I explore these relationships for a sample of 221 conflict cases from 21 countries in Latin America, since this region is endowed with a lot of natural resources and hence conflicts are an important issue (Foster, 2012). This region provides a fruitful base to analyse the effect of community autonomy, because recent constitutional reforms in Latin American countries are opening new possibilities for community autonomy to advance indigenous rights (Bellier & Préaud, 2012; González, 2015). I analyse the effect of community autonomy as well as NGO involvement with respect to conflict resolution, i.e., conflict severity and conflict duration. The former is operationalised by the degree of violence and the number of human rights violated in a conflict; the latter measures the length of a conflict. I hypothesise that community autonomy has a negative effect on both conflict severity and conflict duration, and that NGO participation negatively moderates these relations.

The current study contributes to both theoretical and practical knowledge in several ways. First, my research offers a different angle by showing that the effect of community autonomy is not always unambiguously positive. Second, it deepens the understanding of the interactive effect of the government-community relations on the resolution of conflicts, and more specifically, the extent to which NGO involvement may influence this effect. This combined

(10)

10 relationship has not been studied before, and a stakeholder theory perspective sheds new light on the mechanisms at play in such relationships by illustrating the importance of stakeholders’ relative power, legitimacy, and urgency in determining the influence they can exert on conflict situations. Third, methodologically, by deploying a quantitative approach, this study offers a broader picture of the phenomenon compared to the vast majority of prior studies on conflict dynamics which take a qualitative approach. Finally, this study can be of practical value for both governments and managers, since resource conflicts, if not addressed, have the potential to disentangle the entire structure of a society (Suliman, 1999), and can be costly and time consuming for MNEs (Laplante & Spears, 2008). Therefore, an improved understanding of the factors that drive the resolution of conflicts helps both governments and managers to determine their priorities regarding their contributions towards resolving conflicts.

To answer the research questions, this report starts with a review of the literature in the area of indigenous communities, natural resource conflicts, community autonomy, and NGOs. Building on this, chapter 3 presents a theoretical framework, including hypotheses about the concepts introduced in the research questions. Chapter 4 then discusses the research design by presenting an overview of the sample, the variables, and the methods of analysis that are being used. After this, chapter 5 presents the analysis and results, followed by a discussion of the results and their implications in chapter 6, and a conclusion in chapter 7.

(11)

11

2. Literature review

2.1 Indigenous peoples

Even though there is no agreement on the number of indigenous peoples in Latin America, estimations suggest that at least 10% of the total population consists of indigenous groups (Espinoza, 2015; Layton & Patrinos, 2006). This uncertainty in estimations partly results from the varying methods that are used to calculate the numbers, and partly from the lack of a univocal definition of indigenous peoples (Layton & Patrinos, 2006). Both the United Nations and the International Labour Organisation (ILO) stress the importance of the self-identification of indigenous peoples, thereby indicating that no universal definition is necessary (United Nations, 2009). However, for practical purposes, the following definition, as presented in Martínez Cobo’s (1987) Study on the Problem of Discrimination against Indigenous Populations, is widely accepted and cited:

“Indigenous communities, 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 parts 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.” (Martínez Cobo, 1987, as cited in United Nations, 2009, p. 4)

Despite the lack of a clear-cut definition, Bruijn and Whiteman (2010) argue that the indigenous identity is at least composed of environmental and socio-cultural aspects. Indigenous peoples strongly rely on the geographic location that their previous generations have traditionally been living in, since their survival depends on the use of this territory and the resources it contains (O’Faircheallaigh, 2013). Moreover, indigenous peoples often consider themselves as relatives of the land (Whiteman, 2009), which clearly demonstrates the central place the environment has in indigenous cultures. Furthermore, conserving the natural environment is an important norm in many indigenous communities (Bruijn & Whiteman, 2010; Calvano, 2008).

Despite these commonly shared characteristics, there is considerable diversity in indigenous peoples in different countries. In Guatemala, for instance, 60% of the total population is considered indigenous, whilst in Colombia this is only 3.4%. Most countries accommodate a substantial number of distinct indigenous groups: Panama, for instance, holds 7 different indigenous nationalities, and Guatemala is home to over 20 different groups. In Chile, one group, the Mapuche, dominates the indigenous population; it accounts for 84% of the indigenous population. Furthermore, whilst the indigenous peoples in Colombia and Ecuador are

(12)

12 concentrated in the Andes and other rural areas, three quarters of the indigenous peoples in Chile live in urban zones (Jepsen, Jacquelín & Hansen, 2017).

2.2 Natural resource conflicts

Continuously increasing globalisation pressures have multiplied the amount of activities that MNEs perform abroad. Amongst the numerous theories that explain why MNEs cross borders is the influential eclectic paradigm (Dunning, 1988), which integrates country- and firm-level motivations by addressing firm-specific ownership advantages, location-specific advantages, and internalisation advantages. MNEs hence attempt to maximise their benefits by leveraging their ownership advantages, locating activities in beneficial locations, and internalising certain activities when that is more efficient. For MNEs in the extraction industry, the availability of natural resources forms a substantial location-specific advantage. One of the main motives for such MNEs to enter a foreign country is to gain access to natural resources that are not available in their home country, such as minerals, oil, or gas (Dunning, 1993). The increased liberalisation of markets and technological developments have furthermore enabled MNEs to operate in less developed and more remote areas (Rosenfeld Sweeting & Clark, 2000). Since a large part of the world’s natural resources are in territories where indigenous communities live, contact between MNEs and indigenous communities is inevitable, regularly leading to conflict situations (Calvano, 2008).

In the literature on group conflicts, two thoroughly studied components of conflicts are power and communication. Despite the positive developments on indigenous rights, indigenous communities are still lagging behind when it comes to influence in negotiations and decision-making about extraction projects (Calvano, 2008). From a stakeholder theory perspective, the influence certain groups can exert on decision-making depends on their possession of power, legitimacy, and urgency (Mitchell et al., 1997). Hence, when considering indigenous communities as stakeholders of the firm, the differences in influence in a conflict are the result of the historical lack of power of indigenous communities and a lack of perceived legitimacy of their claims by MNEs (Calvano, 2008; Mitchell et al., 1997). Besides, the claims of indigenous communities often have a relatively low priority to MNEs compared to the claims of other stakeholders (Calvano, 2008) and MNEs financial interests (Palacios, 2004). However, as a result of the growing awareness of the necessity of sustainable development, MNEs start to pay more attention to environmental and social issues, which are important to indigenous communities (Hodge, 2014; Lertzman & Vredenburg, 2005).

Another thoroughly studied component of conflict is communication. As a result of the significantly different ideologies of MNEs and indigenous communities, both parties often perceive actions and intentions of their opponent in a different way than their opponent itself.

(13)

13 Indigenous communities may for instance misunderstand the good intentions a MNE has for contributing to the development of the community; the MNE in turn may perceive the community’s response as ungrateful. Moreover, indigenous communities often do not have access to communication channels in the same way as MNEs do (Calvano, 2008). However, as a result of globalisation, indigenous communities around the world are to an increasing extent able to interact and to share knowledge about matters they typically face. Resulting from the growing confidence this gives communities, it increases communities’ capacity to voice themselves and defend their rights (Hodge, 2014).

Indigenous communities tend to deploy their voice capacity when they experience that their identity is being threatened and that they are treated in an unfair manner (Calvano, 2008; Hodge, 2014). Since the environment plays a fundamental role in forming the indigenous identity (Bruijn & Whiteman, 2010), indigenous communities usually perceive any natural resource project on their land as unfair (Whiteman, 2009). However, this is not always the case; MNEs may contribute to the well-being of the community by providing them with developmental support (Devlin & Tubino, 2012). The discrepancies in communication norms nevertheless imply that an investing MNE and the involved indigenous community often have different perceptions of the fairness of the proposed project, which then intensifies conflicts (Hodge, 2014). The overall level of perceived fairness depends on the fairness of interactions, decision-making procedures, and distributive processes related to the project (McCaslin, 2005; Whiteman, 2009). Therefore, important determinants of perceived fairness, and hence for the escalation and endurance of conflicts, are economic, political, social, cultural, ecological, and spiritual factors (Whiteman, 2009). However, there is some disagreement about which factors are the most important. The prevailing explanation for conflicts is a lack of economic and political benefits related to extraction projects (Acuña, 2015). An investigation regarding the concerns of indigenous peoples about mining projects shows that many of those concerns are directly or indirectly related to economic impact or consultation (International Council on Mining and Metals, 2006). However, others emphasise the importance of indigenous culture by arguing that cultural and environmental concerns, rather than economic and political factors, are most likely to intensify conflicts (e.g. Acuña, 2015; Bruijn & Whiteman, 2010).

Conflicts between indigenous communities and MNEs vary along several dimensions, of which severity and duration most often form a basis for comparisons (Getz & Oetzel, 2009; Humphreys, 2005; Pinkey, 2005 in Calvano, 2008). The severity of conflicts ranges from minor disagreement and peaceful actions undertaken by indigenous communities to express their displeasure about extraction activities, to serious violence such as injuries, destruction, and death (Calvano, 2008; Kemp, Owen, Gotzmann & Bond, 2011). Such a high degree of violence may emerge because of the limited resources of indigenous communities. Due to this, they employ

(14)

so-14 called strategies of the weak, which are meant to gain attention by causing disruption (Scott, 1985 in Calvano, 2008). However, the severity of a conflict is not necessarily constant; it may fluctuate over time (Wenger & Mockli, 1999 in Getz & Oetzel, 2009). Also the extent to which human rights are violated may vary across conflicts (Foster, 2012).

Furthermore, conflicts vary in duration (Getz & Oetzel, 2009). Several factors that affect the duration of a conflict are the extent to which parties in the conflict have access to financial means to continue the conflict, and the benefits that parties, both internal and external, may derive from the conflict. Also the degree of economic integration of the involved parties may reduce the duration of conflicts by facilitating the resolution process (Humphreys, 2005). This may become more challenging when a conflict endures; the duration of a conflict complicates the relations between the parties involved in the conflict (Skippari & Pajunen, 2010). Furthermore, the cohesiveness of an involved group influences the duration of a conflict, although there are mixed results regarding the direction in which this takes place (Humphreys, 2005). Such severe and enduring conflicts may threaten the way of life of indigenous communities (Lertzman & Vredenburg, 2005). However, they may also cause financial and reputational damage for MNEs, since the resolution can be extremely challenging, time consuming and costly (Davis & Franks, 2011; Laplante & Spears, 2008). Furthermore, in the worst cases, conflicts have the potential to disrupt national societies (Suliman, 1999).

Traditionally, researchers considered interactions between indigenous communities and MNEs from a conflict perspective. Following this perspective, conflicts are the result of conflicting goals between the involved parties, which are considered as short-term oriented and unwilling to give in to the goals of their opponent. The involved parties hence view each other with suspicion and reputation is not deemed important. The participants’ behaviour is determined by the interdependence of their interests, the quality of their relationship, their perceived gains or losses, and their relative bargaining power; this portrays a way of coping with rather than resolving a conflict (Danskin, Dibrell & Kedia, 2005; Gladwin & Walter, 1980 in Danskin et al., 2005).

More recently, theories of conflict have taken a different turn. Prahalad and Lieberthal (1998) argue that elevated integration levels in the increasingly global economy have changed the mindsets of MNEs and indigenous peoples from confrontational towards cooperative. Due to the increasing influence of large MNEs at the expense of the state, there is a growing consensus that contributing to peace is part of the responsibility of MNEs, thereby changing their behaviour (Matten & Crane, 2005; Scherer & Palazzo, 2008). In turn, communities increasingly consider outsiders as potential sources of resources, support, and change. Lane and Hibbard (2005) emphasise the importance of such a proactive attitude of indigenous communities towards resolving conflicts; this can shape institutional developments and contribute to the resolution of

(15)

15 conflicts. Therefore, Danskin et al. (2005) propose a theory of conflict and cooperation, which takes into account both antagonistic and collaborative interactions. This theory acknowledges the possibility of conflicting objectives between MNEs and indigenous communities. Yet, it emphasises the importance of balancing conflicting objectives and short- versus long-term goals in order to achieve an agreement that takes into account the full picture (Danskin et al., 2005).

In line with this theory, recent studies often mention collaboration, dialogue, and proactive engagement, when addressing strategies for conflict resolution (e.g. Hodge, 2014; Jamali & Mirshak, 2010). These mechanisms can help in resolving conflicts, because they enable the unification of alternative values that are often held by indigenous communities and MNEs (Hodge, 2014). This enhanced stakeholder participation potentially increases the efficiency and equity of the interaction (Castro & Nielsen, 2001). Indigenous Traditional Law also specifies ways to resolve conflicts that are appropriate for their culture and that are focused on collaboration (Whiteman, 2009). This approach is generally more informal than western methods; talking is the main mechanism to find a solution to a conflict and to re-establish a harmonious balance (Pinto, 2000; Whiteman, 2009). Moreover, since trust is an integral element of many indigenous cultures, it is an important condition for relations between indigenous communities and MNEs to be viable (Franks, 2009; Hodge, 2014).

2.3 Community autonomy

Many authors that address indigenous rights argue that autonomy is the main right for indigenous communities (e.g. Daes, 1996; Espinoza, 2015; Wiessner, 2011); because of the intrinsic value and meaning of every culture, “there is an ethical imperative for self-determination” (Lertzman & Vredenburg, 2005, p. 251). Autonomy or self-determination can be defined as “the institutionally guaranteed capacity of a social entity or territory to rule itself within the ambit of sovereignty of a state” (Assies, 2005, p. 159). A country in which indigenous autonomy is implemented in a successful manner is organised pluri-nationally; every ethnic group has equal access to institutions, and treatment by the government is thus based on equality (Chong, 2010). In this manner, governments can play a mediating role in conflicts between indigenous communities and MNEs (Calvano, 2008), yet this role is not always positive due to conflicting interests or a lack of a strong and supportive institutional framework (Getz & Oetzel, 2009).

In recent years, positive developments have taken place regarding the recognition of formal indigenous rights, on both the international and domestic levels. These changes are the result of a growing indigenous rights movement, which has been supported by the United Nations, NGOs, and scholars (Foster, 2012). One of the key developments at international level is the ILO Convention 169 (ILO 169) of 1989, in which indigenous rights are agreed upon. Although

(16)

16 this agreement seems promising, its effects are not as evident as might be expected, because of its limited applicability. The most recent and influential development towards the recognition of indigenous peoples is the United Nations Declaration on the Rights of Indigenous Peoples (UNDRIP) that came into force in 2007 (Bellier & Préaud, 2012; Foster, 2012). Even though this declaration is not binding, it has its implications for national governments that sign it by obliging them to put the necessary mechanisms in place to enforce its content (Foster, 2012; United Nations, 2007).

Partly because of these international developments and corresponding pressures, a growing number of countries are adopting laws for the recognition of indigenous peoples. Although these laws vary per country, most include the protection of the environment and human rights of indigenous peoples. Moreover, most state the right of indigenous peoples to be consulted before the start of a project on their territory, to share in revenues, and to be compensated if their land is taken from them (Foster, 2012), which are all rights that are recorded in the UNDRIP (United Nations, 2007). Moreover, court decisions informed by the increasing standards of indigenous rights are observed (González, 2015). These legal developments reflect an increasingly positive outlook for the reduction of indigenous peoples’ economic and social marginalisation (O’Faircheallaigh, 2013).

Latin America, in particular, could potentially advance the developments regarding indigenous issues worldwide. The involvement of 15 Latin American states in ILO 169, UNDRIP, and other agreements such as the Inter-American Court of Justice is part of the constitutional changes that have taken place in the area (Bellier & Préaud, 2012). The most recent cycle of constitutional change is aimed at the formation of pluri-national states, in which indigenous peoples are accepted and respected as nations within their state. In Bolivia and Ecuador, these developments towards pluri-national states are most advanced (González, 2015; Tockman & Cameron, 2014).

From a multi-level institutional perspective, a community’s degree of autonomy determines its level of embeddedness in a society. Considering the various levels of institutions and the interdependencies between them, as for instance Williamson (2000) and North (1991) distinguish, various degrees of autonomy exist: formal political autonomy, effective political autonomy, and autonomy as a social entity’s freedom to determine its own destiny (Hoekema, 1996 in Assies, 2005). The first aspect, formal political autonomy, concerns the fixation of autonomy in constitutional law. International initiatives such as the UNDRIP and the ILO 169 provide the legal framework for political autonomy for indigenous communities (Assies, 2005; Foster, 2012). However, Foster (2012) notes that this concerns autonomy within the boundaries of the national state, which thus remains the overarching actor.

(17)

17 Nevertheless, despite the positive legal developments, a gap between law and practice is often observed (Bellier & Préaud, 2012; Stavenhagen, 2007). Tockman and Cameron (2014) illustrate this using the case of Bolivia, where indigenous autonomy was incorporated into the constitution in 2009. However, despite additional regulatory developments in this area, indigenous autonomy lags behind due to a lack of sufficient governmental support. This can for instance be explained by Bolivia’s dependence on the extraction industry. Since resource extraction projects stimulate the national economy (Acuña, 2015), the government may prioritise the entrance of MNEs over law enforcement (Bellier & Préaud, 2012; Espinoza, 2015; Foster, 2012). By not encouraging compliance with the law, the country’s institutional framework constrains actual autonomy (Tockman & Cameron, 2014). In such cases, indigenous peoples are seldom sufficiently politically mobilised to profit from the positive developments (O’Faircheallaigh, 2013). This causes reluctance to collaborate and it also possibly increases violence (Helwege, 2015). This situation clearly illustrates the interconnectedness of the different levels of institutions and the need for informal institutions to complement the formal ones for the latter to be effective (North, 1991; Williamson, 2000). Moreover, in order for formal political autonomy to be effective, care needs to be taken of the implementation of the legal initiatives (Foster, 2012).

The second aspect, effective political autonomy, concerns the distribution of actual decision-making power amongst indigenous communities and external authorities (Hoekema, 1996 in Assies, 2005). Effective political autonomy addresses the desire of indigenous communities to participate economically and politically (Acuña, 2015). When this level of autonomy is in place, communities can share in revenues and have a voice in the governance of their territory (Helwege, 2015). To attain this level of autonomy, countries must develop norms to create operational autonomous entities with well-developed policies and legislation that enable their existence (Chong, 2010; González, 2015). Tockman and Cameron (2014) mention a strong state as a requirement for this to have potential. Next to the government, indigenous communities themselves also have the potential to play a role in obtaining effective autonomy by joining forces with one-another (Cameron, 2010). The Bolivian example illustrates that formal political autonomy can exist without being effective (Tockman & Cameron, 2014); not as extensively discussed but also possible is the existence of effective political autonomy without having the legal framework in place (Assies, 2005).

However, in spite of the seemingly positive consequences of effective autonomy, in some cases where indigenous communities have obtained effective autonomy, indigenous authorities are observed to disengage from regional decision-making (González, 2015). This implies an enduring lack of satisfaction about the effect of autonomy. A case study on the Awajun people in the Peruvian Amazon shows that consultation or shared decision-making is not the main goal for

(18)

18 indigenous peoples. Concerns they have about extraction projects reflect fundamental elements of their political ontology. The Awajun considered the potentially affected territory as ecologically sensitive and since it is their ancestral territory, it also has social and cultural importance (Acuña, 2015). In such cases, conflicts may thus arise as a result of a lack of recognition of the indigenous ontologies and a threat to their identity (Acuña, 2015; Bruijn & Whiteman, 2010). Expressions of such disregard are violations of the environment an indigenous community is living in or the neglect of their history and cultural values (Acuña, 2015). In order to prevent these violations from happening, communities unite their forces against the threat, which may lead to violent conflicts with the involved MNE (Bruijn & Whiteman, 2010).

The third aspect, autonomy as a social entity’s freedom to determine its own destiny, is a socio-culturally composed concept which addresses the extent to which the prior aspects effectively promote the self-determination of indigenous communities (Hoekema, 1996 in Assies, 2005). Even though the previous level, effective political autonomy, is often considered when looking at autonomy, some argue that all indigenous communities’ claims to land, resources, and autonomy should be associated with their desire to let their authentic culture survive and flourish (Espinoza, 2015; Acuña, 2015; Wiessner, 2011). This may thus be considered as the highest level of autonomy (Acuña, 2015; Hoekema, 1996 in Assies, 2005).

2.4 Nongovernmental organisations

NGOs are defined as “private, not-for-profit organisations that aim to serve particular societal interests by focusing advocacy and/or operational efforts on social, political and economic goals, including equity, education, health, environmental protection and human rights” (Teegen et al., 2004, p. 466). This definition is an effort to capture the great diversity that exists amongst NGOs (Fisher, 1997). NGOs are often referred to as the third sector, on top of the private and the public sector (Pharr, 2003 in Teegen et al., 2004). Resulting from several global human rights issues, they arose as a means to temper the power of national governments and thereby fight against the violation of human rights (Florini, 2003; Teegen et al., 2004). Moreover, research recognises globalisation as a driving force behind the expanding role of NGOs, because it enables NGOs to roll out their strategies internationally and thus achieve their goals on a global rather than just a local scale (Doh & Teegen, 2003).

From a stakeholder theory perspective, the success of NGOs as stakeholders of the firm can be explained by the power, legitimacy, and urgency these actors possess (Mitchell et al., 1997). Eesley and Lenox (2006) advance Mitchell et al.’s (1997) stakeholder salience model by incorporating secondary stakeholders such as NGOs. They find that, aside from the power, legitimacy, and urgency that a stakeholder possesses, also the legitimacy of the claim determines whether a firm acts accordingly. Since failing to address stakeholder demands may impose both

(19)

19 direct operational costs and reputational damage, stakeholder actions can form strong incentives for a firm to adjust its actions (Eesley & Lenox, 2006). This illustrates the urgency that NGOs possess in such stakeholder relationships (Guay et al., 2004). To exhibit power in firms’ decision-making processes, NGOs deploy a variety of tactics and strategies (Guay et al., 2004). Peterson (1992, in Teegen et al., 2004) classifies these actions as outsider and insider strategies to advocate certain issues. Outsider strategies are aimed at mobilising the public opinion to pressure governments or MNEs towards more sustainable behaviour. Insider strategies focus on possibilities within the institutional framework by directly affecting decision makers.

2.4.1 The pressuring role of NGOs in resource extraction projects

NGOs are increasingly involved in resource extraction projects, in which they advocate civil society interests (Skippari & Pajunen, 2010). They affect both the public and the private sector in situations where both sectors fail to satisfy the needs of society (Teegen et al., 2004). The public legitimacy of NGOs’ activities is a crucial factor for the success of pressure strategies. This is greater in countries where NGOs have already been active for a reasonable amount of time; in many other countries, the public offers little support (Brown & Kalegaonkar, 2002).

When the national government fails to provide the desired services, representation, or social cohesion, NGOs can come into play by deploying outsider strategies (Teegen et al., 2004). According to Stavenhagen (2009), their role is crucial in making governments accountable by facilitating adequate implementation and reducing the violation of agreements and legislation about indigenous rights.

MNEs in the extraction industry are often criticised for neglecting the environmental and social consequences of their actions (Lertzman & Vredenburg, 2005). However, MNEs increasingly invest in engagement with indigenous peoples, for both political and reputational reasons. This is partly the result of the growing influence of NGOs in conflict situations (Jenkins & Yakovleva, 2006). NGOs have the potential to influence MNEs’ behaviour and policies in a way that induces them to take these aspects into account (Palacios, 2004; Skippari & Pajunen, 2010). Failing to respond to large and influential campaigns by NGOs, such as the Mining Campaign of Oxfam, may impose substantial negative effects on the MNE. Therefore, such campaigns form an often-employed means to pressure MNEs towards more sustainable behaviour (Jenkins & Yakovleva, 2006).

2.4.2 The collaborative function of NGOs in resource extraction projects

In spite of NGOs’ potential to pressure parties, a lot of their efforts are not adversarial (Teegen et al., 2004). In addition, adversarial interactions can change from being confrontational to collaborative (Arenas et al., 2013). One direct way to influence decision-making processes from

(20)

20 within is via shareholder activism; NGOs acquire shares of a corporation to stimulate change through ownership (Guay et al., 2004). Furthermore, NGOs are important actors in the negotiation process of international investments, often serving as advisors or experts (Doh & Teegen, 2002). The elevated attention to cross-sectoral collaborations as mechanisms to solve global, societal issues that cannot be addressed by one individual actor increases the recognition of the important role of NGOs (Bäckstrand, 2006; Kolk & Lenfant, 2012). They can play a bridging role between the different parties that are involved in an investment project, such as the national government, MNEs, and local communities (KimDung et al., 2016). By stimulating cooperation, NGOs are a crucial party in successful co-management projects (Castro & Nielsen, 2001; Plummer, 2009). Kolk and Lenfant (2012) however note that the complex and volatile conditions in conflict areas complicate the process of forming and implementing collaborations, because they call for an even more thorough mutual understanding between the partners. This is illustrated by Skippari and Pajunen (2010), who show that conflicts may arise when a MNE fails to respond to a NGO’s demands in a way that satisfies the NGO.

The role of NGOs in facilitating such collaborations between stakeholders may create value for all parties that are involved (Murphy & Arenas, 2011). A collaboration with a NGO can be of great value to a MNE. The growing interest amongst MNEs regarding corporate social responsibility reflects their aim to address society’s concerns about social and environmental matters (Hodge, 2014). Since NGOs have the potential to influence MNEs’ behaviour and policies (Skippari & Pajunen, 2010), they can help MNEs to act in a more responsible manner and to contribute to social development in the host country (Oetzel & Doh, 2009). Furthermore, NGOs’ experience in and knowledge of the local context enables MNEs to learn about the host country (Kolk & Lenfant, 2012). NGOs’ legitimacy and familiarity with local actors may reduce legitimacy issues for MNEs, and on top of that, it can support MNEs in building relationships with indigenous communities (Dahan et al., 2010; Oetzel & Doh, 2009).

Since NGOs dedicate themselves to, amongst others, “alleviating human suffering, […] environmental protection, human rights, and conflict resolution” (Aall, 2000, p. 124), they can also be of great importance to indigenous communities whose rights are violated in conflicts. These communities can utilise NGOs’ resources to compensate for their own lack of power (Calvano, 2008). Furthermore, NGOs’ unique experience can help communities in dealing with conflict situations (Teegen et al., 2004).

Despite the potential positive effects of collaborating with a NGO, legitimacy issues exist regarding NGOs’ intentions due to the increasing collaboration between NGOs and MNEs (Calvano, 2008). Although NGOs, such as the organisations devoted to human rights, often present themselves as acting on behalf of indigenous communities, they do not always act in the best interests of these communities (Fernando, 2003). They may for instance be funded by

(21)

21 corporations that pursue objectives that are not aligned with NGOs’ supposed interests (Murphy & Arenas, 2011). As a result, the general attitude of indigenous communities towards NGOs may become sceptical (Calvano, 2008; Teegen et al., 2004). However, despite the criticism concerning the effectiveness and accountability of NGOs (Calvano, 2007; Goodhand, 2006), they have proven to be powerful in unstable environments (Barnes, 2005).

(22)

22

3. Theoretical framework

Despite the potential benefits of resolving MNE-community conflicts for communities, MNEs, and governments, the drivers of the resolution processes have not yet received substantial attention. First, previous research addresses the dynamics of conflicts in a fragmented way, either focusing on the severity or the duration of conflicts (Calvano, 2008; Fontana, Sastre-Merino & Baca, 2015), without considering them both as key dimensions of the conflict. Second, existing literature that deals with conflict resolution often focuses on ethnic conflicts without MNE involvement, thereby overlooking an important actor in many conflicts (e.g. Coakley, 1994; Cunningham et al., 2009). Third, previous studies on MNE-community conflicts often limit themselves to exploring one single driver behind the resolution of the conflict (e.g. KimDung et al., 2016; Yiaziji & Doh, 2009); they hence fail to identify interactive effects of several drivers. Finally, several studies explore the drivers of conflict resolution in isolated cases of conflict (e.g. Calvano, 2008; Fontana et al., 2015; Kolk & Lenfant, 2012); they therefore fail to reveal relationships that hold in multiple settings and thereby they do not offer global insights.

Since the absence of violence and human rights violations, and the shortening of conflicts correspond to a decrease in the disastrous effects of conflicts (Getz & Oetzel, 2009), this study addresses the extent to which certain actors drive these factors as a proxy for the resolution of conflicts across a variety of countries. Research suggests that conflicts between indigenous communities and MNEs arise out of conflicting goals between the two parties and perceived unfairness regarding the extraction project (Calvano, 2008; Danskin et al., 2005; Hodge, 2014). A key characteristic of conflicts is the lack of influence of indigenous communities as a stakeholder of the firm in conflict situations, resulting from a lack of power, legitimacy, and urgency to the MNE (Calvano, 2008; Mitchell et al., 1997). This leads to unbalanced, and often violent and enduring, antagonistic interactions between the involved parties (Calvano, 2008). Collaboration and dialogue are considered as the main mechanisms that contribute to the resolution of conflicts (e.g. Danskin et al., 2005; Hodge, 2014; Jamali & Mirshak, 2010).

National governments have traditionally been considered as important actors in controlling the economy and society (Peters & Pierre, 1998). In this respect, the resolution of conflicts between indigenous communities and MNEs is one of their responsibilities. Governments can empower indigenous communities in interactions with other stakeholders, thereby stimulating dialogue and collaboration, and hence the resolution of conflict (Danskin et al., 2005; Foster, 2012). However, there is increasing attention for dispersed authority and shared multi-level governance (Hooghe & Marks, 2003; Peters & Pierre, 1998). Consequently, the importance of NGOs in the resolution of conflicts is increasing (Bellier & Préaud, 2012; Calvano, 2008). Given the strong legitimacy of NGOs, they can positively affect the resolution of conflicts by strengthening communities’ power, legitimacy, and urgency relative to other stakeholders

(23)

23 (Eesley & Lenox, 2006). Following this line of thought, the following subsections evaluate the extent to which community autonomy and NGO involvement may be able to contribute to the resolution of conflicts.

3.1 Community autonomy and conflict resolution

When considering community autonomy from a multi-level institutional perspective, three levels can be distinguished: formal political autonomy, effective political autonomy, and autonomy as a social entity’s freedom to determine its own destiny (Hoekema, 1996 in Assies, 2005). Given the importance of informal institutions to complement the formal ones (North, 1991; Foster, 2012), only the latter two levels have the direct potential to affect conflict situations. Effective political autonomy, the second level, addresses the desire of indigenous communities to participate economically and politically by enabling them to share in revenues that may be generated and to have a voice in the governance of their territory (Acuña, 2015; Helwege, 2015). From this point of view, violence often arises as a result of the neglect of indigenous communities by the national government in decision-making processes regarding the start of extraction projects or the distribution of their revenues (Helwege, 2015). The third level of autonomy takes a socio-cultural approach by focusing on communities’ desire to preserve their authentic culture (Acuña, 2015; Espinoza, 2015). This implies that autonomy has the potential to increase the legitimacy of indigenous’ human rights protection (González, 2015). From this point of view, violent conflicts arise as a result of a lack of recognition of indigenous ontologies and a threat to their identity (Acuña, 2015; Bruijn & Whiteman, 2010).

Major sources of a lack of perceived fairness, and thus of severe conflicts, are a lack of power of the indigenous community and endangerment of the indigenous identity (Whiteman, 2009). By granting indigenous communities autonomy, national governments can enable them to make decisions about their own futures (Bruijn & Whiteman, 2009; Foster, 2012). Thereby, autonomy increases the community’s power as a stakeholder in decision-making processes related to extraction projects (Mitchell et al., 1997). Subsequently, the resulting balance in power dynamics facilitates communication between indigenous communities and MNEs (Calvano, 2008). Community autonomy hence enables proactive engagement and collaboration with the MNE. Autonomous communities may feel more respected, and more open and empowered to participate in peaceful negotiations about extraction projects, which results in a reduced level of violence and a reduction of the violation of their rights (Helwege, 2015).

The potential effect of community autonomy on conflict severity is illustrated by a case study by Whiteman and Mamen (2002) on the activities of the Canadian mining company Tiomin in the copper mine Cerro Colorado in Panama. Since its first exploitation, the Cerro Colorado mine has known continuous conflict resulting from the neglect of the Ngäbe-Buglé people, whose

(24)

24 territory the mine is located in. In 1997, the Panamanian government formally recognised the indigenous community on the condition that the mining operations could continue. During negotiations between Tiomin and the Panamanian government about a license for the mine, the government represented the interests of the Ngäbe-Buglé people. The resulting contract required Tiomin to contribute to the development of the community by investing in a social programme. This achievement resulted in a reduction of the perceived unfairness by the Ngäbe-Buglé people, which in turn reduced the need to employ violent means regarding social issues. It must be noted that, even though this recognition had some positive effects, the broader picture still showed substantial neglect, since recognition lacked in other areas. Furthermore, in a renewal of the contract, Tiomin eliminated the social programme and thus its positive results. However, the example illustrates the potential effect of recognition by the government and moreover, it emphasises the importance of ensuring the effectiveness of this recognition. Therefore, I propose the following hypothesis:

H1: Community autonomy negatively influences the severity of conflicts between indigenous communities and MNEs.

In a similar way in which community autonomy can contribute to the reduction of the severity of a conflict, as explained in detail on pages 23 and 24, it may also affect the duration of a conflict. Considering effective political autonomy, community autonomy provides communities with the ability to rule themselves and participate in decision-making processes (Assies, 2005; Hoekema, 1996 in Assies, 2005). It may hence increase perceived fairness by empowering communities in decision-making processes and enhancing dialogue. Even though this concerns autonomy within the boundaries of the national state (Foster, 2012), it balances the power dynamics between indigenous communities and the other stakeholders in the conflict. Furthermore, increased autonomy and hence enhanced dialogue and collaboration between the indigenous community and the other parties involved in a conflict may simplify and improve the relations between the different parties. Since power differences may enable certain parties in a conflict to benefit from the endurance of the conflict (Humphreys, 2005), and complicated relationships between the parties involved in the conflict may increase the duration of conflict (Skippari & Pajunen, 2010), increased autonomy leads to shorter conflicts.

The effect that recognition by the government and community autonomy can have on conflict duration is illustrated by a case study by Devlin and Tubino (2012) on the activities of the Brazilian mining company Vale in Itabira in Brazil. For several decades, the people living in this area have been expressing their concerns regarding the exploitation of the mines, for instance by community protest marches. Changes were realised when in 1994, the government required Vale to obtain an environmental licence for the exploitation of the mines. This requirement led to public meetings, in which various involved parties expressed their concerns and negotiated about

(25)

25 the conditions of the operating license. The resulting license, which was approved in 2000, obligated Vale to do several concessions to the affected communities, fixed into 52 conditions. The communities experienced these conditions to be suitable reflections of their concerns. By enforcing Vale to do concessions in order to obtain a license, recognition by the government led to the end of the conflict in Itabira. Therefore, I propose the following hypothesis:

H2: Community autonomy negatively influences the duration of conflicts between indigenous communities and MNEs.

3.2 NGO involvement and conflict resolution

NGOs use their power, legitimacy, and urgency to deploy both outsider and insider strategies to advocate particular issues (Mitchell et al., 1997; Peterson, 1992 in Teegen et al., 2004). Regarding outsider strategies, they have the potential to pressure both governments and MNEs towards more sustainable behaviour (Teegen et al., 2004). Insider strategies build on NGOs’ extensive knowledge of and experience in the local context. This knowledge makes them valuable partners for MNEs, national governments, and indigenous communities (Dahan et al., 2010; Oetzel & Doh, 2009). They can fulfil a bridging role between the different actors involved in a resource extraction project (KimDung et al., 2016). This potentially enhances collaboration and increases the long-term sustainability of such projects (Doh & Teegen, 2002; KimDung et al., 2016). Moreover, collaboration and co-management are recognised as important conflict resolution mechanisms (Hodge, 2014). Collaborative approaches to resource management have the potential to unify alternative values held by communities and MNEs and to increase the equality of the interaction (Castro & Nielsen, 2001; Hodge, 2014).

Considering the power, legitimacy, and urgency that NGOs can deploy to influence national governments’ and MNEs’ decisions about for instance indigenous and human rights issues (Mitchell et al., 1997; Stavenhagen, 2009; Teegen et al., 2004), they can play a determining role in pressuring both governments and MNEs towards more sustainable behaviour by creating awareness on these topics. Furthermore, NGOs facilitate collaboration between the involved parties, building on the knowledge they have about each of them (Dahan et al., 2010; Oetzel & Doh, 2009). They hence reduce communication problems and enhance the power of indigenous communities in conflict situations. By stimulating collaboration, NGOs can change actual circumstances and hence serve as institutional actors (Doh & Teegen, 2002). Thereby they can help indigenous communities in obtaining actual autonomy by creating conditions in which formal political autonomy leads to effective political autonomy. The need for indigenous communities to use violence (Calvano, 2008) hence diminishes as their ability to voice themselves increases. In this manner, NGO involvement thus strengthens the negative effect of community autonomy on conflict severity by increasing its effectiveness.

(26)

26 The importance of NGO involvement is illustrated by a case study by Arenas et al. (2013) on the Spanish oil and gas company Repsol and its activities in Bolivia. Despite the positive developments regarding indigenous rights in both national and international law, Repsol’s activities regularly resulted in violations of these rights. The Spanish NGO Intermon Oxfam employed both pressuring and collaborative strategies to influence Repsol’s policies regarding indigenous peoples. In 2007, Intermon Oxfam mobilised the public opinion by launching a campaign to reveal Repsol’s attitude towards affected indigenous communities. In 2008, the NGO directly advised Repsol to implement policies aligned to international standards such as ILO 169 and UNDRIP. Those actions led Repsol to adjust its conduct and adopt the proposed policies, which in turn reduced violations. Thus, the effect of the involvement of Intermon Oxfam demonstrates the potential of NGOs to increase the effectiveness of indigenous rights, such as autonomy, and thereby its potential to reduce the severity of violations. Therefore, I propose the following hypothesis:

H3: NGO involvement negatively moderates the relationship between community autonomy and the severity of conflicts between indigenous communities and MNEs.

In a similar way in which NGO involvement can contribute to the reduction of the severity of a conflict, as explained in detail on pages 25 and 26, it may also affect the duration of a conflict. NGOs’ power, legitimacy, and urgency (Mitchell et al., 1997) have the potential to effectively empower indigenous communities in interactions with MNEs. NGOs influence both governments and MNEs in situations where both fail to meet the needs of society (Teegen et al., 2004). By mobilising the public opinion, NGOs likely stimulate governments and MNEs to act in accordance with the NGOs’ requests, because failing to react to such requests may entail substantial costs. Thereby, the involvement of a NGO in a conflict creates a sense of urgency to respond and resolve the conflict quickly (Eesley & Lenox, 2006). Furthermore, the bridging role that NGOs can play in resource extraction projects enables them to connect the different actors involved (KimDung et al., 2016). Since NGO involvement stimulates collaboration and thereby enhances the efficacy of autonomy, it increases peaceful interaction between the parties involved in the conflict. This may improve the mutual relations and integration of the parties involved, which has the potential to reduce the duration of conflicts (Humphreys, 2005). In this manner, NGO involvement thus strengthens the negative effect of community autonomy on conflict duration by increasing its effectiveness.

This effect is illustrated by a case study by Murphy and Arenas (2011) on the natural resource company BHP Billiton in Peru. In a conflict about the activities of BHP Billiton in the Tintaya Copper Mine, seven Peruvian communities collaborated with both national and international NGOs, under which Oxfam Novib. This collaboration strengthened the position of the communities in negotiations about the mine’s destination, and it thus enhanced their effective

(27)

27 political autonomy. Furthermore, Oxfam facilitated a so-called dialogue table, which enabled peaceful negotiations between BHP Billiton and the indigenous communities. This resulted in an agreement in 2004, which ended the conflict. Thus, the involvement of Oxfam Novib enabled the involved communities to enjoy effective political autonomy, and it eventually led to the resolution of the conflict. Therefore, I propose the following hypothesis:

H4: NGO involvement negatively moderates the relationship between community autonomy and the duration of conflicts between indigenous communities and MNEs.

3.3 Conceptual model

Taking this all together, Figure 1 shows the hypothesised relationships between the concepts as discussed above. Community autonomy is expected to have a negative effect on both the severity and the duration of a conflict. Furthermore, these relationships are expected to be negatively moderated by NGO involvement.

Figure 1 Hypothesised relations between community autonomy, NGO involvement, conflict severity, and

conflict duration Community autonomy NGO involvement NGO involvement Conflict severity Conflict duration - - - - H3 H4 H1 H2

(28)

28

4. Data and methods

4.1 Sample and data collection

The used sample contains 221 cases of conflict between indigenous communities and MNEs from 21 countries in Latin America: Argentina, Belize, Bolivia, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, El Salvador, Guatemala, Guyana, Honduras, Jamaica, Mexico, Panama, Paraguay, Peru, Suriname, Uruguay, and Venezuela. Appendix A shows the distribution of cases across these countries. Latin America is an appropriate region to analyse, because the region is endowed with a lot of natural resources and hence conflicts are an important issue (Foster, 2012). Additionally, recent constitutional reforms in Latin American countries led to considerable achievements regarding the recognition of indigenous communities (González, 2015; Van Cott, 2000). These reforms are opening new possibilities for community autonomy to advance indigenous rights and thereby they potentially contribute to the reduction of the severity and duration of conflicts (González, 2015). Furthermore, the MNEs in the sample are all active in the resource extraction sector. Since this sector is characterised by numerous and severe conflicts which are dispersed over various countries (Lertzman & Vredenburg, 2005), this sample provides a fruitful base to investigate conflict dynamics.

The data used in this study is secondary data; it is collected from third parties that provide databases and reports on indigenous communities and conflicts, such as the Environmental Justice Atlas, Environmental Justice Organisations, Liabilities, and Trade (EJOLT), Intercontinental Cry, Minewatch, and Conflictos Mineros. When necessary, additional data is gathered from other sources, such as newspaper articles or country indices. All gathered data was coded based on a coding manual to ensure completeness and comparability of the information on each of the needed variables. The collected dataset consists of cross-sectional data; this means that it does not contain more than one observation per variable for each conflict.

4.2 Variables

4.2.1 Dependent variables

This study uses three dependent variables to operationalise the concept conflict resolution, namely: the level of violence in a conflict, the number of human rights violated in a conflict, and the duration of the conflict. In this operationalisation, the level of violence and the number of human rights violated are both considered as measures for the severity of a conflict. Whilst the level of violence is a more commonly used indicator for the severity of conflicts (Getz & Oetzel, 2009), attention to the violation of human rights is growing due to the increased awareness resulting from globalisation (Foster, 2012).

The roles of governments and NGOs in the resolution of conflicts between indigenous communities and MNEs