Indigenous communities, MNEs and co-management : the influence of development and corruption in conflict resolution

Indigenous communities, MNEs and co-management

the influence of development and corruption in conflict resolution

Master Thesis – MSc Business Administration: International Management

Name: Josefien (J.J.) de Kool

Student nr.: 11956399

E-mail:

josefiendekool@hotmail.com

Supervisor: dr. Ilir Haxhi

Second reader: dr. Markus Paukku

Institution: Amsterdam Business School, University of Amsterdam

Project: Indigenous Communities, National Governance and MNEs

Date: June 22

_{, 2018}

Statement of originality

This document is written by Josefien de Kool 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.

Abstract

An increasing number of conflicts between indigenous communities and multinational enterprises (MNEs) have stirred academic and public debate on this relationship, because they often result in violence. Previous research explores the factors that influence the resolution of these conflicts: however, the single case study approach these studies take, has impaired our full understanding of these conflicts. In this study, we argue that specifically the cross-country institutional context affects these conflicts at a global level, because they determine the ability of national governments to react in these conflicts and therefore, how long and dense these conflicts are. To analyse institutional factors, we examine how three country characteristics - the level of development, the level of corruption and the rule of law – influence the violence density of the conflicts between indigenous communities and MNEs and the duration of its resolution. In addition, this study will illustrate of the dynamics of conflict resolution and the involvement of civil society organizations (CSOs) by taking a co-management approach as its moderating effect. This will clarify more than just the density and length of these conflicts, but also how these conflicts can be resolved. We test our predictions for a sample of 514 cases of conflict from 58 countries. Our results show three main findings. First, we find that conflict length and degree of violence are related. Second, we find that while the level of development has a positive influence on length and violence; the level of corruption and the rule of law have a negative influence. Third, we find that CSO involvement has a partial moderating effect on the level of development and the level of corruption relating to the degree of violence and the length of the conflict, but no clear effect on the rule of law concerning the same dependents. The results have both theoretical and practical implications. Theoretically, this study shows a trend adding to previous research by illustrating the influence of multiple country-level factors and civil society organizations on multiple conflicts between multinationals and indigenous communities. Practically, this study also has implications for MNEs, indigenous communities, governments and third parties involved in the conflict, because it indicates how the different parties can influence the dynamics of conflict resolution.

Keywords:

Table of Content

Abstract 3

List of tables and figures 5

Introduction 6

Literature review 11

Indigenous communities 11

Conflicts between indigenous communities and MNEs 14

National government and characteristics 16

Third-party involvement 19

Theoretical Framework 23

The dynamics of conflict resolution 23

Country characteristics and their influence on conflicts 24

Co-management and conflict resolution 27

Conceptual Framework 30

Research Design 31

Data and sample 31

Dependent variable 31 Independent variable 32 Moderating variable 33 Control variables 34 Methods 35 Results 38

Descriptive statistics and correlation analyses 38

Regression analyses 42 Discussion 52 Findings 52 Theoretical implications 54 Practical implications 56 Limitations 57 Future research 58 Conclusion 60 References 62

List of Tables and Figures

Table 1. Logistic regression analysis with dependent variable ‘length of conflict’ Table 2. Linear regression analysis with dependent variable ‘degree of violence’ Table 3. Descriptive statistics and correlation between the different variables Table 4. Multicollinearity coefficients

Table 5. Logistic regression analysis for the dependent variable ‘length of conflict’

Table 6. The moderating effect between the independent variables and the dependent variable ‘length of conflict’

Table 7. Linear regression analysis for the dependent variable ‘degree of violence’

Table 8. The moderating effect between the independent variables and the dependent variable ‘degree of violence’

Introduction

The phenomenon of globalization has influenced every aspect of economic integration. Multinational enterprises (MNEs) have more opportunities and other motives to expand their business activities abroad (Calvano, 2008; Pauly & Reich, 1997; Rugman, Verbeke & Nguyen, 2011). Several institutional factors determine the ability, the willingness and the power of a government to protect the local and indigenous communities affected by the practices an MNE undertakes in the host country. The tensions between national governments, indigenous communities and MNEs are most visible within the extractive industries, such as oil extraction and mining (Banerjee, 2008; Yamin & Sinkovics, 2009). Consequently, all three parties find themselves in difficult positions. Especially governments can experience difficulty in protecting their citizens. On the one hand, governments feel the need for foreign investment to increase economic activities and development (Yamin & Sinkovics, 2009, p. 146), while on the other hand; they are bound to protect the interests of their citizens and more specifically, the indigenous communities that find themselves in an even more difficult position (Banerjee, 2008).

Due to their great diversity and ethical incentives, these communities are hard to define (Lertzman & Vredenburg, 2005). Their definition is not always agreed upon, but what is clear is that their relationships with the national governments and MNEs are tense. Research states that MNEs are nowadays predominantly driven by the search for resources and strategic assets (Dunning, 1998; Yamin & Sinkovics, 2009, p. 145), which are still intact in the territories of indigenous communities. Therefore, MNE practices often result in intensified relationships and conflicts especially with indigenous communities (Akee & Jorgensen, 2014; Crawley & Sinclair, 2003; Lertzman & Vredenburg, 2005; Oltremari & Jackson, 2006). One feature about these communities that is clear overall, is that these communities often have their own traditional ways of pursuing economic, social, political and ecological practices. Consequently, indigenous communities are often treated as a separate entity by the nation-state, making it harder for the national government to protect them (Lertzman & Vredenburg, 2005, p. 239). Ever since the International Labour Organization (ILO) first defined these communities in 1989 (ILO), awareness has grown to protect these communities against foreign investments that result in extraction. Hence, these conflicts noticeably turn into violence (McCartin, 2013).

Research has investigated the relationship between indigenous communities and MNEs mainly through single case study approaches (Lertzman & Vredenburg, 2005, p. 247). Consequently, extensive comprehensive overviews on multiple cases are lacking. Moreover, the studies that do include multiple cases do not acknowledge the importance of the national framework in these conflicts, whereas The

United Nations Declaration for the Rights of Indigenous Peoples clearly states that the role of national

governments is tremendous (UNDRIP, 2008, art 1, 19 and 26). These governments have a major influence on the characteristics shaping a country. These features can determine the different political, social, legal and economic standings of the indigenous communities within said country. In addition, country-specific features are inevitably linked to the opportunities MNEs have to invest abroad (Mauro, 1995; Murphy & Arenas, 2010; Oetzel & Doh, 2009; Rodriguez, Uhlenbruck, & Eden, 2005).

Two arguments guide this study. First, the fact that country characteristics affect the conflicts between indigenous communities and MNEs. Previous research has focused on a single country’s effects on the level of development (Akee & Jorgensen, 2014; Anderson, Dana, & Dana, 2006; Cooke et al., 2007; Parlee, 2015; Pedersen, 2014) and the level of corruption (Cross, 2016; Rodriguez et al., 2005; Sethi, Lowry, Veral, Shapiro, & Emelianova, 2011). These two country-level characteristics are inevitably connected through the strength of institutions that shape a country, which is named the rule of law (Yakovleva & Vazquez-Brust, 2018). These three variables are crucial in determining the position of indigenous communities in a country. When their position is not valued or recognized, the investing MNEs will probably experience an even larger liability of foreignness than they do when investing in other areas without such cultural and political differences (Lertzman & Vredenburg, 2005). Previous research has treated most of the country-level factors such as the level of development and corruption in isolation (Blackburn, 2012; Mauro, 1995). Therefore, a combination of level of development, the level of corruption and the strength of institutions enables this study to draw conclusions on the political and economic side of these conflicts to possibly resolve them before escalation. Taking the previous into account, the first research question of this study is defined as:

RQ1: To what extent is conflict resolution between MNEs and indigenous communities influenced by three country characteristics, level of development, control of corruption and rule of law?

Second, countries with a low level of development, a high control of corruption and a low rule of law are usually the subject for civil society organizations (CSOs) (Arenas, Sanchez & Murphy, 2013; Dia, 1995). CSOs see opportunities to protect the economic and social interests of the inhabitants of a country (Cohn, 2014, p. 201). They usually propose other views and sources of advice to the conflicts between MNEs and indigenous communities and are more involved in them (Lertzman & Vredenburg, 2005; Millar, Choi & Chen, 2004; Oetzel & Doh, 2009; Perdersen & Pedersen, 2013; Sethi et al., 2011; McCartin, 2013; Sarfaty, 2005; Spar & La Mure, 2008). Research on indigenous communities and MNE conflicts has emphasized the role of non-governmental organizations (NGOs) as a subject of co-management (Millar, Choi, & Chen, 2004; Sethi et al., 2011; Tedmanson, 2008). This study addresses a much broader group: civil society organizations. This concept provides a much broader notion covering all sorts of parties that are involved in conflict resolution. It reaches further than only measuring the involvement of NGOs; it also takes into account other parties that can be involved in conflict resolutions, such as: international governmental organizations (IGOs), financial organizations, labour unions and the media (Oetzel & Doh, 2009; Perdersen & Pedersen, 2013). These groups are usually more equipped to involve themselves in social causes and conflict resolution, because due to their short lines of communication they are better in maintaining sustainable relationships with multiple parties. Civil society groups can enhance the position of these communities in the face of exploitation and secure their rights and livelihood with more efficiency (Poirier & Ostergren, 2002).

Because of these ideas, this study addresses the nature of co-management. Across decades, third party actors have influenced policy-making and all sorts of factors involved in the institutional environment. First, co-management regarding the effects of (under)development has existed for a longer time period (Acheson, 2013; Kolk & Lenfant, 2013; Spar & La Mure, 2008). Least-developed countries (LDCs) experience a higher engagement of civil society groups than developed countries. Second, it is known that conflicts can be lengthier and more dense if a country has a corrupt government and weak institutions (Kolk & Lenfant, 2013; Owoye & Bissessar, 2012).Therefore, the second research question (RQ2) of this study is defined as:

RQ2: To what extent are the proposed relationships influenced by involvement of civil society organizations?

The purpose of this research is to gain a more thorough understanding of the country-level indicators shaping conflict resolution between indigenous communities and MNEs. Most of the previous research focused on either several socio-economic factors - such as economic development (Cooke, Mitrou, Lawrence, Guimond & Beavon, 2007) – or several political-institutional factors – such as government effectiveness or corruption (Dia, 1995; Cross, 2016). This research focusses on several of these country characteristics in order to determine the effect of these different indicators. Moreover, the study also adds to literature on co-management by using CSOs as the moderating effect to gain understanding about the possible roles of CSOs in conflict resolution. Most of the previous studies have addressed these conflicts by using a single country or conflict. A quantitative approach can secure more universal results and can also help MNEs, governments, indigenous communities and other involved parties to acknowledge specific factors that influence conflict resolution.

To answer these research questions, we run multiple analyses for a sample of 514 conflict cases spread across 58 countries. Expected is that each of the three independent variables - i.e., level of development, control of corruption and rule of law - will influence the duration and degree of violence of a conflict and that CSOs will moderate these relationships. The relationship between development and conflict resolution is expected to be negative, because the higher the development of a country, the better the economic position of the indigenous community, the shorter and less dense a conflict will be. The relationship between control of corruption and conflict resolution is expected to be positive, stating that the higher the control of corruption in a country is, the poorer the political position of the indigenous community, the longer and denser a conflict will be. For rule of law, there is expected to be a positive relationship as well. This means that conflicts in countries with a low trust in institutions – for example in law enforcement – will be longer and denser. In addition, this study tests the moderating effect of CSOs on the aforementioned relationships, because low development, high control of corruption and high rule of law will lead to denser and lengthier conflicts: a subject for CSOs to be involved in (Arenas, Sanchez & Murphy, 2013; Dia, 1995).

This study has some theoretical and practical implications. Methodological, this study will use a multiple-case-study approach in order to generate cross-national results and findings that can enhance previous research on country-level indicators within conflicts between indigenous communities and

MNEs. This approach contributes to the literature because common factors that are underlying conflict resolution in these conflicts can be discovered. In addition, knowledge on conflict resolution can be expanded by the addition of co-management by CSOs. This means that the findings will also add to the debate on third party involvement and civil society interaction within private-public cooperation. Practically, the results will also provide information for indigenous communities, MNEs, national governments, and third parties in resolving and preventing these conflicts. In the future, this could contribute to MNEs investment decisions, collaborations with third parties, and an improved position for indigenous communities.

This study progresses as follows. First, we will review previous studies on the topic of conflict resolution between indigenous communities and MNEs, which will illustrate the need for this study. The next chapter will provide a theoretical framework, where the used concepts in this research will be defined. The theoretical framework will also propose hypotheses guiding this investigation and conclude with a conceptual framework. After that, the data collection, sampling, variables and research method will be described. Thereafter, the results of the different regression analyses will be presented and the final chapter presents a conclusion.

Literature review

Literature has given more attention to international expansion of business activities around the world (Calvano, 2008; Dunning, 1998; Pauly & Reich, 1997; Rugman et al., 2011). This research expanded to include research into the activities of MNEs and their involvement in more rural and undiscovered areas such as the territories o indigenous communities. This involvement resulted in studies of multiple indigenous community – MNE conflicts (Castro & Nielsen, 2001; Lertzman & Vredenburg, 2005; McCartin, 2013; Persson, Harnesk & Islar, 2017).

This chapter gives an overview of current literature on the topic of country-level indicators and CSO involvement on conflicts between indigenous communities and MNEs. First, the chapter will define indigenous communities and the dynamic position they possess in countries. Second, literature on the dynamics of the conflicts between these communities and MNEs is discussed. Last, we discuss the literature on country-specific indicators and the involvement of co-management actors in these conflicts.

Indigenous communities

Far-reaching characteristics of indigenous communities have challenged researchers’ ability to define these communities in a common way (Whiteman & Cooper, 2006). The first attempt to define indigenous communities was during the ILO convention C169 (1989). According to ILO, the need for this definition was that indigenous communities experienced discriminative practices. Therefore, an agreement on the definition of these communities was necessary. The definition stated during this convention enlightens three main aspects. First, the definition highlights a socio-economic part: ‘Tribal

peoples in independent countries whose social, cultural and economic conditions distinguish them from the national community’. Second, a legal and political part: ‘whose status is regulated by own customs or traditions’. Third, a historical perspective: ‘from the populations which inhabited the country at the time of colonialization’ (ILO, 1989).

In 2008, the United Nations General Assembly (UNGA) adopted the ‘Declaration on the Rights of Indigenous Peoples’ (UNDRIP, 2008), where the UNGA states that indigenous peoples suffer from unequal representation and mistreatment. The declaration emphasizes the urgent need for states to

respect and secure the rights of these indigenous peoples. In addition, the literature states that UNDRIP poses another difficulty: the declaration is not legally binding, as opposed to the ILO C169 (ILO, 1989) (Kingsbury, 1998). Hence, the declaration does not create new rights, but its purpose was mainly to create a comprehensive overview on the rights of indigenous peoples within international law. However, this also implies that there will be no legal consequences when a country does not meet the regulations described by the declaration (Foster, 2012; Misiedjan & Gupta, 2014, p. 84).

After the ratification of UNDRIP, many states and regional organisations attempted to adjust national law to imply different articles of UNDRIP (CIGI, 2014; ILO & ACHPR, 2009). The rights of these indigenous peoples are described as: ‘’indigenous people, just as every other person, enjoy the

right of self-determination’’ (Thornberry, 2002). However, research shows that several indigenous

communities do not feel represented or do not experience the required protection by their home state yet (Sierra, 2005). Researchers believe that this is due to difficulties with defining these communities. What is interesting is that the difference between ‘peoples’, ‘people’, ‘community’ or ‘tribes’ is not inevitably clear. Thornberry (2002) describes this first difference as ‘the question of coherence’. To illustrate, special rapporteur Alfonso Martinez (Martinez, 1999) argued that indigenous rights are often confused with minority rights.

Kingsbury (1998) was one of the first researchers to develop a comprehensive overview of the different definitions of indigenous communities. The definition distinguished several essential and relevant indicators. The essential indicators consisted of: first, the self-identification as an ethnic group; second, the historical experience of disruption, dislocation or exploitation; third, the inherent connection with the region and finally, the wish to retain a distinct identity. The less-important, but still relevant, indicators consisted of the non-dominance in the national society and the cultural affinity with a particular territory (Kingsbury, 1998, p. 455; Sanders, 1999, p. 11). The definition specified by another supranational institution - the World Bank - clearly states that the previous definitions lack one aspect to appropriately define indigenous peoples: vulnerability. Indigenous communities do not belong to the major social groups within the nation’s society, which makes them vulnerable to negative practices (Thornberry, 2002).

In the early 2000’s, Thornberry (2002) defines four strands to describe these communities. First, these communities are associated - usually for a long period - with a particular place. Second, the term ‘indigenous’ refers to prior inhabitation - the people inhabited the country before colonialization. Third, the term describes the first, and particularly the original, inhabitants of the country. Last, indigenous communities usually have distinctive cultural, political and economic standards and can therefore be seen as unique societies. These definitions align with the definition stated by the ILO, suggesting a similar approach.

More recent studies on the definition of indigenous communities show several applications of the characteristics as defined by the ILO and the UN before. For example, Lertzman and Vredenburg (2005, p. 244) define indigenous communities as communities that maintain traditional knowledge and experience, linking current practices to its ancient origins. They have distinctive languages, cultures, economic systems, customs law and political sovereignty. In addition, the researchers suggest that the communities are often bio-linguistic and live closely with nature. Hence, Lertzman and Vredenburg (2005) state that indigenous communities can have a crucial meaning to sustainable development. Another example by Crawley and Sinclair (2003) illustrates that indigenous groups are easily defined by their living conditions, thus by naming distinctive statistics of unemployment, education level, criminal activity and health. They state that indigenous communities, in Australia, are often underdeveloped and educated poorly. Other researchers conclude the same in their specific cases (Anderson et al., 2006; Cooke et al., 2007; Garcia-Moreno & Patrinos, 2011; Pedersen, 2014).

In addition to the definitions evolving over time, the research method of single-cases has dominated research into these communities. Three conclusions can be of importance to describe trends in research into these conflicts. First, Sierra (2005) states that in Mexico, the indigenous community experiences a lack in education, as they give most attention to spirituality and ethical and ecological incentives. This illustrates the attention these communities give to living with the environment. Second, Banerjee (2008, p. 70) illustrates that in the case of the Chiapas in Mexico, indigenous lands are now used to grow cattle feed for fast food markets in the USA instead of providing for consumption for the Chiapas, which clashes with the view as proposed by Sierra (2005). Lastly, a study into Canada by Parlee (2015) shows that Canadian indigenous communities are characterized by poor economic growth

and social capital, and therefore he proposes different solutions for governments and companies to improve the living conditions of these communities.

Several studies have combined multiple cases instead of single cases to generate conclusions on the position of indigenous communities. Hall and Patrinos (2004) describe certain developments that take place regarding the position of indigenous communities in Latin-America. They state that indigenous political influence has increased since the 1990s. Additionally, the gap between indigenous peoples’ and non-indigenous peoples’ income is extending and being indigenous increases someone’s probability of being poor (Hall & Patrinos, 2004, pp. 3-4). Another study investigating multiple cases is the study by Cooke et al (2007). This study connects the Human Development Index (HDI) to indigenous well-being. Through the analysis of four cases, Canada, the United States, Australia and New Zealand, Cooke et al (2007) find that indigenous people have much poorer health and social conditions than non-indigenous people. However, they state that education, income, and life expectancy in these communities are improving since the awareness for their position has risen. Hence, these findings cannot be applied for other indigenous communities, because it is a very rough assessment (Cooke et al., 2007). Lastly, an investigation in Russia, Australia and the United States by Poirier and Ostergren (2002) describes the protection and importance of natural resources by these communities. Even though the research covers only three different cases, they suggest that every indigenous community puts a lot of effort into maintaining their natural resources (Poirier & Ostergren, 2002). These illustrations show that several of these factors are commodities.

Conflicts between indigenous communities and MNEs

Different cases in previous research enlighten the tense relationship between MNEs and indigenous communities. MNEs are seen as the bringers of global integration and they are increasingly investing in territories that are not closely related to business hubs or the capital in a foreign country (Pauly & Reich, 1997; Cohn, 2014). This also involves indigenous communities (Tedmanson, 2008, p. 1009). These practices often result in conflicts, due to differences in values, norms, and ideas (Green, 1993; VanderHoff Boersma, 2009). Every community has separate identities, own practices and commons, and these communities expect these features to be taken into account by businesses.

Research highlights that the relationship between MNEs and indigenous communities has particularly been one of conflict (Oltremari & Jackson, 2006). Indigenous communities have the right to self-determination (UN, 2008), which means that they have the right to demand and expect respect for their culture, laws, territory, and authority (Yagenova & Garcia, 2009). These expectations increasingly result into conflicts, because they are not met or taken into account by the investing party (Oltremari & Jackson, 2006). Additionally, these communities usually have a long history of protecting its own territory, laws, and sovereignty, as they were once fighting against colonialism. Often, the mentioned distinctive values, cultures, and practices mismatch with the practices of MNEs investing in these communities (Green, 1993) and as McCartin (2013) states in her research, this mismatch can result in conflicts and a re-emergence of violence.

To understand the conflicts that arise between MNEs and indigenous communities, it is important to acknowledge why MNEs invest in these communities’ territories. As Dunning (1998) defined, two investing motives can be recognized. First, MNEs invest in these communities to seek natural resources. Therefore, current literature mostly covers conflicts within the extractive, or natural resource industry. The second motive concerns MNEs seeking strategic assets. Indigenous communities are widely recognized for their knowledge. For example, Reade and McKenna (2013) illustrate that indigenous peoples use practices from ancient times to resolve conflicts. Therefore, both community leaders and company managers should combine their knowledge as a possible way to avoid conflict. When applied in specific business contexts, the generated common conclusions can prove to be beneficial. This view is also supported by Calvano (2008). She argues that firms should be more accountable for their practices and the outcomes they have on local communities; managers should also understand the roots of the conflict. On the other hand, communities should understand the effect of their possible reactions to conflictive situations. Instead of letting these conflicts evolve into violence, companies and communities should be integrating their knowledge and practices.

In general, indigenous communities know a long history of unemployment, poverty, and poor health conditions (Cooke et al, 2007). MNEs have the possibility to mean a lot to these communities. For example, MNEs can enhance technological change and economic growth and create employment opportunities. Consequently, the economic status of community inhabitants improves, which poses an

opportunity for MNEs to involve in a mutual beneficial relationship with these communities (Akee & Jorgensen, 2013; Spencer & Gomez, 2011). Unfortunately, the investment of MNEs progressively results in negative outcomes. The culture of these indigenous communities is difficult to understand. Consequently, MNEs will experience even a larger liability of foreignness than they do when entering an unknown country (Green, 1993; Shapiro, Hobdari & Hoon Oh, 2018, p. 10). Research describes two possible causes of tension. First, the entering MNE must familiarize themselves with the indigenous practices before entering the community, which is costly and time-consuming (Green, 1993; Shapiro, Hobdari & Hoon Oh, 2018). Second, when the MNE eventually enters an indigenous community it is hard to define the property rights of not only the extracted natural resources, but it is also hard to divide the proceeds that are generated by using indigenous knowledge. Therefore, it is difficult for indigenous peoples to protect their knowledge and resources (Birch, 2016; Marinova & Raven, 2006; Orozco & Poonamallee, 2014; Reade & Mckenna, 2013).

Lastly, MNEs are more involved in less developed countries. These countries usually experience underdevelopment. Therefore, it can be cheaper for MNEs to extract resources from these countries. Research by Imbun (2006) mostly addresses that indigenous communities have their own laws and tax systems. As a consequence, the payment of MNE practices often needs to be made to the land owners. By researching two different mines, Imbun (2006) finds evidence that indigenous communities in developing countries expect socio-economic support from the MNE, for example in constructing infrastructure. This finding is supported by multiple other studies (Akee & Jorgensen, 2014; Crawley & Sinclair, 2003; Lertzman & Vredenburg, 2005; Yamin & Sinkovics, 2009). As corporate social responsibility (CSR) is getting more attention within international business, indigenous communities expect a certain ethical performance from MNEs. However, research by Oltremari and Jackson (2006) shows that too often these expectations are not instantly taken into account by investing MNEs and thus lead to conflictive situations and even violence.

National governments and country specific characteristics

Several authors discuss that nowadays the key tension within political economy is no longer the relation between states and markets, but between states and multinationals (Cohn, 2014, p. 254).

The nature of MNE activities secure regulation of foreign direct investment (FDI) is difficult and the effects of these investments are hard to measure. This is due to the fact that policies on FDI regulation are mainly formulated at the state-level instead of at a transnational level, which could better apply to the nature of MNEs (Cohn, 2014, pp. 278-279). Due to this position of MNEs, the role for national governments is even more crucial in conflict resolution.

The United Nations Universal Declaration of Human Rights (UDHR), describes that ‘‘every

citizen is entitled to have all rights, such as right of language, religion, political and social origin and property’’ (UN, 1948, art. 1 and 2). Hence, states need to protect these indigenous communities as is

stated in article 2 of the ILO convention: ‘‘governments shall have the responsibility to… protect the

rights of indigenous peoples and guarantee respect for their integrity’’ (ILO, 1989). National

governments are therefore obligated to play an important role in the recognition of these communities and the applications of (inter)national laws. The article also states that these indigenous communities are entitled to be seen as equal to the rest of the population, to have the ability to enjoy full realisation of economic, social and cultural rights, and to be protected from socio-economic gaps compared to other people within society. The national government has a crucial role to fulfil its duty to protect the citizens living in the country. As a result, several governments apply these articles in their national legislation, such as South-Africa (ILO, 1989; ACHPR, 2009), Australia (Parliament of Australia, 1992) and Canada (CIGI, 2014).

Several studies investigate country characteristics regarding MNE and indigenous community conflicts. These characteristics and their obligations as stated in the UNHR (1948) and the UNDRIP (2008) usually prove difficult for states to fulfil. First, the level of development has been studied in combination with indigenous communities before. Usually, indigenous communities are poorer than non-indigenous peoples and if the country is already underdeveloped it will only worsen the position of the indigenous community (Cooke et al., 2007). Underdevelopment results in several outcomes for the communities, such as low literacy, poor health conditions, low life expectancy and low living standards. Cooke et al. (2007) use the Human Development Index (HDI) to measure these socio-economic conditions of several indigenous communities and compare them to non-indigenous HDI

rates. They eventually state that governments should be more involved in indigenous education and socio-economic development to improve these conditions.

Regarding the role of the government in improving the lives of indigenous communities’ governments of LDCs have a difficult position. Their populations are usually experiencing low literacy and underdevelopment (Cooke et al., 2007). The governments of these LDCs commonly find themselves at a crossroads. On the one hand, they feel the need for FDI to invest in their own economic development. On the other hand, the protection of their minorities can suffer from their need for investments. Additionally, Yamin & Sinkovics (2009) state that a paradoxical relationship is known between MNE strategies and economic development. MNE activities could have a positive impact on development and poverty reduction, for example by the creation of human capital and infrastructure. However, this involvement could also endorse the MNE with more bargaining power, which is usually dangerous in LDC’s.

Second, political factors shape institutions in a country. Economics and politics are interrelated and therefore, political life influences economic life. The level of incoming FDI is highly dependent upon the quality of institutions shaping a country’s investment environment (Dia, 1995; Luiz & Stewart, 2014; Pierre, 2015). The study by Luiz and Stewart (2014) in Africa supports these arguments. They state that institutional voids and non-regulations can pose challenges for MNEs investing in LDCs. As these markets are highly unexplored, the rules of the game are not very clear. In these countries, corruption influences every aspect of internationalization. They find in their case that MNEs often accept bribes from governments and governments from MNEs. To avoid the bribing, firms are advised to develop codes to deal with corruption, and Luiz and Stewart (2014) find that firms with a written code against corruption are often less likely to accept bribes.

Corruption can be defined as: ‘the extent to which government officials accept bribes’ (Blackburn, 2012). Luo (2011) and Dia (1995) state that corruption is the outcome of institutional weakness. Institutional weakness is a phenomenon that is difficult to address, because it is generally rooted in the development of the state (Luiz & Stewart, 2014). Luiz & Stewart (2014) find that institutional weak governments are likely to be self-serving, which makes it harder for indigenous communities to be protected by the national government. On the other hand, other research states that

MNEs, on the other hand, experience three challenges with respect to corruption. First, the host country’s corruption can increase LOF, making it more difficult for an MNE to get familiar with all a country’s practices and the practices of the indigenous communities (Luo, 2011, p. 377). Second, corruption can blur the margin between the informal and formal economy (Smith & Thomas, 2015). Last, Cross (2016) integrates the bribery of indigenous leaders in international business. She finds that the MNE often does not only need national consent to undertake a project, but also needs the indigenous leaders’ approval. This complicates the relationship between indigenous communities and MNEs, because more actors are involved in the decision-making process. States should seek the approval of the indigenous community, whereas states - especially in LDC’s - often seek the approval of the investing party (Cross, 2016; Yamin & Sinkovics, 2009).

Consequently, a combination between weak regulations and a corrupt government engender an increase in conflicts and a lengthier conflict resolution (Kolk & Lenfant, 2013). Rule of law is a concept that is usually connected to both development and corruption, because it shows the strength of a country’s institutions and a country’s ability to strengthen the help their inhabitants really need. To illustrate, if a country is less developed and corrupt, the institutions usually are not developed as thoroughly (Blackburn, 2012). Therefore, indigenous communities do not gain the help in conflict resolution that certain aforementioned articles derived from ILO and UNDRIP stated.

Third party involvement

Research into conflicts between MNEs and indigenous communities increasingly addresses the topic of collaborative management, in short: co-management (Castro & Nielsen, 2001; Cundill, Thondhlana, Sisitka, Shackleton & Blore, 2013). Research into co-management mostly covers NGOs involving in these conflicts. NGO’s are usually defined as ‘non-profit groups that combine mobilization

resources, information provision and activism to advocate for certain issue areas’ (Spar & La Mure,

2008, p. 79). However, civil society groups, such as media, labour unions, international governmental organizations, pressure groups or lobby groups are also involved in the conflict resolution. Freeman (1984, p. 46) defines stakeholders as: ‘Any group or individual who can affect or is affected by the

only stakeholders involved in resolving conflicts. Research has also focused on involvement of international financial organizations (McCartin, 2013; Sarfaty, 2005), non-governmental organizations (NGOs) (Lertzman & Vredenburg, 2005; Millar, Choi & Chen, 2004; Oetzel & Doh, 2009; Perdersen & Pedersen, 2013; Sethi et al., 2011) and other civil society groups (Oetzel & Doh, 2009; Perdersen & Pedersen, 2013).

A study by Murphy and Arenas (2010) illustrates the importance of treating civil society groups as a stakeholder within conflicts between indigenous communities and MNEs. They state that collaboration with - and involvement of - civil society organizations (CSOs) has improved indigenous-MNE relationships. They find that collaborations between CSOs and corporations exist to create social value and implement CSR onto an MNE’s agenda. As is stated before, indigenous communities appreciate the creation of social value when entering relationships with businesses. In these relationships trust, mutual understanding and communication are crucial. When these relationships are less collaborative, they are more likely to be conflictive (Murphy & Arenas, 2010, p. 107). They therefore propose to research the concept of civil society instead of solely focussing on NGO-involvement when analysing the conflicts between indigenous communities and MNEs (Murphy & Arenas, 2010, p. 117).

The United Nations describe civil society as ‘the third sector of society, along with governments

and businesses’ (United Nations, 2018). Murphy and Arenas (2010) define that ‘Civil society advances social change through public policy.’ They state that the eventual goal of civil society is to improve

accountability on both the business and governmental side of legislation. To achieve that, governments need to be accountable to civil societies, because these groups secure checks and balances, especially in a corrupted or weak institutional government’s decision-making (Dia, 1995). The ultimate consequence of civil society influence is policy change. To achieve this stakeholders could be more involved in conflict and resolve or even prevent conflict (Centre for Civil Society, 2018). Hall (1998, p. 32) conceptualizes civil society as societal self-organization, and he emphasizes one clear characteristic: civil society is based on diversity. The concept enhances multiple institutions and organizations. Hall (1998) states that it is more than just the presence of strong and autonomous groups, because these organizations can evoke change (Hall, 1998, p. 41).

Overall, the literature distinguishes four advantageous practices of third parties with regarding to business-community conflict. First, CSOs are usually able to address the distinctive values, culture and practices of the community, instead of clashing with these practices (Miller et al., 2004). Second, CSOs can invoke a dialogue. As these organizations are generally fair-minded and neutral, they can oversee the conflictive situation. They are able to create a global voice and awareness for pressing causes (Millar et al., 2004). Third, CSOs are able to facilitate multilateral aid (McCartin, 2013). CSOs can provide services that can complement the resources of the MNE and by co-evolution they can help develop and establish institutions in host countries (Oetzel & Doh, 2009). Finally, CSOs can bridge the gap between institutions and MNEs (Millar et al., 2004).

Previous collaborations between MNEs, indigenous communities and CSOs have secured some positive outcomes addressed in research. First, CSOs put CSR on the political and business agenda. Second, CSOs can secure a comprehensive discussion with MNEs and the host country government on its CSR practices. Third, the organizations can convince MNEs to adapt a sustainable business model to serve the indigenous communities. Last, with regarding to the natural resource and extraction industry, CSOs can enforce MNEs to adopt environmentally acceptable standards as regards to the natural resource and extraction industry (Lertzman & Vredenburg, 2005; Millar et al., 2004).

McCartin (2013) investigates the tense relationship between MNEs and CSOs. On the one hand, she finds that these organizations can provide businesses with credibility, cognitive capacity, connections and the capacity of confrontation. On the other hand, businesses can provide CSOs with cognitive capabilities, connections, the capacity to change and resources, such as cash and manpower (McCartin, 2013, p. 74). Therefore, collaborative relationships can lead to a decrease of LOF and an increase in spill-over effects for development and the reduction of corruption (Oetzel & Doh, 2009).

Various studies address the subject of conflict resolution to prevent (violent) conflicts. To illustrate, McCartin (2013) uses the concept of lobbying. As businesses usually approach issues rationally, realistically, pragmatically and professionally, their behaviour clashes with the emotional and culture-related approach of the communities. However, dialogues with third parties can lead to reduced length and density of conflicts. Moreover, she finds that a two-way learning and adaption approach combined with shared power can lead to long-term sustainable relationships (McCartin,

2013). Nonetheless, the MNEs need to reform their business models in order to accomplish these long-term relationships with indigenous communities (Lertzman & Vredenburg, 2005; Oltremari & Jackson, 2006). McCartin (2013) states that unfortunately these findings are derived from reseach based on single case studies, hereby isolating the case from its historical base and other cases. Thus, she states that to look at past conflicts in a cross-comparative way could pose viable solutions and enhancement for conflict resolution. In this way findings could also provide enhancements for conflict resolution, as history tends to repeat itself (McCartin, 2013).

Research shows that CSOs play crucial roles in developing countries, especially regarding to economic development and strengthening political institutions (Boswell, 2018; McCartin, 2013). Boswell (2018) states that the concept of corruption is often accepted as ‘part of the culture’, instead of something that has the ability to change. Consequently, civil society groups create platforms to discuss these negative consequences. Arenas, Sanchez and Murphy (2013) state that civil society and businesses are usually involved in tense relationships. They suggest that civil society-business partnerships are unexplored and unaddressed and the development of a comprehensive framework could address the subject of collaboration between civil society and businesses. They find that the relationships usually start tense, but evolves towards more collaborative partnerships. Specifically, within indigenous community – MNE partnerships does civil society usually plays a crucial role regarding the outcome of collaborations and the devolvement of the conflicts. They address the involvement of civil society in a conflict between an MNE, Repsol, and an indigenous community conflict in Bolivia. In this case, new policies were invoked, social issues were addressed and the company’s business model was adjusted to avoid conflicts. As a result, the MNE was categorised as strategic asset-seeking investor, an actor which is not involved in solution-generation. Instead, Arenas et al, (2013) find a trend that CSOs resolve the conflicts more adequately and quickly (Arenas et al, 2013).

Theoretical framework

This chapter provides a theoretical background on the topic of country characteristics and co-management by civil society organizations within indigenous community-MNE conflicts. Throughout this chapter, hypotheses are proposed based on literature as studied before. Finally, the chapter will provide a conceptual framework, which will be the foundation for this study.

The dynamics of conflict

Conflicts within international business research usually arise when multiple groups of people, companies or states pursue interests that do not align with each other and therefore are difficult to exist together (Calvano, 2008; Lertzman & Vredenburg, 2005). The literature indicates that conflicts between indigenous communities and MNEs exist for two reasons: one, the extraction of land, natural resources and water (Crawley & Sinclair, 2003; Imbun, 2006; Lertzman & Vredenburg, 2005; Oltremari & Jackson, 2006; Sethi et al., 2011; Shapiro et al., 2018), and two, different views on treatment of individuals, sustainability, human rights, social and economic rights and intellectual property (Lertzman & Vredenburg, 2005; Marinova & Raven, 2006; Reade & Mckenna, 2013). The literature illustrates that indigenous communities are overall distinguishable from the major social groups, which make these communities vulnerable to negative practices (Thornberry, 2002). Additionally, Kingsbury (1998) defines four essential indicators: self-identification as an ethnic group, historical experience of disruption, dislocation or exploitation, a long connection with the region, and the wish to retain a distinct identity.

Hence, research needs to be extended addressing the dynamics of the actors within these conflicts (Calvano, 2008), which are not only the indigenous communities and MNEs, but also national governments and CSOs. For example, literature does not show consensus in which country-specific indicators are crucial, but studies do show that country characteristics are important in analysing the conflicts between indigenous communities and MNEs. How these community characteristics are shaped and determined could be dependent upon country characteristics. One fact is overall clear: indigenous communities feel a disconnect with representation (Green, 1993) and therefore, the relationship between MNEs and indigenous communities is overall one of conflicts (Oltremari & Jackson, 2006).

Research covers the conflicts between MNEs and indigenous communities extensively through case study research. This approach allows the researcher(s) to thoroughly investigate all the features of a specific case, usually the indigenous community. The approach emphasizes an intensive examination of the chosen setting (Bryman, 2012, pp. 66-69). However, a case study does not generate multiple theoretical perspectives and is therefore low in external validity. Consequently, this research addresses multiple cases within a single study. This will provide more insights into the nature of these indigenous communities, and their dynamic relationship with MNEs and third-party actors.

Oetzel, Getz and Ladek (2007) argue that conflict resolution can be diverse in both duration and degree of violence. The variables as proposed by Oetzel et al. (2007) will act as dependent variables for this research. Consequently, country characteristics influence the duration and degree of violence. For example, an underdeveloped or institutional weak national government will struggle a lot more with the protection of indigenous communities, which could result in longer and denser conflicts. Hence, governments equipped with the means to protect these communities, are more able to resolve these conflicts (Oetzel et al., 2007; McCartin, 2013).

Country characteristics and their influence in conflict resolution

Previous studies on country characteristics in the conflicts between indigenous communities and MNEs have demonstrated the importance of these indicators in conflict resolution. One difficulty is overall clear: the dominant method in conflicts between indigenous communities and MNEs is single-case study research. Therefore, it is challenging to generate cross-national conclusions, which compare different country-characteristics and their influence on conflict resolution. Previous research has also often discussed only one country level characteristics - for example development (Cooke et al., 2007) or corruption (Cross, 2016). Hence, an extensive overview of multiple cases in diverse countries that cover both socio-economic and political activities is lacking.

First, literature considers the level of development as an important characteristic in conflict resolution. Cooke et al. (2007) state that indigenous communities are frequently poorer than non-indigenous peoples. This means that the non-indigenous communities have an even more marginalized position in developing countries than they have in developed countries. MNE investments could

improve the economic position of these communities. Controversially, investments of MNEs in indigenous communities in developing countries often secure longer and more dense conflicts (Imbum, 2006; Yamin & Sinkovics, 2009). Second, the natural resources, land and water located within the territorial boundaries of the indigenous communities could be the main source of income for these communities (Banerjee, 2008; Foster, 2012). This second effect occurs in developed and developing countries, but could be tremendous for the already weak-positioned indigenous communities in developing countries. Third, economic benefits that derive from MNE investments could occasionally be more important for the government than protecting the rights of the indigenous communities. Especially in developing countries this can lead to violent conflicts when these rights are enforced by the indigenous community (Foster, 2012; Kolk & Lenfant, 2012). Last, the literature does not yet provide evidence of better processes of conflict resolution in developed countries than in developing countries, but state that the dynamics are important to investigate (Bellier & Préaud, 2012).

This study argues that the level of development of a country will influence the ability of indigenous communities and national government to influence the conflicts between indigenous communities and MNEs. It is proposed that the conflicts will last longer and will be more violent in countries with a low level of development. For example, Banerjee (2008) discusses the Chiapas community in Mexico. Their land has been extracted to grow cattle feed for the American fast food industry to resolve food security issues in the United States. He illustrates that this land provides most of the income and economic growth for this community. By diminishing this source of income, the MNE has a negative influence on the economic position of the community. This example illustrates that conflicts can be denser and longer if the indigenous community already has a subordinated position (Banerjee, 2008; Foster, 2012; Kolk & Lefant, 2013). Consequently, it is important to analyse the country level of development, because people in indigenous communities will be more motivated to protect their main source of income. For MNEs, a country with a low level of development could impose more difficulties – such as a larger liability of foreignness (Shapiro et al, 2018). Therefore, we can predict that:

Hypothesis 1 (H1): If countries are less developed, it is more likely that the conflict between MNEs and indigenous communities last longer and will be more violent.

Besides the analysis of a socio-economic concept, this study will also analyse two political concepts. These two concepts together describe ‘the respect the state has for citizens and for institutions that

govern economic and social interactions among them’ (Kaufmann, Kraay & Mastruzzi, 2011; United

Nations, 2015). The first concept - the control of corruption - could be one of the country-level indicators that influence the ability of a government to protect indigenous communities. Previous literature shows different effects of corruption on the position of indigenous communities and the conflicts that occur between them and MNEs. First, Dia (1995) and Spencer and Gomez (2011) state that institutional voids enhance corruption, which can secure unequal representation for indigenous communities. This suggests that it could be easier for MNEs to extract natural resources, because social and human rights are less protected by the national government. Second, corruption can blur the boundaries between the informal and formal economy, which secures difficulties for MNEs to discover the rules of the game (Luiz & Steward, 2014; Smith & Thomas, 2015; Spencer & Gomez, 2011). Third, Blackburn (2014) describes that the practices of weak and corruptive governments could cause violent reactions from their inhabitants, because. Additionally, the so-called curse of resources poses a difficulty, because the different parties – governments, MNEs and indigenous communities – have different interests in their protection, the extraction of resources and income generation. This can lead to conflicts, and more importantly to violence (Luiz & Stewart, 2011; Luo, 2011; Dia, 1995).

This study argues that conflicts in countries with a high level of corruption could be lengthier and more violent than conflicts in countries with a lower level of corruption. The concept ‘control of corruption’ refers to ‘the extent to which public power is exercised for private gain’ (Kaufmann et al., 2011). This study proposes that conflicts in countries with a high control of corruption will less lengthy and less violent. To illustrate this hypothesis, Whiteman and Mamen (2002) describe the effect of corruption on a conflict between an indigenous community and an MNE in the mining industry. They state that states which are dependent upon natural resources have usually lower rates of control of corruption. This implies that in these states MNEs with help of government officials secure to extract these resources (Whiteman & Mamen, 2002, p. 68). Therefore, we could propose:

Hypothesis 2 (H2): A higher control of corruption will secure a less long and less violent conflict between MNEs and indigenous communities.

In addition, the study of Kaufmann et al. (2012) states that the concept of ‘rule of law’ should be included into the analysis in order to draw a full picture of the government’s respect for citizens and the institutions. These two concepts – control of corruption and rule of law - need to be analysed together to fully understand the concept of corruption (Kaufmann et al, 2012, p. 4-6). Rule of law can be described as ‘the extent to which agents have confidence in and abide by the rules of society and in

particular the quality of the police and the courts, as well as the likelihood of crime and violence’

(Kaufmann et al, 2012, p. 4). The concept indicates the strength of formal institutions in a country, which according to the literature is a prerequisite for determining the effect of the level of corruption (Luo, 2011; Dia, 1995). The strength of these formal institutions differs per country and illustrates the ability of a government to apply and execute their rules and regulations. This indicates that rule of law should be included in this analysis as well, because a high rule of law leads to less abuse for public and private gain. Clearly, a country with a higher rule of law will be more able to resolute conflicts and protect every inhabitant of their country.

This study argues that conflicts between indigenous communities and MNEs will be less lengthy and less violent in countries with a high rule of law than countries with a low rule of law. For example, Owoye & Bissessar (2012) find that weak institutions can lead to a rising level of violence. Countries with a low rule of law are often countries do not have the ability to monitor everything that is going on in their territory, because usually the administration is centralized. Also, a low rule of law can negatively influence the marginal position of the indigenous community (Foster, 2012; Kolk & Lefant, 2013) and increase the length of these conflicts and the degree of violence. Therefore, the third hypothesis of this study is proposed as:

Hypothesis 3 (H3): A higher rule of law will secure a less long and less violent conflict between MNEs and indigenous communities.

Co-management and conflict resolution

Literature states that it is hard to understand the dynamics of conflict resolution between indigenous communities and MNEs without involving the concept of collaborative management, in short: co-management (Castro & Nielsen, 2001; Cundill, Thondhlana, Sisitka, Shackleton, & Blore,

2013). The literature on co-management regarding these conflicts distinguishes two important trends. First, the role of NGOs is increasingly - but solely - analysed as important in conflict resolution. However, NGOs are not the only third-party actors involved in co-management; other civil society groups’ also influence conflict resolution (Arenas et al., 2013; Freeman, 1984; Murphy & Arenas, 2010). For example, research refers to media groups, labour unions, international organizations, pressure groups and media as important actors within conflict resolution (Arenas et al.,2013; Persson et al., 2017). Consequently, the concept of civil society intervention is crucial to analyse the conflicts between indigenous communities and MNEs.

This study investigates the ability of CSOs to moderate the different proposed relationships between three country-level indicators that influence the duration and degree of violence of the conflict. First, CSO involvement can increase economic development, as CSOs have the ability to gain awareness for underdeveloped minority groups within society (Boswell, 2018). Third parties are increasingly involved in developing countries, because people in these countries are unable to ascertain the possible economic opportunities to improve their positions. Additionally, McCartin (2013) argues that the involvement of CSOs can offer equal representation within these conflicts. In this case, the indigenous community can gain the protection of their human rights, while the MNE can better understand the causes of conflict. Finally, several researchers highlight the fact that CSOs can secure MNEs to value CSR practices. As Lertzman and Vredenburg (2005) find, MNEs that have CSR policies are usually involved in less lengthy and less violent conflicts.

This study argues that the proposed negative relationship between level of development, and length of conflict and degree of violence will be positively moderated by CSO involvement. To illustrate, Mercer and Green (2013) state that the work of civil society actors in developing countries has changed. Previously, they acted mainly as donors to enhance economic development, whereas nowadays, the efforts of CSOs have spread to community development, conflict resolution and holding governments accountable for economic development of indigenous communities. They find that in Tanzania, CSOs have built communities, created local jobs, and helped foreign investors to enhance in sustainable business practices and development. To determine the effect of civil society involvement

on the relationship between level of development and conflict resolution in different cases, the fourth hypothesis is proposed as:

Hypothesis 4 (H4): Civil society involvement moderates the negative relationship between level of development and conflict length and degree of violence.

Furthermore, the relationship between control of corruption, and length of the conflict and degree of violence is also expected to be moderated by civil society involvement. CSOs are expected to be the bringers of change in countries where the government is highly corrupt (Hall, 1998). Usually, corruption is the outcome of institutional weaknesses rooted in the development of a country (Luo, 2011). Indigenous communities can experience difficulties in protecting their rights when the government is corrupt and institutional features are difficult to understand (Cross, 2016). CSOs can protect the rights and improve the economic and political position of these communities.

To illustrate, McCartin (2013) finds that CSOs can provide MNEs with credibility, cognitive capacity, connections and the capability of confrontation. This suggests that the practices of these MNEs can be conflictive. However, engaging with a CSO can decrease the negative associations and outcomes of the MNE practices within indigenous communities (McCartin, 2013). In addition, McCartin (2013) describes that this business-indigenous communities clash especially occurs in corruptive countries. Dialogues with CSOs could reduce the duration and the density of conflicts, even when corruption is high. Therefore, the fifth hypothesis is proposed to be:

Hypothesis 5 (H5): Civil society involvement will moderate the relationship between control of corruption and conflict length and degree of violence.

The final hypothesis evolves around the concept ‘rule of law’. Rule of law describes the strength of institutions. Different outcomes can be derived from institutional weakness, such as corruption. Oetzel and Doh (2009) clarify that CSO-MNE collaborations could help to develop institutions. This could be beneficial for indigenous communities, as CSOs bridge the gap between institutions and MNEs. This suggests that the LOF decreases through the involvement of CSOs in countries with a low rule of law (Arenas et al., 2013). In these countries, governments can easily ignore their responsibility to improve

the lives of every citizen in their country (Boswell, 2018). Through their objective and pressuring nature CSOs can protect the rights of minority communities.

To illustrate, Arenas et al. (2013) evaluate the country-level institutional effect on different local communities in four case studies. They find that politically weak countries often find themselves to be a subject for CSOs. An indigenous community in Bolivia is facing a difficult political context and with the help of different CSOs, new policies are developed to structure indigenous relations. This conflict resolution also included a risk assessment of social issues, which the government could take into consideration in the development of new policy towards indigenous communities. These illustrations highlight the role of CSOs in conflict resolution. Therefore, the sixth hypothesis is proposed as:

Hypothesis 6 (H6). Civil society involvement will moderate the relationship between rule of law and conflict length and degree of violence.

Conceptual Framework

Figure 1 illustrates the conceptual model. The figure demonstrates the proposed relationships between the three independent variables - level of development, control of corruption and rule of law - and the dependent variables - length of conflict and degree of violence (H1, H2 and H3). Additionally, we predict that the moderating variable - civil society involvement - influences these relationships (H4, H5 and H6). This leads to the following conceptual framework:

Research design and method

This chapter starts with a detailed description of the data collection and the sample. The part afterwards explains the dependent, independent, moderating and control variables. Finally, the chapter presents statistical methods and some justifications for their use.

Data and sample

A dataset existing of 706 cases will provide the opportunity to conduct a comprehensive, cross-sectional analysis involving country-specific characteristics and multiple actors within conflict resolution. The data was collected by fellow students involved in this project. The cases in the dataset are widespread across the world and representing 67 countries. The data was collected through secondary databases, for example NGO and news reports, the World Bank and the UNDP, and coded by a standardized coding scheme. The specific coding used for this study can be found in appendix 1. This year, we collected additional literature on our particular research questions to improve and back-up our proposed hypotheses and future studies.

The sample is based on the entire dataset. However, some cases were excluded because of missing data and to increase the fitness of the study. The exclusions will be explained in the parts on the specific variable (e.g. the dependent, independent, control or moderating variables). This sampling method resulted in a sample of 514 cases over 58 countries, such as Australia, Brazil, the United States and South-Africa.

Dependent variables

Previous research illustrated in the literature review indicates that the length of a conflict and the degree of violence are important factors in conflicts between MNEs and indigenous communities. In this research, the length of the conflict and the degree of violence will act as the dependent variables to illustrate the dynamics of these conflicts between MNEs and indigenous communities.

Length of the conflict will be measured by the months that a conflict lasts. According to Oetzel et al. (2007), violent conflicts can be long or short. The length of a conflict will be coded as either 0, which will mark a ‘short ’conflict that lasts shorter than the median (less than or 84 months and not