Conflicts between indigenous communities and MNEs : the impact of weak institutions

(1)

Amsterdam Business School

Conflicts between Indigenous Communities and MNEs: The

Impact of Weak Institutions

Valentina Guatterini 11085932 23/06/2017 – Final Draft MSc Business Administration International Management University of Amsterdam Supervisor: Dr. Ilir Haxhi Second reader: Drs. Erik Dirksen

(2)

Statement of originality

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

(3)

However, previous literature fails to study from a quantitative perspective, the dynamics of these conflicts, especially the impact of national governments and institutions are overlooked. This study wants to assess the relationship between the institutional strength of governments and the length and violence of the conflict. By analyzing around 700 cases, the study wants to statistically assess whether the health of institutions can be a factor affecting the conflict. Furthermore, we will study the moderation impact of the corporation’s experience and sustainability focus, and the community’s isolation and autonomy.

The analysis shows that healthier governments lead to shorter conflicts, however there is no impact by the institutions on the violence of the conflict.

In addition, the moderators chosen do not provide any significant results. However, the study provides a basis for further literature in detangling the dynamics of the conflict, by providing an empirical basis on which new assumptions will be built.

(5)

Introduction

International business literature provides an extensive analysis from the ethical perspective of doing business abroad. Since the first theorization of the concept of Corporate Social Responsibility (CSR) (1987), corporations have tried to find a tradeoff between the short term profit obtained by the exploitation of the natural resources that often belong to minorities and the long term benefit of performing business with a sustainable model that brings an added value not only to the company but to the overall society. Despite this overall shift in the business towards a fair a sustainable model, conflicts between local communities still represent an unresolved case and they are very hard to prevent. On February 15, 2017, for example, 44 corporations were awarded by the Indian government with a contract for the extraction of hydrocarbures in the region of Tamil Nadu. One day after, protests out broke from a small village in the south of the region, called Neduvasal, the operation has a strong ecological impact on the land and the territories of the community. The case is quite new, and it is just an example of hundreds of similar stories from every corner of the world: 370 million indigenous people live in 90 countries and they represent one of the poorest shares of the global population (World Bank, 2016). From the 1970s NGOs have started to try to find a solution and to promote agreements that would grant more rights and thus more bargaining power to the communities. (Kymlicka,1999) For example, the United Nations has established the UN Permanent Forum on Indigenous Issues, an advisory board that led to the adoption of the Declaration of Indigenous People (2003), a non-legally binding instrument that recognizes the rights of indigenous communities. Despite these efforts, a study provided by the Columbia Center on Sustainable Investment shows that from the 1990s conflicts have increased steadily. Globalization and scarcity of resources have led MNEs to search in more isolated areas and thus encountering indigenous communities with the consequence of destroying their natural habitat, culture and identity (Foster, 2012; Murphy & Arenas, 2010).

(6)

Also, Calvano (2008), states that the increase in conflict between MNEs and local communities is the result of the influence of three factors which are: stakeholder power inequality, stakeholder’s perception gap, and cultural context. In addition to this complexity of factors, there is a complexity of actors involved: among which national governments and institutions play a big role, ranging from the communities itself, national governments and international organizations. In other words, the phenomenon of MNEs and community conflict is complex and requires a good understanding to achieve solutions (Child, 2002). However, literature fails to prove a clear relationship between these actors, especially from an empirical point of view. A leitmotif in many studies (Ikeledgbe, 2005) regarding company-community conflicts is the existence of weak national institutions in those countries where indigenous communities are located: corruption is considered one of the main causes of the conflict. Kolk & Lenfant (2013) and Ikelegbe (2005) argue that national formal institutional weakness decreases the position of indigenous communities through a lack of legal systems, governmental instability and corruption.

In this study we will try to understand if national governments are one of the parties involved, and, in particular, whether the strength of these governments has a direct impact on the conflict. Do governments have an impact on the conflicts between MNEs and Indigenous communities? Does the institutional framework impact the duration and the violence of the conflict? Are there any factors moderating the strength of the institutional framework?

The research question will be twofold:

Our main research question will then be twofold:

1) What is the impact of institutional weakness on the length and type of conflict?

2) What is the moderating effect of sustainability focus, company’s previous experience,

(7)

Finding a solution to the conflicts or at least obtaining a better understanding of the dynamics is fundamental for the corporations: the international Council on Mining and Metals (Mining and Indigenous Issue Review) states that successful mining operations require the support of the communities in which they operate and the relationship with them should be founded on respect, meaningful engagement and mutual benefit. We can translate the statement to any industry: corporations could have delays in the project, leading to higher costs due to these conflicts; a beneficial relationship with the community would certainly decrease the risk. Also, the existence of the predicted relationship would represent an important contribution to the international business strategy, as it builds on the Institution Based View (Peng, 2002). Institutions need to be taken into account when doing an international business activity as they constantly influence the corporations and cannot be considered as an isolated body. (Mayer, Estrin Bhaumik, Peng, 2009). Because national government are so different across the world and some are in better economic, legal and ethical conditions than others, we would like to investigate up to what extent the type of government, and especially the health of this government (the strength), influences the conflicts between communities and multinational corporations. If the government strength has a direct impact on the conflict, it means that the change has to start in the political environment first. NGOs and supranational organizations could address their agendas towards lowering corruptions and enhancing a fairer and stronger institutional framework.

With our study we want to provide for the first time an empirical evidence of the relationship between weak institutions and conflicts. We will also contribute to the literature by providing an analysis on a global level rather than just focusing on a single case or a single industry as most of the studies in the field do. Most of the studies focus on a particular geographic area, or in a particular industry, especially the extractive one. In our studies we will have data from several countries and from seven different industries. Finally, we will provide an empirical

(8)

analysis on the dynamics of the conflicts which is not common in the literature. Most of the studies on conflicts present a qualitative analysis that focuses on a particular case; we will run a statistical analysis to assess the exact impact of institutions on the length and type of conflict.

The aim of the research is to understand which are the factors that might lead to the solution of the conflict and those that instead might be a cause of it. In addition, we will try to underline the importance of national governments’ institutions when the rights of local communities are at stake. To answer the research questions, we will analyze approximately 350 cases from 41 countries and 18 industries. All data is secondary and is gathered through third party reports from organizations that study indigenous communities and from non-governmental organizations and then we coded it according to a coding scheme.

The thesis is structured in the following way: firstly, a literature review is presented. In particular, an extensive analysis of the main concepts of the study will be presented: conflict, communities, MNEs, institutions and those characteristics that will be the moderators of the relationship. Secondly, the theoretical framework with five hypotheses is presented. Finally, the statistical analysis and the results will be provided. This section is followed by a discussion of the results, on the basis of previous literature.

(9)

Literature Review Literature Review

The following chapter provides an analysis on the main actors involved in the conflicts: communities, multinational corporations and institutions, based on previous literature.

Also, it includes an analysis on some characteristics that we assume as moderators of the relationship between institutions and communities.

Communities

The concept of community has been researched by scholars of different disciplines. Literature shows that there does not exist a unique definition for the concept, but it is defined differently according to the dimension in which it is studied. In other words, the literature presents studies on communities that range from a geographical dimension to a sociological one; religious communities, occupational communities, even internet communities. The concept adapts to different dimensions with the common trait of “sharing something”. Kapelus (2002) points out that identifying a community is a complex task; any definition implies imposing an order that often does not fit the experiences of the people in question. This applies in particular for indigenous people that have radically different world views that may clash with the capitalistic view of corporations (Wheeler, Fabig and Boele, 2002).

For the purpose of our studies, we will try to analyze specifically the indigenous communities and the values they share. The term indigenous has been itself defined in many ways probably because there are more than 370 million indigenous people across 70 countries. Because they are spread in every corner of the world it is difficult to provide a definition that includes the characteristics of such a variety of people and cultures.

The UN argues that the most effective approach with indigenous communities is to identify rather than define indigenous people (United Nations, 2014). Identity is one of the crucial values of indigenous communities as most of them have been subject to colonialization by

(10)

Western countries that ignored their traditional culture and imposed their value. Furthermore, the common sharing of values, as Brint (2001: 8) points out, is an intrinsic characteristic of these communities. Brint (2001) also argues that most of the communities are built on informal constraints, thus formal institutions are not fundamental for the surge and the survival of the community. With informal constraints we refer to traditions, customs, and taboos. (North, 1992)

Another study, presented by Dunham, Freeman and Lietdka (2006), tries to define communities on four levels: firstly, communities of place; secondly, communities of interest, thirdly virtual advocacy groups and finally, communities of practice. In communities of place the physical proximity of the members characterizes the community itself; firms have to take into account the consequences that their operations could have on these communities. Such conceptualization falls into the stakeholder theory and in the past years many multinational corporations have been working to actively participate in supporting the communities. Communities of interest fall as well into the stakeholders’ theory as they are advocacy groups that share a common focus mainly targeting business firms. Virtual advocacy groups are similar to communities of interest in the sense that they have a shared focus, but they are more “Opposition-driven” rather than “Agenda-driven”. One of the most common examples is the “No Global” movement that emerged in 1999 at the Seattle WTO Meeting (Bernard, 1999). Finally, communities of practice are those professional working groups of people that are linked by a sense of shared interests, values and purpose. Also, they are characterized by a shared identity and a mutual obligation (Dunham et al. 2006).

Nonetheless, Dunham et al. provided an extensive analysis of the concept of community from different perspectives. We will follow Calvano (2008) and we will focus on communities of place as the aim of the study is to understand how conflicts between MNEs and Indigenous communities relate with the strength of institutions in a particular geographical location. So,

(11)

if we follow the concept of community of place, the question that arises is: are these communities recognized by national governments and do they have to be recognized by them to be considered as a community? To which extent is the physical proximity of the members sufficient to characterize them as a community?

The ILO Convention 169 (Entered into force on 5/09/1991), one of the most important documents concerning the rights of indigenous communities, defines indigenous first as “tribal peoples in independent countries whose social, cultural and economic conditions

distinguish them from other sections of the national community, and whose status is regulated wholly or partially by their own customs or traditions or by special laws or regulations”; and

then as “peoples in independent countries who are regarded as indigenous on account of

their descent from the populations which inhabited the country, or a geographical region to which the country belongs, at the time of conquest or colonization or the establishment of present state boundaries and who, irrespective of their legal status, retain some or all of their own social, economic, cultural and political institutions.”

Thus, according to the ILO, it is not necessary for a community to be recognized by the local government as indigenous; self-identification can be enough as long as the community has kept some cultural, social or political institutions that were existent before some historical contingencies like colonialization. In the definition of the Convention there is a combination of an institutional perspective with a geographical and territorial one. In fact, the two main elements for being considered indigenous are either special regulations granted by the national government or the possession and occupancy of a particular territory, in spite of conquests by other populations, in which they kept the same institutions (formal and informal). To sum up, the definition provided by the ILO convention is a summary of what previous literature has provided, and shows that not only the territory (community of place)

(12)

but also a common institutional trait, whether it is formal or informal, is part of the community’s definition.

Communities’ Governance

The definition of community provided by the ILO convention explained in the previous section underlines the need to a certain extent an autonomy of the community, which means that every community should have a governance body managing the local issues with ad hoc legislation. This autonomy is granted by the national government itself and Nelles and Alcantara (2013) provide a paper that analyzes the dynamics of communities’ governance. The authors point out that most of the communities function on the basis of multilevel governance which means that governments engage with a broad range of actors at a different territorial level to pursue collaborative solutions to complex problem.

Thus, in an ideal world, communities’ governance is made up of different actors that work together in finding solutions to a common issue. The multilevel governance is used especially in Canada, where the indigenous presence is quite high and the government is trying to find a solution to ensure respect of their rights. However, even in the case of Canada this type of governance fails to provide a solution to all the issues that the Canadian aboriginal communities have (Nelles and Alcantara, 2013). It is therefore necessary to zoom out and provide a deeper understanding of what is good governance in general and how communities rights and needs are usually addressed.

As Hunt and Smith (2006) point out, good governance means creating an environment where ordered rule and collective action take place. Governance is particularly important for Indigenous communities because they usually have the highest rates of poverty, unemployment, early mortality and the lowest levels of education of the country (Hunt and Smith, 2006; Canadian Centre for Policy Alternatives, 2013); good governance could lead to a sustainable development of the community. The National Centre for First Nations

(13)

Governance (NCFNG), a Canadian NGO that tries to develop self-governance in First Nation communities1 across the country, defines governance as the “[t]raditions and institutions that a community uses to make decisions and accomplish its goals”. An effective governance is based on five pillars (people, land, laws and jurisdiction, institutions, and resources) that represent a mixture of the traditional values of the indigenous communities with the modern concept of self-governance. Also, the NCFNG defines seventeen principles that guide and support the abovementioned five pillars. It is important to underline that the five components of the effective governance are presented in a hierarchical order: good governance starts with the people and ends with the resources (NCFNG, 2009).

Although the definition and the governance characteristics presented are focused on Canadian indigenous communities, a similar outcome was obtained by Hunt and Smith (2006) who studied the governance of Australian Aboriginal Communities. The two authors point out that an effective governance depends on a mix of factors that represent a complex web of issues that cannot be simplified. But among those factors, is the ability of the governance to share information and to listen to the voice of the people. Also a good governance has to be able to protect people’s land. In addition to these endogenous factors, meaning directly dealing with the community, there are also exogenous factors, as national government’s funding plans to minorities, which do not depend on the indigenous governance itself, but on the overall wealth of the country in which the community is developed.

An increasing trend in the community governance and especially to avoid conflicts on natural resources is co-management. Co-management connotes a collaborative institutional

1

A term that came into common usage in the 1970s to replace the word "Indian," which some people found offensive. Although the term First Nation is widely used, no legal definition of it exists. Among its uses, the term "First Nations peoples" refers to the Indian peoples in Canada, both Status and non-Status. Some Indian peoples have also adopted the term "First Nation" to replace the word "band" in the name of their community (Indigenous and Northern Affairs Canada, 2012)

(14)

arrangement among diverse stakeholders for managing or using a natural resource. (Castro and Nielsen, 2001)

However, a degree of conflict may be necessary before the state and other stakeholders are willing to enter into negotiations for a co-management agreement (Castro and Nielsen, 2001). To sum up, it is hard to define the concept of good governance in the context of local communities. The literature suggests that there has to be an effort from both the members of the community and the national government itself. The main outcome from the previous literature is that communities’ government is always linked to other important actors: firstly, the national government itself and then everyone that has interests in the territory.

In order to reach effective governance agreements, most of the communities have to go through conflicts, while it should be the other way around; indigenous communities should already have a well-established governance in order to provide stronger claims on their territory.

After a brief analysis on the first actor of the conflict - indigenous communities - we will try to give a definition to multinational enterprises, or at least we will try to provide some insights on their functioning and their main dynamics.

Multinational Enterprises (MNEs)

MNEs are coordinated systems or networks of cross-border value-creating activities, they differ from normal corporations, by the fact that they have set up part of their value chain abroad, thus involving a foreign direct investment.

Many scholars have studied the reason why MNEs decide to go abroad and set up a new business in a completely different country. First, Coase (1937) with its idea of Transaction Costs Economies, sets the basis for the possibility of an internalization of the activities with the aim of reducing costs.

(15)

rather than international exchange, and it exists when it possesses a countervailing advantage over local firms and the market for selling such advantage is imperfect. On the other hand, the “Reading School” (Buckley and Casson, 1976, 2009; Rugman, 1981; Hennart, 1982) argues that the existence of an MNE depends on efficiency properties rather than monopolistic advantages. This means that companies want to maximize their profit by internalizing their activities rather than leaving them in the hands of intermediaries that are external to the organization.

Other scholars, such as Dunning (1977) with his eclectic paradigm, contend that FDI is influenced by three types of advantages: 1. Ownership; 2. Location; and 3. Internalization advantage. Dunning incorporates in this way different theories of international business such as Rugman’s (1980) idea of Firm Specific Advantage (FSAs) and Hennart’s (1982) internalization theory.

Another important model on the internationalization theory is the Uppsala Model by Johanson and Vahlne (1977) who see the FDI expansion of an MNE as a stage process that depends on the company’s past international experience. In other words, according to the Uppsala Model or Internationalization Theory, an MNE will expand first in countries that are more similar to their home country and once it has achieved a certain international experience, it will expand in a country which is further away and less similar to its home country.

The diffusion of the above mentioned theories led corporations to a higher internationalization, especially after the collapse of the communist bloc and with the increase in globalization from the 1990s. While the elites of newly developing countries show a consensus to the activity, local communities represent the main voice against the operations. (Calvano, 2008).

(16)

in the host country. This phenomenon has been studied by various scholars and results in different approaches in the literature; in particular, Zaheer (1995) identifies the problem as Liability of Foreignness (LoF),

The LoF depends on four main sources: 1. Costs directly associated with spatial distance; 2. Firm-specific costs related to unfamiliarity with the host location; 3. Costs resulting from the host location and 4. Costs from the home location (Zaheer, 1995).

However, with the increase of the experience of the MNE in a particular area, the LoF will decrease and, with time, the company will face less problems in adapting to the different context. Several researchers in the international business literature have shown an increase in performance in the international activity due to prior experience. (Li, 1995; Shaver, Mitchell and Yeung, 1997). Perkins (2014) points out that prior experience pays when the institutional contexts are similar.

We can conclude that multinational corporations, in order to benefit maximally from their international strategy, focus mainly on decreasing the liability of foreignness. In our study we can define the arousal of a conflict with indigenous communities as a direct consequence of the liability of foreignness, as MNEs usually do not take into account that these communities might have peculiar interests that are not considered in the strategy of the corporation, and these interests may cause frictions in pursuing the final outcome.

Setting up ad hoc departments in trying to overcome the liability is becoming a major trend, this is why companies often have a corporate social responsibility section with the goal of finding a bridge between the “capitalistic” goal and the respect of the interests of weaker parties.

In the next section we will further discuss this concept of sustainability, as we consider it an important tool to avoid friction with local communities.

(17)

Sustainability focus

The first definition of sustainability development sees the concept as a marriage, an interdependence between economy and ecology with the objective of saving the planet from its environmental problems. This concept was first developed by the United Nations Conference on Human Development in 1972, but it became more popular in 1987 with the publication of the so-called “Bruntland Commission Report” by the World Commission on Environment and Development. Our Common Future, the formal name for the Report, did not only cover environmental concerns, but also social issues; cooperation was considered the key means to overcome sustainability problems. Also, the Bruntland Commission Report for the first time was focusing on the centrality of the role of business organizations in affecting the change in the global environmental problems (Lertzman and Vrendenburg, 2005).

Although the original definition represents a big step in the field of sustainability, as it recognizes the need for different actors to participate with the goal of improving the planet, nowadays the concept of sustainable development is defined as “the development that meets the needs of the present without compromising the ability of future generations to meet their own needs” (Lertzman and Vrendenburg, 2005, p.242). In addition, from 1992, the concept of sustainable development is considered as the combination of three pillars: the economic, the environmental, and the social one. This tripartite definition is also referred as the “triple bottom line” or “capital approach”. It considers sustainability as the maintenance of three different types of capital (economic, environmental, and social) (Lehtonen, 2004). In fact, economic sustainability alone is not sufficient for the overall sustainability of a business. (Gladwin, Kennelly and Krause, 1995). For the purpose of this study, we want to narrow the focus on sustainability to the business level, rather than keeping a global framework with the ultimate goal of the satisfaction of human needs, as the previous definitions suggest. On a corporate level, sustainability has the aim to meet the needs of a company’s direct and

(18)

indirect stakeholders (Dyllick and Hockerts, 2002).

As well as in the global environment perspective, the corporate perspective of sustainable development can be based on the aforementioned three pillars and, as a consequence, on the maintenance of a sustainable capital (economic, environmental and social).

Economic capital comprises financial capital, tangible capital, and intangible capital; economic sustainable companies are those that can keep a sufficient cash flow that enables them to obtain above average returns (Dyllick and Hockerts, 2002).

Environmental or natural capital can be in two different forms: natural resources and ecosystem services (climate stabilization, water purification…). From the perspective of Ayres (1994, 1999), an industry is a living organism that consumes inputs like energy and resources to create both desired (products and services) and undesired outputs (waste emissions). If the latter are more than the former the organism is ecologically unsustainable. Thus, economically sustainable companies are those that use natural resources at a rate which is lower than the natural reproduction, those that do not cause emissions in the environment and finally those that do not engage in activity that can deteriorate eco-system services (Dyllick and Hockerts, 2002).

Regarding the social dimension of sustainable development, there is extensive literature on the topic. It is important to provide an extensive analysis of the social dimension of sustainable development as it could be a powerful tool in avoiding or reducing the risk of conflicts with communities. The need for companies to manage their social capital is not a new concept; already in the 1960s with the development of the concept of “Corporate Social Responsibility” there was an interest in the topic, but it is only in the recent years that the concept has become more popular. Lehtonen (2004), argues that the social dimension of sustainable development can be analyzed from two different perspectives: the individual capabilities one has and the social capital perspective. The former approach was first

(19)

developed by Sen (1987, 1993) who defines capabilities as alternative combinations of functioning that any human being can achieve. Ballet, Dubois, and Mahieu (2003) extend the definition on a societal level: not only individuals can possess the capabilities, but also societies: the combination of capabilities depends on the adaptation of the society’s external constraints. Socially sustainable development is one that “guarantees for both present and future generations an improvement of the capabilities of well-being (social, economic or environmental) for all, through the aspiration of equity on the one hand - as intra generational distribution of these capabilities - and their transmission across generations on the other hand.” (Ballet et al. 2003, p.6).

While the capabilities’ approach focuses more on the improvement of social conditions from one generation to another, the social capital approach sees social sustainability as the maintenance and increase of a valuable social capital. In general, social capital refers to the network of social relations characterized by trust and reciprocity, and if this network is well functioning it can improve the efficiency of the society.

Dyllick and Hockerts (2002) define two different types of social capital: human and societal. The former entails primary aspects of human beings like skills, motivation and loyalty; the latter type of social capital focuses more on the quality of public services.

Gladwin et Al. argue that a firm, to be socially sustainable has to internalize its social costs, maintain and grow the capital stock; avoid exceeding the social carrying capacities encourage structures for self-renewal; foster democracy; enlarge the range of people’s choices; and distribute resources and property rights fairly (1995, p.42).

Thus, socially sustainable companies, are those that add value to the community in which they produce or have their operations. They add value by both increasing the human and societal capital of the community. Also, socially sustainable companies aim at the satisfaction of the needs of their stakeholders and are considered by them as fair and trustworthy as they

(20)

agree with the company’s value system. (Dyllick and Hockerts, 2002)

After a brief discussion on the main actors involved in our study, we will provide a deep analysis on the concept of conflict itself, in order to unveil its dynamics and to have a better insight before running the statistical analysis.

Conflicts

Franks, Davis, Bebbington et Al.. define social conflict as the “coexistence of aspirations, interests and worldviews that cannot occur simultaneously, or that actors do not perceive as matters of simultaneous satisfaction.” (2014, p. 7576) In other words, conflicts arise when the needs and the interests of people living in the same context crush against each other as they cannot be satisfied at the same time.

Another view of conflict is the one provided by Desloges and Gauthier (1997): they argue that conflicts are crucial for the social change and for continuous creation of the society itself. Thus, they should not be entirely viewed from a negative perspective but as a dysfunctional relationship between individuals and communities, that of course should be avoided but that also can be a source for change and growth.

Regarding the nature of the conflict, Bebbington et al. (2008) argue that not all conflicts are violent or lead to a civil war between the communities and the other actors. Most of the conflicts are “socio-environmental struggles” to gain the control over a particular area with specific resources or in a strategic position for the company rather than armed civil uprisings. It is hard to understand what causes this types of conflicts: as Calvano (2008) points out, literature mainly focuses on the negative outcomes of the business activities rather than the dynamics of the conflict itself. In addition, most of the articles regarding conflicts usually present particular cases rather than taking a broader and general perspective.

Mining land use conflicts are more intense in the developing world, where the issue of who owns the land is the cause of most problems (Hilson, 2002).

(21)

Three are usually the main causes for conflicts according to the literature. Firstly, the nature of the industry: some types of industries have an environmental impact that offsets the economic benefits that it provided by operating in the area. Secondly, the mining industry in particular has a long history of human rights violation and abuse of indigenous land rights. Finally, the corruption existing in unstable political systems represents the third element that might cause social conflicts (Fontana, Sastre-Merino, Baca, 2015).

Bebbington et Al. (2008) consider the need for natural resources as one of the main reasons for the conflict: evidence shows that most of the conflict arises with MNEs that are in the extracting or mining industry, thus there has to be a correlation with the need for local resources.

Other authors also include contextual factors like the cultural distance (Lotila, 2010) as one of the causes of the conflict.

Hook and Ganguly (2000) in their study argue that there are three main theories that could explain the existence of such conflicts. The first perspective is the primordialist which underlines the importance of the original identity of groups. This ethnic identity is given by nature as individuals who are born in networks based on relationship and share the same beliefs. When the identity is threatened, ethno-nationalist awareness arises, in particular when a particular group is alienated from its homeland. The alienation can be as a result of the invasive MNEs operations that, by endangering the environment, lead communities to develop new ways of living. For example, when MNEs occupy arable land, indigenous communities have to adapt to the new situation and find a new way to provide for themselves (Hook and Ganguly, 2000). The second perspective is the constructivist that, in contrast with the primordialist, reject the hypothesis that ethnic identity is given by nature. It is rather a “social construction”, determined by human actions and choices. In particular, ethnic identity is derived from a cultural construction of descent. The descent can be social or genetic (Hook

(22)

and Ganguly, 2000). The third perspective is the ethnic competition which argues that ethnic conflicts are driven by competition for resources between ethnic elites. MNEs in the extracting and mining industry could exploit this situation by bringing wealth to the region and giving incentives to ethnic elites to compete more (Hook and Ganguly, 2000).

However, they argue that it is impossible to explain the phenomenon just from one of the three standpoints: ethnic conflicts are complex and it is hard to find a uniform cause. They are the result of “convergent casual conjunctures”. However, the scholars recognize the strong impact of governments on the conflict, rather than MNEs; the role of central governments in the conflict, rather than only of indigenous communities and companies is often overlooked in the international business literature. (Hook and Ganguly, 2000).

We can then assume that there is a third party involved in the conflict: national government and especially their institutions, thus in the next chapter we will focus on them by providing some previous literature insights.

Institutions

Institutions are the humanly devised constraints that structure political, economic and social interaction. They can be both formal (constitutions, laws, property rights) or informal (sanctions, taboos, customs, traditions and codes of conduct) (North, 1991).

Institutions arise to avoid transaction costs and uncertainty: North (1991) argues that due to the increase in long distance trade during the XV century, contracts were needed to provide safe conditions to the merchants.

Before the 1990s the role of institutions had been overlooked in the IB literature, but then scholars started to realize how important they were for the business strategy and for the internationalization process of the MNE. Peng (2002) developed the Institution-Based-View, a business theory that sees the relationship of institutions with organizations as a dynamic interaction, whose outcomes are strategic choices. Thus, institutions are fundamental for the

(23)

profitability of an MNE and it is critical to include them in a business analysis. (Gao, Murray, Kotabe, Lu, 2010)

According to the contemporary institutional theory, organizations must adapt to the rules and belief systems (formal and informal institutions) of the local environment because through

isomorphism they will earn legitimacy.

Isomorphism is a constraining process that forces one unit in a population to resemble other units that face the same set of environmental conditions (Di Maggio and Powell, 1991). The concept is fundamental because companies not only compete for resources and customers but also for institutional and political legitimacy.

Institutions can be weak or strong. By “weak” we mean those institutions that do not ensure the effective functioning of markets, for example, because of corruption or because of a weak intellectual property right enforcement (as in China). Strong institutions are those institutional arrangements that support an effective functioning of the market mechanisms, when institutions are strong their role is almost invisible. (McMillan, 2008).

Bebbington et al. (2008) find that one of the main reasons for the bad relationship between the mining industry and local communities is the lack of transparency and the political corruption: many corrupted governments allocate resources to companies that favor those politicians that are in power. Mineral rents can also feed the over-expansion of bureaucracy, and induce patronage, clientelism and graft that corrode the quality of government (Auty and Gelb, 2001).

Also, Franks et al. (2014) argue that a strong national government that fosters collaboration between MNEs and the community has a positive impact on the social development and in the management of the risk of conflict.

Institutions are still a fundamental actor in the settling disputes of the communities, and in granting their rights.

(24)

For example, in the case of drawn-out negotiations between the Haida Gwaii and the Canadian government, ‘the consensus-seeking forum was constrained by the broader institutional and political structures [in] which it remained firmly embedded’ (Takeda and Ropke, 2010: 187).

Some governments have recently strengthened the environmental and social requirements of companies in extractive industries. Australia, considered to have a really strong institutional framework, shows one of the highest percentages of indigenous population. Native title, Indigenous Land Use Agreements or combinations of both now cover 23% of Australia (NNTT, 2011).

In Australia a participation agreement exists between Argyle Diamonds and the Miriuwung, Gidja, Malgnin and Woolah peoples on whose lands an Argyle mine sits; in Canada, indigenous peoples have signed three-tiered agreements with the government, the local administration and relevant corporations, with the obligations of each partner clearly outlined and agreed upon from the outset of a development project (UN, 2007).

Reterritorialization is pronounced also in Brazil where ‘over one million km2 of protected indigenous lands already exist, serving as a barrier to ... deforestation and ... also promot[ing] cultural survival’ (Vadjunec et al., 2002: 6)

A study provided by Columbia Center on Sustainable Investment, 2014 found that those countries that set mandatory requirements for the engagement of mining companies with the local communities have a lower number of conflicts compared to those that do not have any. Despite an escalating number of settlements, indigenous claimants often maintain that community aspirations have not been addressed (Booth and Skelton, 2011).

Ikelegbe (2006) points out that one of the main sources for conflicts between indigenous communities and multinational corporations is the quality of the national government: those countries with poorer governance, where corruption and ineffective governments are often

(25)

seen as one of the main sources for the increase in the conflict. Ikelegbe points out that the strength of government institutions can impact the dynamics of the conflict itself. In fact, in contexts with low institutional quality, governments are more interested in the economic benefits for multinational corporations rather than in social programs to protect minorities (Ikelegbe,2005 and Ite, 2004). We will base our study on Ikelgbe (2005) and Ite (2004) and we will try to assess whether governments with low institutional quality are actually affecting the conflicts dynamics between MNEs and indigenous communities.

Theoretical Framework

The main research question wants to assess if there is a relationship between institutional strength and the type and length of the conflict.

The strength of institutions in a country is an important variable to take into account when doing a FDI: when institutions are stronger the LoF decreases as there is less uncertainty. Many studies point out the need to analyze the institutional strength before MNEs enter in the market. Meyer et al. (2008) underline how weak institutions impact the entry mode choice. We will start from the work of Ikelegbe (2005) who points out the weakness of national governments as one of the main sources for the increase in conflicts dynamics between MNEs and indigenous communities.

With weak institutions, MNEs have more possibility to use corruption for their interests with no respect to the voices of local communities: in other words, the benefits of corporations come first rather than the benefits of indigenous communities.

The study provided by the Columbia Center on Sustainable Investment (2014) shows that when a national government intervenes in the relationship between mining companies and communities by imposing mandatory requirements of engagement with the local population, the number of conflicts decreases.

(26)

The study only considers the mining industry, we are going to assess the relationship in a more general perspective, focusing on different companies, from different industries that are involved.

In the presence of strong institutions there should be less conflicts, or at least the government should implement some policies to prevent and protect indigenous rights.

However, in the dynamics of the conflict, not only the national government has an active role, but also MNEs and communities. We want to understand whether there exist some important characteristics of an MNE that can mitigate the possibility of a conflict even in the presence of weak institutions. Also, the community itself could influence the length and type of the conflict; whether a community is more autonomous or more isolated can be factors of important influence for the conflict.

Our framework will have five hypotheses: one for the general research question, two for the company’s characteristics as moderators and two for the indigenous communities’ characteristics.

Hypothesis 1

As many scholars in the literature point out, national governments have an active role in conflicts between indigenous communities and multinational corporations. For example, most of the good governance cases in indigenous communities show a common effort of national governments, NGOs and corporations in pursuing a common goal. (Alcantara and Nelles, 2013). Calvano (2008) argues that indigenous protection depends on a mix of different factors, among which she lists national government support: institutions are fundamental to protect and grant rights to the population, especially to minorities, but when these are not strong enough, conflict can easily arise.

In a strong institutional context, with weak corruption and law enforcement, MNEs have lower bargaining power on governments, thus it will be less probable that they will violate

(27)

human rights or just in general communities’ rights, thus conflicts should be shorter and less violent. However, in most developing countries, where there are high levels of corruption and a weak rule of law, the economic profit is always considered more important than the social aspects.

Newell (2005) states that weak institutions, are often preferred by MNEs to invest in, as the power balances in these context often shifts towards MNEs. The reason behind this, is that in countries with high levels of corruption and weak rule of law, foreign direct investment is always favored more than the protection of the rights of local communities (Narula and Dunning, 2000). Thus, our study will investigate the national government strength as main cause of the conflicts and will try to research whether there is a linear relationship between the two concepts.

An example of how strong institutions are a driver for a decrease in the conflict violence or conflicts itself, it provided by the case of the Tsilqhot’in Nation of British Columbia, where indigenous rights are protected by the national institutional context. The community has the possibility to veto any activity in its territories as it was granted by the Canadian government as a form of self-governance. The multinational company Taseko, was prohibited by the community to start its extracting activity. The study provides empirical evidence of what we expect: institutional strength must positively impact the length and violence of the conflict. On the other hand, communities that are granted some autonomy and that can rely on governments support will be less likely to protest.

We expect then shorter and less violent conflicts in countries with higher World Governance Indicator scores.

Hypothesis 1: The stronger the institutional context, the shorter and less violent the conflict between MNEs and Indigenous Communities.

(28)

MNEs’ Characteristics Hypothesis 2

In the previous part we argued that according to the contemporary institutional theory, organizations must adapt to the local institutions (formal and informal) in order to obtain legitimacy.

Mayer et Al. (2008) state that institutions have a strong impact on the final performance of the multinational corporation. Together with the firm resources and capabilities they shape the strategy that then leads to the profit. But how these institutions influence the strategy? Organizations adapt their choices to the institutional framework. When facing a new institutional framework, companies often undertake partnerships with local firms to gain experience on the informal and formal constraints of the country (Mayer et Al. 2008). Thus, experience is considered an important weapon to counterbalance the liability of newness. Perkins (2014) points out that experience is fundamental to overcome the liability, especially if the company has already experienced a similar context. Singh, Tucker, and House (1986) found that the lack of institutional support experienced by young organizations was found to be an important reason for the "liability of newness". The literature fails to provide concrete examples of cases in which experience has a positive impact in the particular context of indigenous community, however we discussed before that the community is a complex construct in which multiple actors play a role. Thus, we can consider the context of indigenous communities as a complex institutional framework with specific dynamics. If a company has already operated in a community, it means that it has developed some relationships with the locals, and thus transactions with them will be easier. In addition, if a MNE has already conducted some projects in an indigenous community and has already dealt with conflicts, it will be probably easier to manage them in the future. Also, weak institutions do not impact the company anymore, or at least they have a weaker impact on the duration

(29)

and violence of the conflict as the company can rely on the relationships with local people. On the basis of previous literature we can then conclude the following:

H2: The higher the MNE’s experience in dealing with the indigenous community, the lower

the impact of weak institutions on the type and length of the conflict will be.

Hypothesis 3

In the past years most companies had to modify their business model in order to have a focus on sustainability, in particular, social sustainability. We argued that socially sustainable companies tend to meet the interests of all their stakeholders and to add values to the community in which they operate. (Dyllick and Hockerts, 2002) When a company focuses on having a sustainable impact on the area in which its plants are located, the relationship with local people should be more peaceful than in the case in which the production does not take into account the external environment. Despite there possibly exist weak institutions in the national government that foster the exploitation of natural resources in indigenous land, when a company tries to be socially sustainable its impact on the area should only be positive. Thus, the probability of having a conflict with a local community should be lower in the case where a company is socially sustainable.

The Galore Creek project in British Columbia represents an example of how a sustainable focus of the corporation leads to a better and more sustainable management of the resources exploited and of the local communities (Fidler, 2010). The case shows how aboriginal needs can be taken into account by pursuing a combination of multilateral agreements between community, nation and corporation and, in particular, through an environmental impact assessment of the project (Fidler, 2010).

We can then argue the following on the basis of the example and on the existing literature on the topic:

(30)

H3: The higher the sustainability focus of the company, the lower the impact of weak

institutions on the type and length of the conflict will be.

Communities’ Characteristics Hypothesis 4

The third hypothesis is built on the assumption that when an indigenous community is culturally isolated, it will be harder for MNE’s to enter the market and establish its plants there. We build our assumption on the notion of distance: a fundamental concept for the International Business literature. Distance between two countries leads to friction and complexity in cross-border operations. For example, in China, where the culture is very different from Western countries, MNEs often opt for a Joint Venture as entry mode in order to learn from local companies and gradually adapt to the new environment. (Meyer et Al, 2008). The concept has been studied from various points of view: cultural distance, institutional distance, geographic distance, economic distance. There is a new trend in the literature (Berry et al.,1999; Ghemawat,2000) where distance is studied as a multidimensional construct. When a community is isolated we expect that the distance will be even higher: not only from a geographical perspective, but also from a cultural and institutional one. Thus, conflicts and frictions between the multinational and the indigenous will be stronger.

Isolation of a community generally depends on a geographical isolation of the community itself, although in some cases communities can also self-isolate themselves by raising language and cultural barriers. However, in many communities self-isolation does not work when there are bigger interests at stake: for example the Nanti community in Peru, located in a zone with vast oil and gas wealth, could not keep its isolation as its territories are currently exploited by a major international gas and oil consortium. (Montenegro and Stephens, 2006)

(31)

Montenegro and Stephens (2006) further argue that self-isolation can only be possible when there are strong governmental programs that safeguard indigenous interests.

According to the previous literature, and by translating the concept of distance on the concept of isolation we can formulate the following hypothesis.

H4: The higher the cultural isolation of the community, the higher the impact of weak

institutions on the type and length of the conflict will be.

Hypothesis 5

One of the main causes for the conflict is the inability of one of the two parties to control its own resources. (Ballard and Banks, 2003) The control over the resources can be granted, as mentioned before, from national governments. Especially the governments that have stronger institutions are more willing to grant a certain extent of autonomy to the community. If the community is not granted a certain extent of autonomy the power balance will shift towards the MNEs, and communities could feel a higher level of inequality Ballards and banks (2003).

We argued before that in many countries, mainly developed ones, like Australia or Canada, governments are trying to implement solutions that leave indigenous community a certain amount of autonomy in their jurisdiction. We mentioned the example of the Tsilqhot’in

Nation of British Columbia where, thanks to the autonomy granted by the national

government, the community has veto power on any decision on its territory. A similar example, is reported by Parlee (2015). In 2014 Canadian communities won a lawsuit against one of the major extracting oil company that was trying to obtain access on the communities’ territory.

We thus argue that, when the community benefits from a certain degree of autonomy in the decision making, conflicts will arise less.

(32)

H5: The higher the autonomy of the community, the lower the impact of weak institutions on

the type and length of the conflict will be.

Framework Model

Figure 1 presents the framework that we will use to perform our analysis.

(+) (-)

(H1)

(-) (-)

Methodology

Data Collection and Sample

The sample consists of 706 cases collected by fellow students involved in the project. The data is cross-sectional as it has been collected for a period of time.

The cases cover almost all the corners of the globe as they present conflicts in around 65 countries of the world, located in every continent.

Weak Institutions Community Isolation (H4) Community’s Autonomy (H5) MNE’s sustainability focus (H2) MNE’s experience (H3) Type and length of conflict

(33)

The companies involved in the conflicts belong to different industries: we were able to collect 120 different NAICS eight digit codes, that relate to around seven different industries that range from mining to leisure services.

The analysis that we run is on the firm level, as it assesses the relationship between MNEs and indigenous conflicts.

Dependent Variables

To assess the impact of institution on the indigenous conflict, we chose two different dependent variables that will be investigated in two different regressions.

The first dependent variable is the Length of Conflict: longer conflicts should be located in those countries with weaker institutions that do not enforce laws to protect minorities and are more keen to corruption. Length of conflict is measured in months. I decided not to code the variable into a dichotomous variable (i.e. long and short) to better capture the variance. For ongoing conflicts, the final date was set for November 2016, the last month of the data collection. Outliers were checked, in order to standardize the distribution of the variable, and were not taken into account in the analysis, reducing the sample from 706 to 668 cases. The second dependent variable is the Type of Violence, measured by a Likert scale from 1 to 7, where 1 indicates peaceful negotiations and 7 indicates a high-level of violence including deaths.

Independent Variable

To analyze the strength of institutions the World Bank Governance indicators (WGI) were used.

The WGI reports aggregate and individual governance indicators for over 200 countries and territories over the period 1996–2015, for six dimensions of governance: Voice and

(34)

Accountability, Political stability and absence of violence, Government effectiveness, Regulatory Quality, Rule of Law, Control of Corruption.

These aggregate indicators combine the views of a large number of enterprises and citizen and expert survey respondents in industrial and developing countries. They are based on over 30 individual data sources produced by a variety of survey institutes, think tanks, non-governmental organizations, international organizations, and private sector firms. (The World Bank, 2016)

They are measured on a scale from 1 to 6 where countries belonging to the 90th-100th percentile scoring 6 and those belonging to the 0-10th percentile a 1.

The variables were check for multicollinearity and unfortunately the tolerance was quite low: thus I decided to create a new independent variable which I coded as Institutional Strength, computed as the mean of the six dimensions.

Moderators

We chose four moderators for the relationship we want to analyze: two moderators refer to specific characteristics of the community while the other two refer to features of the corporation.

Community autonomy and cultural isolation were chosen to investigate whether there are intrinsic characteristics of the community that can moderate the impact of weak institutions on the conflict. Both variables are measured by a scale: community autonomy takes values from 1 to 4, with 1 meaning no autonomy at all; cultural isolation also takes values from 1 to 4 with 1 meaning high cultural isolation. The Number of Project that a company has carried in an indigenous community is a good summary of the experience of a company and is used as our third moderator. Finally, to investigate the sustainability focus we included the Corporate Social Responsibility ratings offered by CSRHUB (CSRHUB, 2016).

(35)

Control Variables

We included three control variables in our model: The Human Development Index (HDI), the GINI coefficient and the Experience of the companies in an indigenous community.

The HDI is defined for country j as follows:

𝐻𝐷𝐼𝑗 = 1 − (∑ 𝐼_𝑖𝑗 𝐾 𝑘 𝑖=1 )

With Iij being the ith indicator of human deprivation in country j. Human deprivation is calculated on the basis of life expectancy, adult literacy and the logarithm of purchasing power adjusted by the GDP per capita (McGillivray, 1991).

The GINI coefficient is a measure of statistical dispersion intended to represent the income distribution of a nation’s residents and is the most commonly used measure of inequality. A coefficient of 1 represents maximum inequality.

The experience was calculated in years: the difference between 2016 and the first time the company had a project in an indigenous community.

Method

A linear regression was chosen to investigate the impact of the independent variables on our dependent variables. A regression analysis allows us to predict the outcome of the dependent variable from multiple independent variables.

Two different regressions will be run: the first investigating the impact of the independent variable and the moderators on the Length of the conflict, while the second on the Type of violence.

For each dependent variable we will test 5 models, with the first only including the control variables, the second also including our independent variable strength and finally the last three models testing the moderating effect of the variables I chose.

(36)

Table 1 and 2 give a schematic description of the models and how each variable is added at each step.

Table 1 Model Summary wih dependent variable Length of conflict

Control Independent Moderators

GINI HDI Exp_Yrs Institutional

strength Geographical Isolation Community autonomy CSRHUB NR of projects Model 1 x x x Model 2 x x x x Model 3 x x x x x Model 4 x x x x x Model 5 x x x x x Model 6 x x x x x

Table 2 Model Summary wih dependent variable Type of Violence

Control Independent Moderators

GINI HDI Exp_Yrs Institutional strength Geographical Isolation Community autonomy CSRHUB NR of projects Model 1 x x x Model 2 x x x x Model 3 x x x x x Model 4 x x x x x Model 5 x x x x x Model 6 x x x x x

The four assumptions of the linear regression

In order to proceed with a linear regression, there are some assumptions that have to be respected. First, the model was tested for linearity, which means that the data can be summarized in a straight line (Field, 2009). A qq-plot was run for each variable taken into account for the analysis in order to test whether the variables were normally distributed. Only the variable MM and Nr. Of project were not following a normal distribution. Thus, I normalized the variables by computing two new variables as follows:

LogMM = log₁₀𝑀𝑀

(37)

The new variables obtained follow a normal distribution when plotted in a QQ-plot.

Furthermore, the QQ-plot obtained in the test do not show heteroscedasticity: which means that all the variables have the same finite variance. Homoscedasticity is the second fundamental assumption to be respected when running a linear regression.

Third, multicollinearity was checked. Multicollinearity means that two or more variables are strongly correlated, leading to non-accurate results. To test for multicollinearity, we look at the Variance Inflation Factor (VIF) (tanble 1 and 2) and the Tolerance level, provided in Table x for the dependent variable Length in Months and in Table y for the dependent variable Type of violence. The rule of thumb suggests that the VIF has to be lower than 10 and that the Tolerance does not have to be too low: in both cases there are no multicollinearity problems.

Fourth, we checked for autocorrelation with the Durbin-Watson test. For both dependent variables the Durbin Watson test takes values between 1.5 and 2.5 which, as a rule of thumb, suggest that there is no autocorrelation between the variables.

Furthermore, I checked for multivariate normal distribution, with both the Mahalanobis Distance and the Mardia’s coefficient test.

A multivariate normal distribution is a generalization of univariate normality when the variables in account are more than 2. In general, there is multivariate normality if 1) all the variables have a normal distribution; 2) all the combined distributions are normal and 3) all the linear combinations of the variables are normal. Both the tests show a multivariate distribution in our case as the plot of the Mahalanobis Distance follows a linear distribution and its mean of squares is lower than p(p+2) (where p=number of variables).

(38)

Results and Analysis

Descriptive statistics and Correlations

Table 5 reports the descriptive statistics for the variables that we are analyzing. The table shows a mean of around 100 months for the length of the conflicts, as we detected the outliers and eliminated them.

The mean for the type of violence is close to three: most of the conflicts are then peaceful or are under court action, but the tendency is towards a low degree of violence.

The mean of the GINI coefficient is 44: it is close to the average of the OECD countries (The World Bank, 2016). On the other hand, the mean of the HDI lies below the OECD countries average which is 0.8.

By looking at our independent variable we can conclude that the average of the countries investigated range between the 25th and the 75th percentile, while most of OECD countries belong to the 90th and 100th percentile.

Another variable to look at is the Corporate Social Responsibility index. Unfortunately, there is not enough data to cover all 668 cases; only half of the scores are available in our dataset. Following the descriptive statistics, the table presents the bivariate correlation coefficients. The Pearson correlation coefficient indicates the strength of linear association between the variables. The matrix shows that most of the correlations are significant. There is a negative significant correlation between the dependent variables (MM and Type_vio) and the independent variable strength suggesting that an increase in the institutional strength is correlated with a decrease in the length and in the violence of the conflict. In other words, if variable A increases, variable B decreases. The variables follow what predicted in hypothesis 1 and 6: an increase in the institutional strength would lead to shorter and less violent conflicts.

Conflicts between indigenous communities and MNEs : the impact of weak institutions

Amsterdam Business School

Conflicts between Indigenous Communities and MNEs: The

Impact of Weak Institutions

Table of Contents