Relations between multinational firms and indigenous governance : an analysis of indigenous and corporate governance mechanisms in conflict resolution

(1)

Relations between Multinational firms and

indigenous governance

An analysis of indigenous and corporate governance mechanisms in conflict resolution

Student: Fenna Eleonore Aukes Student number: 10826009

Supervisor: Dr. Ilir Haxhi Study: MSc Business Administration Second Reader: drs. Erik Dirksen Track: International Management Date: 29/06/2015

(2)

Statement of Originality

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

(3)

List of Tables

Figure 1 Framework Model ... 25

Table 1 - Summary of regression models ... 30

Table 2 - Descriptive statistics: means, standard deviations, and correlations ... 35

Table 3 - Logistic regression analysis results ... 36

Table 4 - Linear regression analysis results ... 37

(5)

Abstract

Conflict resolution is an important skill for multinational enterprises (MNEs) in preventing unnecessary risks and costs. For many firms, the identification of stakeholders plays an important role in the prevention and resolution of conflicts. Indigenous communities – connected to a history of colonialization preventing their self-governance – have gained attention recently in the media as a significant stakeholder to be considered by MNEs, especially within the resource extractive sector. However, in recent years these communities have been successfully establishing their land rights and acting as stewards of the environment which in turn brings them into conflict with multinational firms looking to exploit natural resources. This thesis builds upon previous studies of which the majority have only been qualitatively exploring the conflict but not systematically identifying their determinants by examining which multi-level governance mechanisms employed by both indigenous communities and MNEs successfully aid in resolving conflicts. We argue that both community-level and firm-level governance characteristics play an important role in resolving conflicts. Using a sample of 50 North American cases, within five resource extractive industries, we first explore the relationship between indigenous (i.e. community’s relationship to the government and their geographic isolation) and governance (i.e. previous experience of the MNE in dealing with these communities) characteristics and the length of conflict, and second we analyze the moderating effect that different types of industry may have on these relationships. While the indigenous and governance characteristics examined were not supported in the statistical analysis this study takes the first step in systematically discovering what determines conflict resolution and poses possible characteristics for consideration in future research in this field.

(6)

1. Introduction

Conflict resolution is an important skill which multinational enterprises (MNEs) must be able to employ. This involves the firm being aware of stakeholders in order to first prevent conflicts from arising, and if unavoidable, resolve conflicts. One of the best practices of prevention of conflict lies simply with the recognition of each of the parties. Conflicts are more likely to arise because of gaps in perceptions between the two parties at stake (Calvano 2008). It is more beneficial for both parties to resolve the conflict quickly, preventing it from becoming more violent. As conflicts become increasingly more violent, there is a greater risk to the firm’s performance. Stakeholder relations is a large determinant of a firm’s social performance and by neglecting such relationships, in particular with indigenous communities, firms can expose themselves to a multitude of risks. While maintaining stakeholder relations remains an important issue for firms, neglecting the concerns of indigenous communities poses many risks to a firm’s well-being. Poor relations with indigenous communities can hinder access to capital, force the imposition of new rules and regulations, lead to litigation or damage the reputation or brand of the firm.

Identity plays an important role in the recognition of Indigenous Peoples and the establishment and defense of community rights. However, identification proves problematic as there is no concrete definition of what constitutes Indigenous People. Indigenous communities are generally characterised by traditional lifestyles, customs and laws, a distinct culture from the national population and their own social organization (International Labour Organization 2014). Indigenous People are generally characterized by having a unique relationship to the environment which provides a major source of power and identity. Indigenous communities act as stewards of the environment and as such, their identity is defined around their land rights, positioning them as important stakeholders for organizations interested in developing projects among those lands. As a result of growing recognition of indigenous identity and the bargaining power that comes with this, communities have begun to exhibit forms of governance in order to establish their rights and build relationships with firms and governments. Governance mechanisms employed by indigenous communities are crucial in

(7)

providing a framework and structure in which they can establish identity, recognition, and as a result bargaining power.

However, the question still remains as to which practices and characteristics are most successful in resolving conflicts between MNEs and communities. While there have been many studies on the relationship between multinational firms and indigenous communities (Boutilier, Black 2013, Bruijn, Whiteman 2010, Calvano 2008, Crawley, Sinclair 2003, Huemer 2014, Lertzman, Vredenburg 2005, Lertzman 2009, Le Meur, Horowitz & Mennesson 2013), the majority of these studies have been qualitative in nature. These studies focus on a select number of conflicts to determine which practices work best; however, most of the previous studies have tackled this issue by exploring isolated cases in a specific context, overlooking important relationships in a broader picture across countries or industries. Previous research has qualitatively explored conflicts but not systematically identified its determinants, whereas this study examines the multi-level governance characteristics employed by both indigenous communities and MNEs that successfully aid in resolving these conflicts. Therefore, our research question is twofold: First, to what extent the characteristics of indigenous communities and MNE governance affect conflict resolution and second, to what extent is this relationship moderated by industry type?

First, we identify two factors that affect the length of conflict: Recognition and Previous Experience. Recognition plays an important role in legitimizing the stakeholder and their claims on the firm. The more legitimate the stakeholder the more likely their issues will be addressed by the MNE in order to minimize risks and costs (Alcantara, Nelles 2013a) and in turn reduce the length of conflict. This study measures recognition through Relationship to Government and Geographic Isolation. A community’s relationship to the government is crucial in determining their identity and the recognition granted by firms as a legitimate stakeholder (Alcantara, Nelles 2013a). Geographic Isolation also affects recognition as well as the community’s ability to access resources necessary to establish themselves as legitimate stakeholders (Bebbington 1999). These two characteristics are essential in a community gaining recognition from an MNE. Finally, a firm’s previous experience

(8)

with such conflicts lends itself to established practices and greater chance of successful relationships in the future (Alcantara, Nelles 2013a). This study measures previous experience through the number of previous projects of an MNE. We hypothesize that the more established the relationship with the government and greater firm experience with indigenous communities the shorter the length of conflict. Furthermore, we predict that the more geographically isolated the community is, the longer the length of conflict will be.

Second, firms that operate within the resource extractive sector are heavily dependent upon the supply of natural resources (Adamson 2014). As a result, these firms are more exposed to indigenous communities, who take responsibility for the sustainable development of the environment. Firms that ignore this stakeholder group face risks associated with capital, regulation, reputation, and the legitimacy of their project (Voss, Greenspan 2012). Within the resource extractive sector there are many different industries that operate within indigenous territory and use different practices in their relations with indigenous communities. Therefore, it would be worth examining whether different industries employ different practices in resolving conflicts.

Using a sample of 50 North American cases, within five resource extractive industries, we first explore the relationship between indigenous (i.e. community’s relationship to the government and their geographic isolation) and governance (i.e. previous experience of the MNE in dealing with these communities) characteristics and the length of conflict, and second, we analyze the moderating effect that different types of industry may have on these relationships. In this study conflict resolution is measured as duration under the assumption that longer conflicts increase the likelihood for violence (Bond 2014). The statistical analysis uses a stepwise logistic regression to measure the direct and moderating effect and show that the relationship to government, geographic isolation, and previous experience did not significantly affect the length of conflict. Each of the characteristics was tested first independently along with control variables and then with the moderator to evaluate its ability to determine whether the conflict would be long or short. Because the analysis refuted the proposed hypotheses, a second linear regression was used to measure the direct and moderating effects on

(9)

length of conflict, coded as number of months. This regression also found no significant effect on the length of conflict, confirming that the indigenous and governance characteristics examined in the analysis were not determinants of conflict resolution.

While the majority of previous studies have been qualitative in nature, this study contributes by providing a multi-level analysis as to what determines conflict resolution. Past studies in this field have examined specific cases of conflict in isolation and as a result fail to recognise important relationships that can be extended across countries or industries. There have been few studies that have taken a quantitative approach in establishing which indigenous and governance characteristics are determining factors in conflict resolution. Furthermore, most studies focus on either corporate governance mechanisms in stakeholder conflict resolution, or indigenous governance. This study incorporates both indigenous and corporate governance to see which have an effect on conflict resolution. In addition, this study will enlighten as to what communities or firms may need to employ in order to successfully resolve conflicts.

The structure of this study is as follows: First common stakeholder conflict resolution practices used by MNEs will be introduced. This will be followed by an examination into who Indigenous People are and what governance mechanisms they employ. The relationship between indigenous communities and multinational firms will then be explored. This will lend itself to the theoretical framework which sets out the hypotheses to test which indigenous and corporate governance factors affect the length of conflict. A summary of the statistical analysis and discussion of the findings, implications, and limitations will follow.

2. Literature Review

2.1 Stakeholder Conflict Resolution

The possibility of a firm engaging in a conflict with an indigenous community is often unavoidable. Furthermore, it is important for managers in firms to handle and resolve these conflicts with their stakeholders. For firms in the resource extractive sectors, indigenous communities are

(10)

considered to be one of the primary stakeholders. However, compared to other stakeholders, communities are rarely given as much priority as other stakeholders despite the fact that their opposition is quite costly and time consuming (Calvano 2008). These communities are only considered when they pose a significant threat to the firm’s survival or directly impact the bottom line. Conflicts between MNEs and communities often arise from a gap in the perceptions between each of the parties. For instance, to an MNE, communities may not appreciate the benefits of economic development that come from the development of MNEs, whereas communities may feel that MNEs are simply imposing their will (Calvano 2008). This gap in opinions only increases the likelihood of conflict between the two parties. This gap can be diminished by promoting meaningful relationships between MNEs and indigenous communities. Fostering growth and promoting these sustainable relationships relies on factors such as the quality of institutions in the country and regions, the existence of social contracts, transparency, and the quality of governance practiced (Bebbington et al. 2008). Attending to these relationships in a timely manner is important for firms as lengthier conflicts may reduce the chances for a positive resolution. Firms that effectively and efficiently manage their relationships before they deteriorate promote sustainable peace while reducing the likelihood for violent behaviour (Bond 2014).

While conflicts are generally considered to be negative experiences, firms can find positive resolution through conflict transformation. If a firm has resolved a conflict, positively or negatively, it is often considered that this issue is permanently solved, however this may not be the case and a firm can enhance its adaptability through conflict transformation (Bond 2014). This takes place by viewing the conflict as a constructive opportunity for positive change and can be accomplished by focusing upon building peaceful relationships rather than pacifying any opposition. This, in turn results in the growth and maintenance of meaningful relationships between indigenous communities and MNEs.

Effective governance practices play a large role in the ability of a firm to manage conflicts before they become too detrimental or damaging. Corporate governance enables MNEs to build

(11)

sustainable organizations that benefit from trust, reputation, investor confidence, access to capital and employee satisfaction (Jenkins, Yakovleva 2006). Corporate Social Responsibility (CSR) practices often become a manifestation of corporate governance mechanisms for resolving conflict between managers and stakeholders (Jo, Harjoto 2012). However, whereas CSR is an important link to creating relationships between MNEs and communities, it can also be used to manipulate this relationship. Firms can use corporate philanthropy to silence opposing communities, pit individuals within the community against each other, or use CSR as a greenwashing tool by claiming to promote sustainability while not following through in practice (Calvano 2008). Firms should then engage wholeheartedly in CSR as a governance mechanism to confront conflict.

One effective method of conflict resolution comes in the form of mediation. Mediation makes use of a third party who promotes cooperation between the two parties to resolve the conflict (Lampe 2001). By introducing the mediator into the conflict, power imbalances can be cancelled out as the mediator allows the other parties to better understand each other’s positions. Mediation can be a very useful tool in reducing the perception gap that often occurs between indigenous communities and multinational firms. Both non-governmental organizations (NGOs) and the government can take the role of mediator. NGOs function as an excellent mediator since MNEs view them as a more legitimate stakeholder than communities. Furthermore, these communities use NGOs to ensure that their issue is brought to light and considered legitimate. Governments can also act as mediators in the relationship between indigenous communities and multinational firms by offering a state mandated level of protection (Calvano 2008). Both NGOs and the government can work together with the conflicting parties to increase the likelihood of a successful conflict resolution.

2.2 Indigenous Peoples

Indigenous Peoples have become a salient stakeholder for MNEs, especially those operating within the resource extractive sector. As a result, it is in the best interested of MNEs to fully understand these stakeholders in order to effectively address any conflicts that may arise. The term “Indigenous” has been defined in many different ways. The United Nations Declaration on the Rights

(12)

of Indigenous Peoples (UNDRIP) shies away from a universal definition, and instead provides a description of common characteristics of Indigenous Peoples. In this description, Indigenous Peoples are those who have faced prior colonialization that removed them from their lands, territories and resources, prohibiting any form of self-governance (Cope 2012). In a similar vein, the International Labour Organization (ILO) Convention No. 169 refrains from providing a definition and instead supplies a selection of criteria to help identify these communities. Such criteria includes traditional life styles, a distinct culture from the national population, their own social organization through traditional customs and laws, and finally, practising a similar lifestyle to that prior to colonialization (International Labour Organization 2014). Communities or groups of people that exhibit the criteria set forward by the UNDRIP and ILO 169 are then considered Indigenous Peoples. Furthermore, the history and past experiences of Indigenous Peoples has substantially shaped the identity and social structures of these communities.

Many indigenous groups around the world share a history of being the subject of colonialization and as a result, the identities and livelihoods of indigenous communities were ignored by colonialists. Identity is of crucial value to indigenous communities and is a fundamental tenet in the international law on the rights of Indigenous Peoples. Crawley and Sinclair (2003) outline the different stages that aboriginal groups are generally exposed to when colonialists deny or devalue their culture. In this framework, indigenous culture is initially rejected in favour of the more dominant, colonial culture. However, when faced with a threat or external pressure, it has been observed that these groups band together to create a collective identity to balance the power (Bruijn, Whiteman 2010). The cooperation of several different groups of Indigenous Peoples working together provides a much more formidable opposition than dealing with independent groups. For instance, in the 1970s in Quebec, Canada, a hydroelectric project was proposed in James Bay. The native communities in the area consisted of highly individual groups that had little contact among themselves. All of the communities in the area were opposed to the development of the James Bay hydroelectric project and chose to form the Cree Regional Authority, an organization representing all

(13)

eight communities, as a way to promote solidarity and formal representation. By aligning the common interests of all communities, the Cree Regional Authority was able to negotiate an agreement with the governments of Canada and Quebec which explicitly stated their rights as Indigenous Peoples (Wertman 1983). This idea is extended through the concept of “glocalization” in which indigenous communities identify globally as stewards for the environment (Bruijn, Whiteman 2010). The identity of indigenous groups relies strongly on their relationship and history with their land. By emphasizing their relationship to the environment, indigenous groups create a concrete identity and, as a result, a stronger role in the stakeholder relations of multinational firms.

The identity of indigenous communities also relies on the recognition of their basic rights. While these rights stem from the use and occupation of certain areas, it is difficult to exactly define what these rights consist of. Hanson (2015) has defined it as “rights to land, rights to subsistence resources and activities, the right to self-determination and self-government, and the right to practice one’s own culture and customs including language and religion”. Although these are self-defined rights, they may or may not be recognized by the governments within the countries where these indigenous communities reside. For instance, it was not until 1982 that the federal government of Canada included Aboriginal rights into the Canadian Constitution and Charter of Rights and Freedom (Hanson 2015). Even then, these rights were not specific enough and as a result must be determined through the court system. As states, institutions, and corporations recognize the rights of Indigenous Peoples it reflects the growing recognition of indigenous political capacity and the ability to disrupt, delay, and force the abandonment of projects that go against these rights (O'Faircheallaigh 2013). The recognition of an indigenous community’s right over the land they live upon plays an important part in forming an identity, as well as validating their sense of responsibility over protecting the ecosystem.

A common theme among indigenous or aboriginal groups is the importance placed on stewardship for the environment. These groups often see themselves as guardians of nature and its resources. Indigenous communities are especially dependent on the ecosystem for their survival, and

(14)

as such, have built a relationship with the environment that promotes a sustainable way of life. The land and its resources also play an important role in defining the identity of Indigenous People. Their relationship with the ecosystem is important, not only for economic survival, but also for their social and cultural identification (O'Faircheallaigh 2013). Stewardship implies a responsibility in the conservation and sustainability of the environment (Lertzman 2009). The lifestyle practiced by indigenous communities was developed over time to be harmonious with nature, placing emphasis on using resources rationally and sustainably (Marsden 1991). The beliefs and values of these groups have been shaped by their relationship with the ecosystem, which in turn has influenced their social institutions (Lertzman 2009). Because of this intimate knowledge and their tried and tested practices, indigenous communities have an important role in the discussion with multinational firms over resource extraction (Lertzman, Vredenburg 2005). Without the participation and involvement of indigenous groups, multinational firms can face many obstacles.

2.3 Indigenous Governance

Once identity and rights have been established it is the prerogative of the indigenous communities to promote their interests and goals. When confronting MNEs, the use of governance practices can increase a community’s bargaining power and help to ensure their interests and rights are recognized. The National Centre for First Nations Governance (2009) provides a list of seventeen principles of governance practices that aid indigenous communities in exercising self-determination. These seventeen principles fall within five hierarchically ordered components: People, Land, Laws and Jurisdictions, Institutions, and Resources.

Governance begins with the people that make up the community and their strategic vision, ability to meaningfully share information and participate in the decision making. These three areas are important to ensure that all voices are heard within the community and that decisions are made in a fair and just manner. Location and recognition of land rights also plays an important factor in developing governance practices. Communities must demonstrate territorial integrity, in which they take responsibility and stewardship for the land, recognizing that they have the rights to develop the

(15)

land into sustainable economies while still respecting the spirit of the land. In order to establish these land rights, governance practices are reliant on the law and jurisdiction that these communities fall under. The rule of law states which acceptable behaviour benefits the community and who can learn to expand their traditional jurisdiction by accepting offers of delegated authority, negotiation, and self-governance. The development of indigenous institutions is also an important factor in developing governance. In order to be most effective, these institutions must be transparent and fair, become results-based in order to measure success, align with the culture and promote collaborative relationships with different governments. Finally, identifying and developing resources is an important facet of indigenous governance. This includes recognizing the potential of their human resources and allowing them to grow and develop in the community through mechanisms such as performance evaluations. Adequate financial management, accountability and reporting, as well as diversifying revenue sources reduces reliance and vulnerability of the community (National Centre for First Nations Governance 2009).

2.4 Interactions between Indigenous Peoples and Multinational Firms

Indigenous communities hold a close relationship with the environment they live upon and as a result impart a sense of responsibility to the land. Indigenous communities hold control over very valuable resources as a result of their land rights, creating the need for positive indigenous relations when firms want to pursue resource extraction in these areas. However, resulting from their reliance on the environment for cultural and life-sustaining purposes, indigenous communities are especially susceptible to marginalization and the loss of their livelihoods as a result of resource extraction (O'Faircheallaigh 2013). As the supply of natural resources becomes increasingly scarce in times of increasing demand, MNE exposure to indigenous groups increases as they seek out new and achievable sources (Adamson 2014). A successful relationship with indigenous communities incorporates the concept of social performance into the firm’s performance, expanding on the traditional measurement of success through economic terms. However, quantifying social

(16)

performance and the social risks of stakeholder relationships is considered a difficult task (Adamson 2014).

An important facet in the relationship between multinational firms and indigenous groups is the recognition of Free, Prior, and Informed Consent (FPIC). This is an internationally recognized legal framework stating that indigenous communities must be informed about oil, gas, and mining projects well in advance so as to give their approval or rejection before the operation begins (Voss, Greenspan 2012). While FPIC is not legally enforceable, it is recognized as a right that indigenous communities can choose to act upon. Firms that invest time into building relationships with indigenous communities often adopt FPIC as a best practice in order to increase the legitimacy of the project (Voss, Greenspan 2012). Without taking the time to recognize the input and opinions of indigenous communities, firms can expose themselves to unnecessary risks and costs.

The importance of good relations with indigenous communities moves well beyond maintaining stakeholder relationships. In the report on indigenous rights by Experts in Responsible Investment Solutions (EIRIS) (2009), four major areas were identified as possible risks faced by multinational firms: access to investment capital, increased regulation, litigation, and reputation risk. First, in order to gain access to financial resources, institutions often require a level of transparency and responsibility in business practices (Stefani 2009). Access to capital can be hindered by a poor reputation of indigenous relations. Furthermore, a history of inadequate relations with indigenous communities may force the government to impose new rules and regulations against the industry. Poor behaviour can also lead to litigation against the firm, and finally, damage the brand and reputation of the firm (Stefani 2009). It is clear, that there are many risks faced by organizations that do not spend the time fostering relationships with important stakeholders such as indigenous communities.

2.5 Building Relationships between Multinational Firms and Indigenous Peoples

As a result of the unavoidably close relationship between MNEs and indigenous communities, there is a greater emphasis on building sustainable relationships. In recent years, engagement with

(17)

these communities has become a political and reputational necessity that has been amplified by trends of global opposition to multinational firms, increased capacity of NGOs, social movements, and indigenous rights (Jenkins, Yakovleva 2006). In response to these issues there has been an increase in the global initiatives to promote sustainable development, CSR, and corporate community investment, especially within the resource extraction sectors (Esteves 2008). These methods are crucial for building meaningful relationships between indigenous communities and MNEs.

The responsibility for building and maintaining positive relationships with indigenous groups is often appropriated to the management level. This group is responsible for balancing the public purpose of behaving ethically while still ensuring private benefit to the organization (Sanders 2008). However, in order to build these relationships, parties must be both capable and willing to work together to enter a partnership (Alcantara, Nelles 2013b). A community’s capability to bargain and negotiate is derived from institutions that encourage autonomy and reduce detrimental intervention, as well as the availability of resources. Once capability is established, then the parties must be willing. One’s willingness can be influenced by historical context (in which colonialism plays a large role), polarizing events that can affect the relationship, respective motivation of the parties involved, and finally, the community capital (Alcantara, Nelles 2013b). Each of these factors can positively or negatively affect the capacity and willingness of either indigenous groups or multinational firms to work together.

While capacity and willingness create an important foundation in building relationships between indigenous communities and multinational firms, the parties involved must follow through with their actions. Persistence and commitment to social initiatives is an important facet of this, as stakeholders may perceive social initiatives as instrumental and lacking in sincerity (Esteves 2008). Using CSR as a social initiative is purely a short-term move and contributes very little strategic benefit to the firm. MNEs should seek strategic social investments which add value to the business while promoting sustainability (Esteves 2008). These social investments should contribute to the competitiveness of the firm rather than simply fulfill a requirement. For instance, Crawley and

(18)

Sinclair (2003) criticized Australian mining firms for simply allocating indigenous affairs under their public relations mandate instead of integrating it into the strategy of the firm. If a firm’s relations with a community is not integrated into the human resource management or strategy of the firm, they may lose out on both current and future opportunities. How well firms integrate indigenous affairs into their organization can be evaluated through four criteria: two-way learning and adaption, long-term sustainable relationships, power sharing, and placing value on indigenous communities (Crawley, Sinclair 2003). In order to truly integrate indigenous relations into the organization, management must recognize the value that they can teach and employ indigenous communities but, more importantly, also learn from them. Firms increase their ethical behaviour by sharing and gaining knowledge from indigenous groups, which can then be incorporated into the firm. Furthermore, firms should emphasize the value of building long term relationships as this will strengthen the relationship between parties by demonstrating the firm’s commitment to the aboriginal community. The distribution of power also plays an important role in integrating indigenous relations into the firm. Since most indigenous communities share a history of colonialism, firms must share power with these communities to ensure that the relationship is not conceived of as tokenistic or paternalistic (Crawley, Sinclair 2003). Finally, the firm must demonstrate that the indigenous population is not nearly a means to an end but a valuable partner and stakeholder. Through their investigation of Australian mining corporations, Crawley and Sinclair (2003) found that those organizations that developed meaningful relationships with indigenous communities were most successful in gaining access to land to develop their business.

While Crawley and Sinclair examined the specific context of Australian mining corporations and their relationship to indigenous communities, their findings can be extended to other locations and industries. Lertzman and Vredenburg (2005) undertake this task by applying the prior framework in order to examine sustainable development through cross-cultural interaction with resource extractive firms. With the growing emphasis on corporate social responsibility, an organization’s behaviour becomes unethical if it ignores its stakeholder relationships. Lertzman and Vredenburg

(19)

(2005) argue that an organization’s sustainable development is dependent on natural, social, and cultural capital. A firm’s natural capital consists of natural resources and the biophysical services that firms can provide. Social capital is derived from the trust and public opinion that an organization builds through its business. Finally, a firm’s cultural capital stems from how an organization transforms natural capital into social capital (Lertzman, Vredenburg 2005). By incorporating this viewpoint when interacting with indigenous communities, firms are more likely to act ethically by placing more emphasis on learning the indigenous culture. When a firm focuses on avoiding deficits among these sources of capital, then they are on their way to behaving ethically and developing sustainably with indigenous communities.

3. Theoretical Framework

3.1 Mechanisms to Resolve Conflicts between Multinational Firms and Indigenous Peoples While there have been many prescriptive models and frameworks proposed in the previous literature on how to successfully interact with indigenous communities, there has been a lack of analysis into what determines successful conflict resolution. Generally, studies will examine cases in isolation, for instance, Lertzman and Vredenburg (2005) study the case of sustainable forest practices in Clayoquot Sound while Bruijn and Whiteman (2010) reviewed the role of community identity in the Machiguenga tribe in Peru. While these provide useful insights into what has or has not worked in that context, to a large part it ignores important relationships in conflict resolution regardless of country or industry. It is clear that a quantitative study that uses several cases of conflict between indigenous communities and multinational firms could elucidate the most effective and successful indigenous and governance characteristics. An evaluation of the different characteristics employed by both indigenous communities and multinational firms can enlighten as to which are the most effective in resolving conflicts.

The duration of the conflict acts as a valuable measure of the effective resolution of a conflict. As the length of conflict increases, partners may become increasingly frustrated as the relationship

(20)

deteriorates and the likelihood for violence increases (Bond 2014). This in turn will only encourage conflicts to drag on and tensions to rise. In addition, longer conflicts only incur more costs to the parties involved. In the transaction cost view of stakeholder analysis, external stakeholders incur costs in order to gain recognition from firms (Mitchell, Agle & Wood 1997). Costs could then be reduced if the firm is able to recognize and prioritize salient stakeholders. This study argues that there are two factors that affect the conflict duration: Recognition and Previous Experience. Recognition concerns stakeholders such as indigenous communities and is an important factor in reducing conflict duration. The more recognition granted to a stakeholder the more legitimate their claims are to the firm, meaning the firm will prioritize these stakeholders to minimize associated costs and risks (Mitchell, Agle & Wood 1997). Greater recognition of indigenous communities should then reduce the length of conflict and will be manifested in this study by the community’s relationship to the government and their geographic isolation. Previous experience from the MNE is also an important factor in conflict duration. The greater the number of positive experiences an MNE has in dealing with indigenous communities, the more likely they are to engage again and apply any lessons learned (Alcantara, Nelles 2013a). More experience, as measured in this study by the number of previous projects an MNE has in indigenous territories, should then shorten the length of conflict. These relationships with the length of conflict will be examined in further detail below.

3.1.1 Relationship to Government and Autonomy

One major characteristic contributing to recognition in the resolution of conflicts is a community’s relationship to the government and the autonomy granted to them. Recognition of identity and self-government is crucial in determining the community’s ability to negotiate with other parties. The identity of a community can act as a buffer against corporate interests and place communities as an important stakeholder (Bruijn, Whiteman 2010). The capacity of a community to negotiate is formed by the structures that define the rules of the game and tools that are available. If a community exists in an environment that grants them the ability to govern themselves they become

(21)

more empowered (Alcantara, Nelles 2013a). This empowerment plays an important role in the recognition of communities and the authority granted to them.

The recognition of an indigenous community is important in their ability to become legitimate stakeholders to MNEs. The greater the recognition given to a community, the greater their empowerment which in turn leads to the ability to negotiate and commit to partnerships (Alcantara, Nelles 2013b). The recognition granted by the government then plays an important role in not only determining how seriously an indigenous community’s claims are taken, but also in the duration of a conflict. Communities with autonomy and identity are found to be more likely to propose and successfully negotiate partnerships (Alcantara, Nelles 2013a). This leads to the assumption that communities who are granted recognition are more successful in negotiating partnerships and reducing the length of conflict. Autonomous communities can more easily communicate their needs and issues to MNEs and in turn, resolve conflicts more quickly. If firms can easily identify these communities as legitimate stakeholders, they will also be more incentivized to resolve conflicts more quickly and reduce any costs and risks incurred by lengthy negotiations.

The increased ability to negotiate can be seen in the case of the Tsilqhot’in Nation of British Columbia, Canada. When the Canadian constitution transformed indigenous rights from common-law to a constitutional right, indigenous communities were granted the right to self-government (Alcantara, Nelles 2013a). While this gave communities the ability to govern themselves, their constitutional rights did not given them the ability to govern their lands. However, in 2014 the Supreme Court of Canada ruled that First Nations owned their traditional lands (Fine 2014). This ruling granted the Tsilqhot’in Nation the ability to veto any developments upon their land that they do not approve of, aiding in the rejection of Taseko’s New Prosperity mine (Baker 2014). The recognition granted to the Tsilqhot’in Nation allowed them to advocate for their rights and clearly communicate their issues. Although the resolution did not favour Taseko, the conflict was resolved as a result of the Tsilqhot’in Nation’s ability to successfully stand up for their right over the land. This case demonstrates how an indigenous community’s relationship to the government plays a

(22)

crucial role in determining how effectively they can act and negotiate with other parties. From this we can propose:

H1: The stronger the relationship between the government and indigenous community which grants autonomy the shorter the length of the conflict.

3.1.2 Geographic Isolation

The effectiveness of a community in resolving conflicts through recognition is also represented through the geographic isolation of a community. The more geographically isolated a community is the less access this community has to resources such as credit, skills, labour, markets, and networks of social organizations and non-governmental organizations (Bebbington 1999). Access to resources aids communities in two major ways: first, they give meaning to the community, and second, they act as a source of power. Resources give communities significance by endowing on them the ability to meaningfully engage with other parties and establish their rights and issues. In addition, access to resources is crucial to granting bargaining power to communities, enabling them to stand up for their rights (Bebbington 1999). Isolated communities have less access to these resources and as a result have more difficulty in effectively protesting against development in their territories.

The more effectively a community can advocate for their interests, the greater recognition they will receive from other parties. Access to other actors is crucial in a communities’ ability to utilize mechanisms which distribute resources (Bebbington 1999). The greater the geographic isolation of these communities, the harder it is to create relationships with other actors and the more difficult it is to establish recognition. This in turn creates complications to both indigenous communities and MNEs when trying to resolve conflicts. If a community has access to more resources and actors they will gain recognition. Once recognized, firms will see these stakeholders as contributing to unnecessary costs, increasing the urgency of resolving the conflict (Mitchell, Agle & Wood 1997). The greater the isolation of the community, the less recognition granted to them, making it more costly and difficult for the firm to engage with them which could increase the length of conflict.

(23)

Geographic isolations of communities can complicate the negotiations of a conflict. This is seen in the example of the Lutsel K’e Dene, who protested the development of a diamond mine near their community (IDEX 2006). Located in the Northwest Territories of Canada this community is inaccessible by roads the whole year round and as a result of their isolation, the Lutsel K’e Dene have little access to resources such as credit or markets for goods and services (Akaitcho Treaty 8 Tribal Corporation 2014). This in turn provided difficulties in establishing themselves as stakeholders to be considered and consulted when De Beers Canada proposed their Snap Lake Diamond mine. The geographic isolation of the community hindered their ability in the beginning of negotiations with De Beers Canada. The Lutsel K’e Dene had difficulty establishing themselves as legitimate stakeholders with De Beers Canada, lengthening the duration of the conflict. Therefore we can predict:

H2: The greater the geographic isolation of the indigenous community the longer the length of conflict.

3.1.3. Experience in Number of Previous Projects

While mechanisms such as an indigenous community’s relationship to the government and their isolation can affect the duration of the conflict, there are also factors relating to the firm that are also worth consideration. Previous experience with indigenous communities is an important factor in determining the length of conflict. First, previous experience means that there are employees who are familiar with interacting with indigenous communities. These employees become “bi-culturally literate” and are crucial to easing the interaction between parties (Lertzman, Vredenburg 2005). These bi-culturally literate employees use practices that have worked in the past and apply them to current conflicts. In addition, success in previous encounters with indigenous communities encourages firms to engage again and implement partnerships, whereas negative experiences make partners wary to try again (Alcantara, Nelles 2013a). This implies that firms which continue to engage with indigenous communities were successful in previous endeavours. Therefore, the greater the number of projects an MNE has embarked upon significantly influences their ability to resolve conflicts.

(24)

Firms are more likely to engage in a project in indigenous territory if they have previous positive experiences dealing with such communities. Previous experience contributes to conflict length in two ways. First, prior experience through projects builds knowledge within the firm of best practices which can be applied to future situations. The greater the knowledge and familiarity with indigenous communities the shorter the conflict should be. Second, the number of previous projects is a signal that prior projects were successful and firms are more likely to engage in meaningful partnerships. This also increases the familiarity and ability to successfully interact with communities, and in turn shorten any possible conflicts. Due to this previous experience, firms recognize the importance and legitimacy of these communities and as such have likely developed practices to successfully engage in partnerships, reducing the length of conflict.

This can be seen with one of the largest mining firms in the world, Rio Tinto. Their mining projects range all over the world and as a result often deal with indigenous communities over land rights. Furthermore, because of their extensive experience with such communities, Rio Tinto has developed an explicit commitment to FPIC. This was seen in practice with Rio Tinto’s Jabiluka mine in Australia, in which Rio Tinto respected the needs of the Mirrar people by waiting for their prior informed consent before developing the mine (Doyle, Whitmore 2014). Firms such as Rio Tinto are well aware of the costs associated with developing projects in indigenous territories and therefore are experienced in treating these communities as legitimate stakeholders. We can therefore predict that:

H3: The greater the experience of MNEs, as seen through number of previous projects in indigenous territories, the shorter the length of conflict.

3.2 Indigenous Relations within the Extractive Sector

As a result of the high exposure of firms in the resource extractive sector with indigenous communities there are numerous cases of conflict. The level of exposure between indigenous communities and MNEs is expected to intensify as resources become scarcer and firms must resort to operating in new areas (Adamson 2014). While organizations pursue projects on indigenous land, the communities are often faced with costs and excluded from any benefits of development. However,

(25)

firms are often faced with additional costs as their development projects are delayed or cancelled because of opposition from indigenous communities (Voss, Greenspan 2012).

As previously discussed, the duration of the conflict may be influenced by factors such as an indigenous community’s relationship to the government or geographic isolation, as well as a firm’s experience with these communities. However the question arises as to whether it is worth considering how the length of conflict may differ amongst these factors depending on the industry the firm operates within.

While studies have examined specific industries within the resource extractive sector, there has been little work done comparing best practices between industries. It is more often the case that only one industry is examined in isolation. For instance, Crawley and Sinclair (2003) studied the mining industry while Lertzman and Vredenburg (2005) examined the forestry industry. While these studies provide useful insights into which practices are successful in resolving conflicts, they do not offer any indication as to which industries have adopted the most successful practices. Oil and gas firms operate in a much different manner than mining or forestry organizations, and as such, may utilize different methods in resolving conflicts. For the purpose of this study, industries within the resource extractive sector will be categorized as mining, oil and gas, logging, electric power generation activities, as well as a category of ‘other’ in which activities such as hazardous waste disposal or support activities fall under. Firms within these different industries may employ different practices and may excel in some areas of conflict resolution compared to firms in another industry. Therefore it would be worthwhile to evaluate how the industry in which the conflict takes part moderates how indigenous and corporate governance determine conflict duration.

H4: Individual industry categories will affect length of conflict differently from each other

4. Framework Model

Figure 1 represents the framework for the analysis of duration of conflict. The variables of

(26)

Governance Mechanisms while Corporate Governance Mechanisms are characterized by Number of Previous Projects. Both Relationship to Government/Community Autonomy and Number of Previous Projects have a negative effect on the Length of Conflict. In this scenario, greater amounts of

community autonomy or experience will reduce the length of conflict. In contrast, Geographic

Isolation has a positive effect on the Length of Conflict in which greater isolation will increase the

length of conflict. Finally the effect of these variables on the length of conflict will be moderated by the industry in which the conflict takes place.

(27)

5. Data and Methods

5.1 Sample

This analysis of the relationship between indigenous and governance characteristics with length of conflict was conducted at the level of the firm. For the purpose of this study the sample was drawn from a number of conflicts between multinational firms and indigenous communities. A conflict between an MNE and indigenous community was considered as any occurrence of opposing interests that resulted in some sort of protest by the indigenous community. This study focused on conflicts arising within Canada and the United States from 2004 onwards in the resource extractive sector, including industries such as mining, oil and gas, logging, electric power generation activities, as well as activities such as hazardous waste disposal or support activities. Cases that filled the requirement of the applicable geographic area, start time of project and industry sector were examined.

5.2 Data collection

For the purpose of this study, the data collected consisted of a selection of cases that were collected and codified according to similar characteristics. The cases of conflict between indigenous communities and MNEs and were sourced from third party and non-governmental organization (NGO) groups that advocate indigenous sovereignty. These sources provided databases and reports of cases of conflict arising between MNEs and indigenous groups. This study gathered its cases from the following sources: Agreements, Treaties and Negotiated Settlements Project, Native Nations Institute Database, Canadian Business Ethics Research Network Case Studies Database, and Intercontinental Cry reports on Indigenous Struggles. Supplemental information on the specific cases were also found through media sources such as newspaper articles and UN Reports.

5.3 Method

The analysis of the individual cases of conflict was done through the codification of different variables and characteristics. Each case of conflict was analysed and the relevant information was

(28)

coded as common variables and characteristics. While there was a lack of previous quantitative studies in this field, variables were derived from qualitative studies. These studies qualitatively evaluated case studies of conflict between indigenous communities and MNEs while recommending possible factors leading to resolution. This study incorporated these factors and to quantitatively examine their relationship with the duration of conflict. Data was gathered from 50 cases of conflict between indigenous communities and multinational firms within Canada and the United States from 2004 onwards from the aforementioned databases. The relevant variables and characteristics found in these cases were extracted and coded into a table.

The data collection consisted of two phases: the pilot phase and the data collection phase. In the pilot phase five cases were selected to be codified and compared to the independent results of the other project members. Once these cases had been codified and compared, the list of variables was revised to ensure consistent coding among all members of the project and data collection was carried forward for the remaining 45 cases. The initial five cases were included with the 45 cases coded outside the pilot phase to create the dataset of 50 cases.

5.3.1 Dependent Variables

The dependent variable measured in this analysis was the length of conflict. Length of conflict can be related to the success in resolving conflicts. Longer conflicts are characterized by increased potential for violence and costs incurred to the firm (Bond 2014, Mitchell, Agle & Wood 1997). As communities gain recognition from firms, needs and issues are more easily communicated and resolved (Mitchell, Agle & Wood 1997). The length of conflict was originally coded as beginning month, beginning year, end month, and end year. To measure the length, number of months were counted from the beginning to the end of the conflict. However, for ongoing cases, number of months was counted to the end of the time frame, December 2014. A new variable was then created to classify duration of conflict as either long or short. Analysis of the length of ongoing cases showed that the median length of conflict was 20 months. Coding the length of conflict as long if duration was over

(29)

24 months would keep the significant variation in the dependent variable and the majority of ongoing conflict cases could easily be coded as either short or long.

5.3.2 Independent Variables

The independent variables considered in this analysis can be grouped into two categories: Recognition and Previous Experience. Recognition consists of the relationship to the government and the autonomy granted to the community, as well as geographic isolation of the community. Previous Experience, consists of previous MNE experience as measured through the number of previous projects.

5.3.2.1 Independent Variable 1 – Relationship to Government

The relationship to the government acts as a measure of recognition granted to indigenous communities. The more recognition and autonomy granted to a community the greater their ability to negotiate and commit to a partnership (Alcantara, Nelles 2013b). The community’s relationship to the government was coded on a scale ranging from no autonomy at all to almost full recognition of community rules by the government. Within Canada and the United States this was evaluated by whether the community was recognized by the federal government as well as if the type land rights granted to them. As stated in the first hypothesis the greater the recognition of a community, the shorter the duration of the conflict should be.

5.3.2.2 Independent Variable 2 – Geographic Isolation

Isolation affects the ability of a community to gain access to important resources which grant them bargaining power and identity (Bebbington 1999). Geographic isolation was coded as either Highly Isolated or Most Community Members Live Within Broader Municipalities with Mixed Populations. The isolation of a community was first evaluated by the geographic location of the community. In addition census data from the Aboriginal Affairs and Northern Development Canada (AANDC) community profiles, as well as the U.S. Census Bureau provided information as to where the majority of the community resided. As the second hypothesis stated, it can be expected that the

(30)

greater the geographic isolation, the longer the duration of the conflict as a result of greater difficulty in gaining access to resources.

5.3.2.3 Independent Variable 3- Previous MNE Experience

The previous experience of an MNE in resolving conflicts with indigenous communities can affect the duration of a conflict. Previous success in resolving conflicts encourages firms to engage in similar projects and teaches them useful practices that can be applied in the future (Alcantara, Nelles 2013a). Previous experience was coded as the number of projects that the firm has in indigenous territory. This was then refined into six ranges were created with intervals of 20, and the last range for greater than 100 projects. As stated in the third hypothesis duration should be shortened when MNEs have many projects which affect indigenous communities.

5.3.3 Control Variables

The analysis was controlled using variables at the community, firm, and country level. At the community level, geographic size in terms of hectares was used to control for the relationship between size and duration of conflict. The size of the firm, in terms of number of employees was used as a firm level control. Finally, average country governance was used to control for diversity between countries. The average country governance was calculated as an average of the World Bank’s 2013 figures for World Governance Indicators including: Voice & Accountability, Political Stability and Absence of Violence, Government Effectiveness, Regulatory Quality, Rule of Law, and Control of Corruption (The World Bank 2013).

5.3.4 Moderator Variable

As a moderator variable, industry was used to examine the effect industry would have on the length of conflict. In the data collection, industry was coded based on 6 figure NAICS codes. These industries were then refined into five categories: Mining, Oil and Gas, Logging, Electric Power Generation Activities, as well as a category of Other in which activities such as hazardous waste disposal or support activities fall under. Each category was studied to see whether length of conflict differed between the varying industries.

(31)

5.3.5 Methods and Analysis

The relationship between dependent and independent variables can be expressed in the equation below.

𝑌 = 𝛽₀ + 𝛽₁𝑋₁+ 𝛽₂𝑋₂+ 𝛽₃𝑋₃ + 𝜀

Here 𝑌 represents the binary length of conflict as either Short of Long. Regression coefficients are represented by 𝛽0, the intercept, and 𝛽1, 𝛽2, and 𝛽3, the slope of the line relating the dependent

variable Y, to independent variables 𝑋_𝑖,. The independent variables Relationship to Government, Geographic Isolation, and Previous Experience are represented by 𝑋₁, 𝑋_2,, and 𝑋₃ respectively. Finally, 𝜀 stands for the difference between the estimated 𝑋_𝑖 and the actual 𝑋_𝑖 (Sekaran, Bougie 2009). These relationships were further explored through a regression analysis. For this study a logistic regression was chosen. As the dependent variable is categorical by returning two outcomes, short or long, and the independent variables are both continuous and categorical a logistic regression is an appropriate measure. Regression was run in a stepwise approach to test the different effects of the variables on the length of conflict. A stepwise approach was appropriate for this study as they are recommended in situations where there is little previous research for which to base the hypotheses for testing (Field 2005). As there have been few quantitative studies in this field in which to base the hypotheses upon, a stepwise regression is an appropriate test. First control variables were included, and then each independent variable separately. After this, the moderator variable, Industry Category, was included as an interaction variable. As Industry Category was a categorical variable, each of the categories was ran along with the independent variable against the length of conflict. Table 1 shows a summary of the models ran in the regression.

(32)

6. Statistical Analysis and Results

6.1 Descriptive Statistics and Correlation

The descriptive statistics and correlations between the variables under examination can be found in Table 2. The descriptive statistics return a mean value for Relationship to Government of 3.76 showing that within Canada and the United States, communities generally had limited to partial recognition from the government. In addition, the mean for Range of Project was 4.56 implying that firms generally have experience with between 60 to 100 projects in indigenous territories. These two results are supported by the Average Country Governance value of 5.62. Average Country Governance was measured on a scale where higher values correlated to higher percentiles in terms of governance indicators. It follows suit that countries that are found in the higher percentiles of governance performance would be characterized by higher community autonomy granted to indigenous communities and become an ideal location for firms needing to do business in indigenous areas.

Table 2 also displays the correlations between variables. The Variance Inflation Factor (VIF) was used as a measurement of the possible multicollinearity between variables. Multicollinearity appears when two or more variables are strongly correlated in a regression model (Field 2005). For this analysis multicollinearity is not an issue as all VIF values returned were below 5. To test the relationships between variables a Kendall tau’s bivariate correlation was chosen. This test was chosen as the Kendall’s tau is suitable for non-parametric data from a small sample where many scores have a tied rank (Field 2005). The data used in this analysis was a sample of 50 cases, from which many of the variables were categorical. As Table 2 shows there were no significant correlations between the dependent variable, Length of Conflict, and the independent and control variables. However there was a significant correlation between Size of Community and Relationship to Government, meaning that larger community sizes were correlated with greater recognition from the government. There were also significant negative correlations between Country Governance and the Relationship to

(33)

Government, Number of Employees and Range of Project, as well as Country Governance and Size of Community.

6.2 Regression Analysis

Results of the logistic regression analysis can be found in Table 3. For this analysis a stepwise regression analysis was run to determine the relationships between the independent and dependent variables. The dependent variable, length of conflict, was coded as a dummy variable with the option of either a short or long conflict. As seen in Table 1, the regression analysis was first ran with the control variables and then each independent variable was ran. A total of seven models were run in which Hypothesis 1 was tested in Model 2, Hypothesis 2 in Model 3, and Hypothesis 3 in Model 4. The effect of the moderator in Hypothesis 4 was included in Models 5-7. To analyze the results the significance values are considered as they suggest whether the result is reliable and in turn, whether the hypothesis is supported. The beta values returned in the analysis give an impression of how important the predictor is in the model (Field 2005). Hypothesis 1 predicted that a stronger relationship between indigenous communities and the government would result in a shorter length of conflict. However, as seen in Table 2 the regression analysis returned that there was no significant relationship between the two variables, even more so when the industry moderator was included. Hypotheses 2 and 3 were also not supported by the regression analysis implying that none of the included variables have any explanatory power since none of the significance factors fall below the threshold of .05. Unlike a linear regression, the value for R2 must be estimated. In this analysis we estimate the R2 using both the Cox and Snell and Nagelkerke methods. The Cox and Snell R2 is based on the log-likelihood of a model, the log-likelihood of the original model, and the sample size. However, this estimation’s major drawback is that its upper bound never reaches 1 (Field 2005). The Nagelkerke R2 addresses this issue by adjusting the formula. The R2 values generally must be between .7 and .9 to confirm the line best fits the data. The Cox and Snell, and Nagelkerke R2 values all range from .041 to .409 showing that the line does not demonstrate goodness of fit.

(34)

It is apparent that when length of conflict is measured as a dummy variable with either a long or short conflict that there are no significant relationships with the independent variables. A second regression analysis was run in order to confirm that the independent variables had no significant effect on the length of conflict. In this regression, the dependent variable is measured in terms of magnitude instead of either long or short. This is done by measuring the length of conflicts in months from the beginning of conflict to the resolution. For ongoing conflicts, the number of months was counted from the beginning of the conflict until the end of the time period examined, December 2014. In this regression the dependent variable is continuous, therefore a linear regression was used. This was done in a similar stepwise method as the previous logistic regression (see Table 2) where first control variables were tested, and then independent variables were tested individually in sequential order. The results of this regression can be found in Table 4.

Despite the change in measure of the dependent variable, there are again no significant results in the regression analysis to support the hypotheses. We can conclude that none of the indigenous and governance characteristics are determinants of conflict length as the significance values for all of the proposed hypotheses fell above .05. Hypothesis 1 is represented in Model 2, with an R2_{of 0.078,}

showing that the model only accounts for 7.8% of variation in length of conflict. Hypotheses 2 and 3 show similar results with an R2 of 0.019 and 0.150 respectively. From this, it can be concluded that these characteristics are not good determinants in the variation in length of conflict. The addition of the moderating effect of industry category also did not significantly change the results of the regression. By including industry category as a moderator the R2_{was only 0.178, 0.140, and 0.212}

for the each of the independent variables. It is then apparent that regardless of how the dependent variable is coded, there are no significant relationships between length of conflict and relationship to government, geographic isolation, and previous MNE experience.

One conclusion that can be drawn from these results is that time is not a function in resolving conflicts between indigenous communities and multinational firms. However, there are two additional reasons why there were no significant results in the regression analysis. First, it could be

(35)

that the variables chosen for analysis were not correlated to the duration of conflict. Second, there may have been other underlying factors that are more dominant but were not considered in this analysis. If this is the case, a larger correlation matrix of all variables collected should be created to see if there are any correlations with the dependent variable, length of conflict. If none of these variables are significantly correlated, it is more than likely that there are factors that were not considered in the data collection which are influencers in the length of conflict.

As the logistic and linear regression analyses returned no significant results to support the hypotheses, a second correlation matrix was assembled. This matrix examined the relationships between the dependent variable, length of conflict, and all variables collected in the data collection phase. The results can be seen in Table 5. While the original correlation analysis returned no significant relationships between the dependent and independent variables, this second correlation analysis returned a few relationships worth further examination. Significant relationships occurred at the .05 level between length of conflict and Cultural Isolation, Experience in Months, MNE Experience in Months, and the Poverty/development of the country. Possible future research could examine these relationships in greater detail.

Relations between multinational firms and indigenous governance : an analysis of indigenous and corporate governance mechanisms in conflict resolution