The acquisition of Social License to Operate: Create trust through dialogue and receive acceptance

The acquisition of Social License to Operate

Create trust through dialogue and receive acceptance

Tor Grimsvik & Viktor Tornberg

Bachelor Programme in International Business Administration 2021

Luleå University of Technology

Department of Social Sciences, Technology and Arts

Acknowledgments

We would like to express our gratitude to the people who have made it possible for us to complete this thesis. Without you, we would not have been able to complete this journey.

We would like to extend our eternal gratitude to:

Manoj Chiba, our supervisor. Your knowledge in statistical analysis, your guidance, and your inspiration helped us complete this thesis. Thank you for being the best supervisor one could ask for.

Thank you to all the respondents from Kiruna, Gällivare, and Svappavaara who choose to answer the survey. Without your input, we would never have been able to complete the thesis.

Luleå, June 2021

Abstract

Today, the concept social license to operate (SLO) has been gaining traction among companies. If a

company wants to establish a business in a new area or keep their current one, they must cooperate with the stakeholders to acquire and maintain a SLO or be forced to shut down. In the mining and extractive (M&E) sector the negative effects on the environment and local area are so apparent it has made the industry focus on SLO. Previous research draws different conclusions on what factors impact the SLO the most and this presents a research gap to analyze.

The purpose of this research is to investigate how companies within the M&E sector need to interact with local communities to build trust and acquire a SLO. This is done by an exploration of how individuals living in or near mining operations perceive mining operations by asking how they feel about the distributional fairness, the procedural fairness, the confidence in governance, the dialogue, the trust towards the mining industry, and the acceptance level towards mining. This research is of a quantitative explanatory character and primary data was gathered from an online questionnaire distributed among two Facebook groups that is connected to Kiruna and Gällivare. A total of 190 responded and their answers were analyzed with the help of statistical techniques.

The results indicate dialogue to be an efficient way to communicate for companies towards communities.

Procedural fairness and confidence in governance lead to trust while distributional fairness did not and a company that is trusted will receive a SLO.

Keywords:

Sammanfattning

Konceptet social license to operate (SLO) har blivit populärt bland dagens företag. Om ett företag vill etablera en verksamhet i ett nytt område eller behålla sin nuvarande, måste de samarbeta med intressenterna för att erhålla eller bibehålla sin SLO, annars tvingas de stänga ner verksamheten. Inom gruvindustrin är de negativa effekterna uppenbara på miljön och närområdet och det har fått gruvindustrin att lägga större fokus på SLO. Tidigare forskning drar olika slutsatser om vilka faktorer påverkar SLO mest och det innebär att det finns ett forskningsgap att studera.

Syftet med denna uppsats är att undersöka hur företag inom gruvindustrin behöver interagera med

lokalsamhällen för att bygga tillit och erhålla en SLO. För att få reda på detta undersöks individers, som bor nära gruvdrift, attityder mot gruvdrift. Det som undersöks är rättvis fördelning, rättvis process, förtroende mot staten, dialog, tillit mot gruvindustrin och acceptansnivån av gruvdrift. Undersökningen är av

en kvantitativ och förklarade karaktär. Primära data samlades in via en online enkät som publicerades i två Facebook grupper kopplade till Kiruna och Gällivare. Totalt svarade 190 personer och deras svar

analyserades med hjälp av statistiska tester.

Resultatet indikerar att dialog är ett effektivt sätt att kommunicera för företag med lokalsamhällen. Rättvis process och förtroende mot staten hade störst påverkan på tillit medan rättvis fördelning inte hade någon påverkan. Slutligen, ett företag som lokalsamhället litar på, kommer att erhålla en SLO.

Table of Content

1. Introduction ... 1

1.1 Background ... 1

1.2 Problem Discussion ... 2

1.3 Overall Purpose ... 3

1.4 Delimitations ... 4

2. Literature Review ... 4

2.1 The Social License ... 4

2.2 Stakeholder theory ... 5

2.3 Dialogue ... 6

2.4 Distributional fairness ... 6

2.5 Confidence in governance ... 7

2.6 Procedural fairness ... 8

2.7 Trust ... 8

2.8 Acceptance & Approval ... 9

2.9 Conceptual Framework ... 9

3. Methodology ... 10

3.1 Research purpose ... 10

3.2 Research approach ... 11

3.3 Literature search ... 12

3.4 Data collection ... 12

3.4.1 The survey ... 13

3.5 Sample selection ... 14

3.6 Data Analysis... 14

3.6.1 Statistical techniques ... 14

3.7 Quality standards ... 17

3.7.1 Reliability ... 17

3.7.2 Validity ... 18

4. Data Result ... 19

4.1 Profile of respondents ... 19

4.2 Cronbach’s alpha ... 21

4.3 Pearson’s test for correlation ... 21

4.4 Multiple Regression... 22

4.5 Chi-squared test ... 24

5. Data analysis ... 25

5.1 Is dialogue an efficient communication tool? ... 25

5.2 What leads to trust? ... 26

5.3 Does trust in a company result in a SLO? ... 27

6. Conclusion ... 28

6.2 Implications for practitioners ... 30

6.3 Limitations & future research ... 30

References ... 32

Appendix ... 37

List of tables

Table 4. 2 Pearson’s correlation... 22

Table 4. 3 Adjusted R square ... 23

Table 4. 4 Coefficients and model summary ... 23

Table 4. 5 Cramer’s V ... 24

Table 4. 6 Chi-Square Tests ... 24

List of figures

Figure 4. 1 Gender distribution of respondents ... 19

Figure 4. 2 Geographical location of respondents ... 21

1

1. Introduction

This chapter presents a background that provides theories and information about corporate social responsibility and social license. Then a problem discussion will follow and be concluded into an overall purpose with a research question and the authors hypotheses.

1.1 Background

The goal of a business is to provide value and even the first businesses in ancient markets provided others with goods and services in the exchange of bartered goods, giving both parties value (Schaltegger et al., 2017). Businesses are challenged to adapt their value creation process as societal attitudes and expectations change; with overuse of natural resources the environmental impact of businesses have become more important. This has led to new disciplines of business studies such as corporate social responsibility (CSR) (Schaltegger et al., 2017). CSR has been a subject of discussion long before the emergence of the concept of globalization (Crane et al., 2008). CSR is defined as a company’s debt towards all social actors involved in achieving their economic activity, how company’s address and manages economic, social, and

environmental issues (Dănescu and Popa, 2020; Su et al., 2016). The earliest notes concerning CSR can be found in literature from the 1930s however, evidence of actively applying CSR can be traced back several centuries (Carrol, 1999; Carrol, 2008). As the social expectations of corporate behavior have changed during the years, so has the concept of CSR (Agudelo et al., 2019). CSR has gained more attention from marketing scholars around the world (Zou et al., 2019). It has become a top priority for half of the Fortune Global 500 companies, with $20 billion dollars being spent on CSR activities yearly (Lenz et al., 2017).

Engaging in CSR activities stems from stakeholder pressure on companies. The stakeholder is a group or an individual that affects or is affected by an organization. The term stakeholder is an umbrella term for

suppliers, customers, stockholders, employees, communities, political groups, governments, media and more (Freeman, 2015). Evidence about CSR stakeholder pressure can be found on the political level, at the World Economic Forums, where corporate leaders are told to have collective and company goals that must focus on sustainability (Wilburn & Wilburn, 2011).

Stakeholder engagement is a fundamental part of CSR, which involves communication practices to build and maintain this interaction. The Social license to operate (SLO) concept was built on CSR foundations and emerged as a way to build, sustain, and engage with social and local communities (Hall & Jeanneret, 2015). Also, as a counter-response to minimize opposition for company operations (Owen and Kemp, 2012). The SLO concept has been gaining traction alongside CSR (Bice et al., 2017). Bice et al (2017)

2

describes SLO as “the ongoing acceptance and approval of a [project] by local community members and other stakeholders that can affect its profitability”. Mining and extractives (M&E) companies have realized their profitability from operations can be negatively affected by stakeholders if there is no SLO, and they have thus increasingly adopted the term to minimize the associated business risks (Zhang et al., 2015). The perceived role of a SLO has protected M&E companies operating all over the world against costs related to conflict (Bice et al., 2017).

The results from not obtaining a SLO or losing it are well documented, ranging from interruptions in the mining operations to operations shutting down permanently (Zhang et al., 2015; Franks et al., 2014). To understand the SLO and what is required to achieve one, research has been conducted on what constitutes a SLO on a local level by analyzing the relationships between local communities and single mining

operations. This research has in turn evolved into examining the SLO in a larger context, within a regional or national context, to find out if only a single operation decides the social license or if entire industries can earn a SLO from the whole society (Zhang et al., 2015).

Research from Moffat & Zhang (2014) points out that trust between local communities and companies is crucial to obtain a SLO and that certain factors being performed by companies can lead to trust building.

The factors are also found to be affecting trust differently, depending on the social, political, and

economical context of where a mining operation takes place (Moffat & Zhang, 2014). Therefore, identifying what factors affect trust the most in a developed country, such as Sweden, allows companies to understand what is maintaining their SLO and what factors they need to allocate most resources to.

1.2 Problem Discussion

Because mining companies have an apparent effect on the environment acquiring a SLO is not the only challenge but maintaining it so it won’t be lost is just as challenging (Zhang et al., 2015). Prno (2013) states that SLO is built on relationships, but how much community support that is required to acquire a SLO remains unsolved. Despite mining companies gaining political support, regulatory approval and focusing on CSR, they are still being questioned by the local communities and conflicts occur (Bice et al., 2017). If companies do not adhere to the requirements for a SLO, they will lose it and the right to conduct business in that area. Companies can also face further backlash from its stakeholders and the image of the companies can be tarnished as a result of losing the SLO (Demuijnck & Fasterling, 2016).

3

According to Davis and Frank (2014) when relatively small community concerns are being ignored by the companies, the community can decide to escalate the situation to help draw attention to their concerns.

Making early communication with the ones raising the concerns vital (Davis & Frank, 2014). Effective communication between companies and communities is necessary to establish a successful relationship between the two parties. Successful relationship communications have distinctively higher communication quality, information sharing between the parties, and higher participation in planning among both parts (Mohr & Spekman, 1994). Dialogue is a key mechanism in a relationship which strives towards mutual learning. Dialogue is therefore affecting the process of attaining and maintaining a SLO (Mercer-Mapstone et al, 2017).

According to Walton et al (2013) dialogue between company and community is seen as a mechanism for building trust and acquiring, while also maintaining, a SLO. Moffat and Zhang (2014) argue that community trust is one of the core factors of getting acceptance and acquiring a SLO and that contact quantity, contact quality and procedural fairness are some of the factors that affect community members’ trust. According to research by Zhang et al (2015), distributional fairness, procedural fairness, and confidence in governance were found to together affect the level of trust within the mining industry.

While findings from Zhang et al (2015) concludes that the combination of all three factors have the same weight when trying to acquire a SLO, Moffat and Zhang (2014) come to another conclusion, their factors affect trust differently and showed procedural fairness was the strongest predictor of trust.

1.3 Overall Purpose

The purpose of this thesis is to investigate how companies within the M&E sector need to interact with local communities to build trust and acquire a SLO. To be able to answer the overall purpose, one research

question is listed below:

RQ1: What does companies within the M&E sector need to communicate with local communities to acquire a SLO?

To be able to answer the research question, three hypotheses will be tested.

Hypothesis 1: Is dialogue an efficient way for companies within the M&E sector to communicate with local communities?

4

Hypothesis 2: Can distributional fairness, procedural fairness, and confidence in governance lead to trust in a company?

Hypothesis 3: Does trust in a company result in a SLO?

1.4 Delimitations

This study will focus on the mining communities surrounding Kiruna, Gällivare, and Svappavaara. No other mining communities in Sweden is included in the sample. These delimitations are due to time constraints and lack of resources.

2. Literature Review

In this chapter the main theories and authors this thesis will emanate from are presented. Firstly, an introduction to the social license. Thereafter, an overview of stakeholder theory. Then the importance of dialogue will be explained, followed by trust and how it is affected by distributional fairness, confidence in governance, and procedural fairness. Lastly approval and acceptance is presented followed by a conceptual framework.

2.1 The Social License

According to Boutilier & Thomson (2020), the social license is defined as the ongoing approval of a local community and other stakeholders to a project. The social license stems from the beliefs, perceptions, and opinions carried by the local community and other stakeholders about the project and thus the social license is granted by the community. It is dynamic and ever-changing due to beliefs, opinions, and perceptions changing when new information comes into play. Therefore, the social license must be earned and

continuously maintained. The granting of a social license is usually site-specific, meaning a company might have a social license for one project but not another. It gets more difficult to acquire a social license if social, economic, and environmental impacts from a project are substantially large, therefore, opening a factory close to a village can be a hard challenge to face as a company (Boutilier & Thomson, 2020).

Due to the social license being based on the levels of expectations a community or the wider society has of a company and its activities rather than legal requirements, officials in companies express how potent social demands could be. An environmental manager at an American mill mentioned that: “The Environmental Protection Agency (EPA) is such a large organization that it takes decades for it to make something happen.

Rather, it’s the community that dictates the term as they are doing much more to keep companies

5

accountable and as such, it gives me an incentive to go to management and tell them we need to do better or the community will cripple the company” (Gunningham et al., 2004).

To be able to enforce the social license, there are a plethora of options to use. Firstly, companies are worried about the preservation of their “reputation capital,” which the local community can drain through critique or grow by approval. These “punishments” and “rewards” can make companies go beyond their regular

obligations and make them take “good citizenship” measures, which would be defined as expenditures not justified based on quantitative analyses of profit but, rather, measured on grounds that enhances a

company’s reputation which in the short or long term can be considered as “good business” (Gunningham et al., 2004). Prno (2013) confirms this by stating a social license to operate is built on relationships.

Secondly, the demands for a social license granting can be coerced through augmentation of legal

enforcement mechanisms. By pressuring regulators to enforce the current regulatory and legal agreement, it adds weight to the existing regulatory and legal requirements and pushes the minimum expected

requirements towards full legal compliance and even makes companies invest into measures beyond compliance to provide a large margin of errors towards violations (McClelland & Horowitz, 1999). This could be a reduction in the amount of a specific pollutant used in a process, well beyond the set limit in a regulatory permit. Thirdly, if social demands are not met, they can be translated into new legal requirements that companies have to follow. And finally, the social demands can be enforced via the market by for

example customer boycotts (Gunningham et al., 2004).

2.2 Stakeholder theory

Stakeholder theory suggests managers of firms have wider obligations towards its stakeholders more so than traditional economic theory has assumed. Stakeholder theory has, due to continuous research on the subject, proceeded along four lines of senses. Firstly, the instrumental stakeholder theory, which deals with the objective function of firms and how firm managers wishing to maximize said objective function, has to take stakeholder interests into consideration. Secondly, the descriptive research on how managers, stakeholders, and firms interact with one another. Thirdly, the normative sense of stakeholder theory which details what managers should do face to face with stakeholders. Fourthly, the metaphorical use of stakeholder, meaning the depiction of the idea as a figure in the corporate life narrative. The first two lines combined is the analytical approach of stakeholder theory while the two last is the narrative approach of stakeholder theory (Freeman, 2015).

6

The number of stakeholders can make it difficult for a company to discern which stakeholders to turn towards to get consent for a SLO, but Wilburn & Wilburn (2011) provides a way by dividing up all possible stakeholders into two categories, vested and non-vested stakeholders. Vested stakeholders have a voice at the table and have a vote in the awarding of a SLO to companies. Non-vested stakeholders do not have a vote in the awarding of a SLO, but they still have a voice which they can use to influence the vested stakeholders SLO awarding decision (Wilburn & Wilburn, 2011).

2.3 Dialogue

Dialogue has been presented as a key factor for obtaining and sustaining a SLO and failure to engage stakeholders in meaningful ways will result in high costs due to social conflict (Mercer-Mapstone et al., 2017). Mercer-Mapstone et al (2017) states meaningful dialogue as core to stakeholder engagement practices. Mercer-Mapstone et al (2017) shows in their study that there has been a shift in how to build relationships with stakeholders in recent years. From a transactional approach toward an approach focused on “doing with” stakeholders instead of “doing to”, a more collaborative and involving relationship (Mercer-Mapstone et al., 2017). Relational aspects of social license such as the quality of contact between the company and community are important and driving factors of achieving a SLO (Mercer-Mapstone et al., 2017). Mercer-Mapstone et al (2017) implies that key relational factors such as trust can be developed through different mechanisms and dialogue is one of them. Moffat and Zhang (2014) states that the way companies engage and treat communities will affect the trust communities have towards companies.

2.4 Distributional fairness

Moffat and Zhang (2014) states that distributional fairness influence on trust and acquiring a SLO is an interesting area to examine. Zhang et al (2015) describes distributional fairness as “fairness in terms of the outcome of resource allocation”. Distributional fairness and its importance in creating social relations, decision acceptance and cooperation has been documented. Research in distributional fairness has its focus on outcome allocations being fair and individuals’ reaction to what a specific relationship gives them.

People who believe they receive a fair share show considerably more satisfaction and in situations where distribution seems unfair, they reject it (Zhang et al., 2015). Zhang et al (2015) explains that if the people believe there is a high degree of distributional fairness, the benefits from the company’s activity is being distributed equitably, the trust in the company is also high. What is viewed as fair distribution also differs from several societies. Individualistic societies perceive fair distribution as allocation based on merit, in

7

other words, distribution based on the amount of work put in. On the other hand, collectivistic societies view fair distribution as an equal share among the group, as they prioritize the groups wellbeing over the

individual’s wellbeing. As such, people of a collectivistic society strive towards equal share of the benefits while the individualistic society proportionally distributes the benefits (Roy et al., 2018).

2.5 Confidence in governance

Governments play a major role in regulating industries and how company activities should be conducted within a government’s jurisdiction. This is especially prevalent within the M&E industries as frameworks have been developed to meet development objectives and manage social, economic, and environmental impacts. The frameworks ensure the responsibilities of the mining companies to produce the necessary conditions to live up to the development goals (Zhang et al., 2015).

As governments try to manage industries’ social and environmental impacts through regulation and legislation, they also seek to create a favorable environment for industries to operate in due to their significant contribution to society in the form of government revenue and employment. This creates a

delicate balance where the government must achieve both goals without compromising the other. Due to this complexity, governments are heavily monitored by the public and interest groups to make sure the

government holds industries accountable or else be met with strong public objection (Zhang et al., 2015).

For the public, regulation and legislation is seen as the main mechanisms to manage social and environmental impacts from industries. If the public’s perception is that there is no proper regulation towards certain industries, then the public’s attitude towards those industries is less favorable. Research points out that among the citizens who believe in strong regulation and legislation to keep industries accountable, in other words strong governance, there is an increased probability that they will accept such industries. As such, the confidence in governance will directly impact to what extent communities will accept industries (Zhang et al., 2015). Zhang et al (2015) found that lower confidence in governance resulted in decreasing trust in companies, and vice versa. A higher confidence in governance arrangements means they believe they will manage the operating companies’ social and environmental impacts better, which results in higher trust in the companies (Zhang et al., 2015).

Evidence of weak governance affecting the SLO has been pointed out by Cruz et al (2020) where

politicians’ lackluster management of royalties and taxes related to mining created social discontent among

8

the local community in the Canaã dos Carajás municipality. Zhang et al (2018) points out that companies that communicate that they will adhere to government regulations and guidelines when applying for a SLO, significantly increased the local community’s confidence in governance and in turn increased the likelihood of a SLO acquisition.

2.6 Procedural fairness

A company’s decision procedures and the community’s level of participation in the procedures will affect the community’s acceptance of the company (Moffat & Zhang, 2014). Moffat and Zang (2014) describes procedural fairness as “whether individuals perceive that they have had a reasonable voice in a decision- making process.”. When communities feel that the decision makers have treated them with respect, and they are participating in decision making processes they will regard the whole procedure as fair which will lead to trust (Moffat & Zhang, 2014; Wright & Bice, 2017). This is supported in a study by Cremer et al (2005) which found that communities who believe a company’s procedures are fair will trust it and cooperation will be easier. This correlation means that a high level of perceived procedural fairness by the community will lead to a higher level of trust and vice versa. This is strengthened by Besley (2010) that suggests people who participate in fair decision-making processes will more likely accept the outcome, no matter who favors the outcome.

2.7 Trust

Trust is defined as having credence that the behavior of a person or group will match the presumptions of the trust holder. In addition, trusting someone or a group is to presume they will look to collaborate and not take advantage of one’s vulnerability. If looking at it through a social license perspective, public trust

represents the collective amount of trust the general public as a group holds towards an organization (Moffat

& Zhang, 2014).

Several types of trust exist within the professional relationship sphere and knowledge-based trust is one of them. It is based on the predictability of behavior - an assessment of the probability of the other’s expected choice of behavior. This form of trust transpires when one has sufficient information about others to understand them and correctly anticipate their expected behavior. As such, there is an understanding of the decision maker’s actions, thoughts, and intentions well enough to predict their actions (Lewicki & Bunker, 1994).

9

Trust can be seen as a psychological state where a company and the community rely on each other and they both expect the other’s intentions and behavior to be positive (Rousseau et al., 1998). Trust is a crucial element needed in the acquisition of a SLO, due to how it is based on the perceptions of impact on social infrastructure, dialogue quantity and quality, the view of procedural fairness, distributional fairness and confidence in governance. The trust mechanism relates in turn to the acceptance and approval of a project/operation (Moffat & Zhang, 2014; Zhang et al., 2015). Billing et al (2021) noted that community opinions on whether to trust given information was heavily affected by who provided the information. The local community considered information, evidence, and advice provided by science as being biased and in favor of industry as the very same industry had likely requested it.

2.8 Acceptance & Approval

When a community perceives that a company lives up to their expectations and that trust has been established between the community and the company, the company has gained either acceptance or approval for their project/operation. Without acceptance or approval, a company cannot acquire their SLO (Boutilier & Thomson, 2020). Acceptance from a community can be viewed differently, either a community does not actively oppose a company operation, or it actively shows its support for an operation. As such, acceptance is viewed, in this case, as the same as “non-opposition” (Diantini et al., 2020). Noteworthy, studies point out that some communities do not express their discontent for operations due to it being viewed as culturally inappropriate and therefore companies need to understand that a lack of open conflict may not mean they have the acceptance of the community (Diantini et al., 2020).

2.9 Conceptual Framework

A conceptual framework has been made to help illustrate how perceptions about distributional fairness, procedural fairness, and confidence in governance influence trust which will lead to a social license to operate. It also illustrates that dialogue is a key mechanism which relational factors that affects trust can be developed through, by communicating with the community.

10 Figure 2. 1 The conceptual framework

3. Methodology

In this chapter research purpose, approach and strategy will be presented. The method of data collection will also be presented followed up by the statistical techniques used as well as the validity and reliability of the study.

3.1 Research purpose

The research purpose can be classified into three possible categories: descriptive, exploratory, and

explanatory. They can either be used separately or combined to answer the purpose of the study. Descriptive research aims to describe phenomena and provides a basic template on how we think the world is. It does not try to understand why a certain type of behavior is the case. Exploratory research aims to investigate a phenomena or problem that there is little prior knowledge about through descriptions of social systems or relationships between events. Lastly, explanatory research describes a phenomenon and tries to understand

11

why a certain type of behavior is the case. Explanatory tries to analyze relationships between variables and form them into general statements, attempts to build, test or revise a theory or explain social relations or events (Adams et al., 2014).

The purpose of this study is to examine if dialogue is an efficient way to communicate distributional fairness, procedural fairness, and confidence in governance and how these factors influence the

communities trust in companies, and if trust leads to the acquisition of a SLO. Due to the purpose of the study, where the goal is to find and explain relationships between variables by analyzing quantitative data, an explanatory approach will be used.

The research purpose is of an explanatory character as it tries to explain what factors affects a community’s trust towards companies as well as the relationship between trust and SLO, and the relationship between dialogue and procedural fairness, distributional fairness, and confidence in governance. The research question is answered by breaking it down into three hypotheses, which aims to explain the relationships between the variables. Descriptive statistics was used to describe the population in order to provide further insight into the gathered data prior to the explanatory analysis. Therefore, the study combines an

explanatory and descriptive research purpose.

3.2 Research approach

Depending on the research purpose, the research approach can vary between either inductive or deductive.

The inductive approach is exploratory and focused on gathering data and, from that data, trying to build new theories and constructs to explain the data collected. The starting point with an inductive approach is the data and not a predetermined theoretical framework. The deductive approach has its starting point in literature and seeks to define hypotheses from existing theoretical frameworks, therefore the deductive approach demands more structure when gathering data. When trying to test a hypothesis different variables are often measured and relationships between the variables are often sought to be explained (David &

Sutton, 2016).

When collecting data different approaches can be used, quantitative or qualitative. Qualitative data focuses on the meaning of words and is often used along with the inductive approach. Qualitative studies often use interviews with small sample sizes and participants. Quantitative data on the other hand fits a deductive approach and is based on meanings derived from numbers, it aims to measure variables while being

12

objective to the hypotheses (David & Sutton, 2016). Quantitative data collection enables gathering a larger amount of data which facilitates statistical analysis (Saunders et al., 2006). A non-experimental quantitative method gathers data on different variables with the purpose to examine the possible relationships and connections they have. Quantitative data is often gathered through questionnaires or structured interviews (David & Sutton, 2016).

This study’s conceptual framework is built upon the social license to operate and how it is affected by trust, which variables affect trust, and how those variables can be communicated. Since this study seeks to

describe and explain the relationships surrounding SLO, a deductive approach is most suited to answer the research question. Based on the choice of a deductive research approach and an explanatory purpose, a quantitative data collection is the most suitable approach. This study seeks to gather large amounts of data in order to execute statistical analysis, and quantitative data collections enables it (Saunder et al., 2006).

3.3 Literature search

The literature search has been conducted primarily through Google Scholar and the LTU Library database.

Certain criteria have been used for the literature sources, such as peer review and originating from high rated journals.

Specific keywords have been used to find literature and they are: Social license to operate, Dialogue, Stakeholders, Trust, Distributional fairness, Procedural fairness, Confidence in governance.

3.4 Data collection

A data collection method was chosen that suited the research purpose and the research strategy, which was a survey. Surveys have the potential to gather enormous amounts of quantitative data in a potentially short amount of time while being cost efficient (David & Sutton, 2016). A survey can either be distributed directly, through email, or indirectly via platforms like social media. A direct distribution can decrease the chance of missing out on possible respondents but due to Covid-19, it will not be possible to do at this time.

As such, the indirect distribution was chosen, and the survey was published on the social media Facebook.

Furthermore, it was published on the Facebook groups “Kiruna” and “Händer i Gällivare Kommun”. The survey could be responded to for one week and the authors reposted the survey every second to third day to make sure it was the first post people saw when they accessed their Facebook group. This was done to increase potential respondent frequency.

13

The survey was created with the help of Google Surveys, a tool provided by Google that can create surveys and analyze the data being collected from them. The upcoming section will more thoroughly describe the outline of the survey.

3.4.1 The survey

The survey is based on a previous survey by Zhang et al (2015) with some newly added questions from Moffat & Zhang (2014) and questions regarding age, gender, and location added by the thesis authors. The layout of the questionnaire has been considered to bring about a sense of purpose to the respondent which in turn can increase the likelihood of the respondent completing the survey. If not structuring the layout in such a way, it can cause the opposite reaction, where the respondent does not feel obliged to finish the

questionnaire (David & Sutton, 2016).

An introduction is provided to the survey which states the reason behind the survey, who created the survey, that respondents will remain anonymous, the number of questions, the approximate time it will take to complete, and contact info to the survey creators.

The survey consists of 17 questions and all responses, except the age, gender, and location response, were provided on seven-point and five-point Likert scales ranging from:

• 1 (strongly disagree) to 7 (strongly agree)

• 1 (not at all) to 5 (very much so)

• 1 (none at all) to 5 (a great deal)

• 1 (very unpleasant) to 5 (very pleasant)

• 1 (very negative) to 5 (very positive)

Last question was a Likert scale as well, with the exception that 1 to 5 were coded as:

1. Reject mining 2. Tolerate mining 3. Accept mining 4. Approve mining 5. Embrace mining

14

The questions adopted from Zhang et al (2015) and Moffat & Zhang (2014) were originally in English but have been translated to Swedish. To make sure the core message was transitioned properly from English to Swedish in the translation, the authors consulted a group of swedes with academic background and a tryout group. Both gave feedback on the translation and if they understood the questions. The questions in their full entirety can be found in appendix IV and V.

3.5 Sample selection

The purpose of this thesis is to investigate how companies within the M&E sector need to interact with local communities to build trust and acquire a SLO.. Since an online survey has been chosen as the data collection method it is important to choose a sample size which can help us answer our research purpose. Since there are both practical limitations, such as Covid-19 not allowing on site collection of responses to take place to complement the online survey, and a time limit, the sample size has been limited (David & Sutton, 2016).

Our targeted population are Swedish individuals living in, or near a mining community and the sample has been limited to individuals 18 years or older living in, or near, either Kiruna, Gällivare or Svappavaara where they have open pit mines or underground mining operations. The amount of people living in Kiruna, Gällivare, and Svappavaara are 22 841, 17 511, and 417 respectively (SCB, 2020). The total sample size amount is 40 769. Since the survey was conducted online through different Facebook groups which are connected to Kiruna, Gällivare, and Svappavaara and everyone who saw it could choose to participate, it is a non-probability convenience sampling (David & Sutton, 2016).

3.6 Data Analysis

One can analyze quantitative data in two possible ways according to David & Sutton (2016), by either using a descriptive approach to interpret and understand the data or using statistical tests to analyze the

hypotheses. The analysis of this thesis uses statistical tests in order to comprehend the gathered data and test the hypotheses. The analysis was conducted by using the SPSS program and it was used to test correlations and the reliability of variables. The specific methods used for data analysis will be expanded upon further.

3.6.1 Statistical techniques

Descriptive statistics were used to determine some of the respondents’ demographic profiles. This research sought to understand the gender, age, and location distribution of the respondents. The survey consisted of 17 questions and three of those questions were to determine their demographic profiles. Two questions each

15

were used in order to gather the data surrounding distributional fairness, procedural fairness, and confidence in governance. The data for dialogue were gathered with four questions, two that measured the contact quantity and two which measured the contact quality. Three questions were used when gathering the empiric data for trust and one question was used to measure the acceptance level of mining. The data within each category has been added together and then divided by the number of questions, resulting in the mean being used when all the statistical tests have been done. The questionnaire showing all the questions can be found in the appendix. To be able to certify the collected data and to test the hypotheses, some statistical

techniques were chosen and used for this research. The following statistical techniques were used:

Cronbach’s Alpha

Cronbach alpha measures the internal consistency of a scale or test. It measures how

well all the data measures the same concept or construct, and it ranges from 0 to 1. A value closer to 0 can be because of too few questions or a heterogeneous construct. A value to close to 1 can be interpreted as items testing the same variable and that might be unnecessary (Tavalok & Dennick, 2011). According to Tavalok and Dennick (2011) a Cronbach Alpha score between 0.7 and 0.95 is acceptable.

Pearson Product-Moment Correlation

This research used a Pearson product-moment correlation (Pearson’s correlation), and it is a

statistical technique that measures the correlation between two or more quantitative variables, it tests the association between variables (David & Sutton, 2016). When performing a Pearson test the Pearson’s correlation indicates the strength and the direction of the correlation between the variables.

The Pearson’s correlation ranges from -1 to +1 where -1 indicates a negative correlation between the variables whereas +1 indicates a positive correlation among them and closer to 0 indicates there is no correlation. The p is a measure of significance and if it shows less than 0.05 it means that there is a less than 5% chance of the found correlation to have happened by coincidence and the test is statistically significant (David & Sutton, 2016).

Hypothesis 1 is “Is dialogue and efficient way for companies within the M&E sector to

communicate with local communities?” therefore this study seeks to find and measure the strength of the relationship between dialogue and three variables, distributional fairness, procedural fairness, and confidence in governance. As mentioned earlier, Pearson’s correlation excels at measuring correlation between two or more quantitative variables, and thus it was used to test Hypothesis 1.

16 Multiple Regression

Multiple regression analysis is used to predict an estimated value of a dependent variable based on the value of two or more independent variables. It also determines the level of impact each

independent variable have on the dependent variable (David & Sutton, 2016). When performing a multiple regression test, the R value measures the quality of the prediction of the dependent variable.

The value ranges from 0 to 1, where the higher the value is of R, the better the independent variables are at predicting the dependent variable. The R Square value tells the proportion of variance in the dependent variable that can be explained by the independent variables. The p value checks if the proposed model is a good fit for the data, if p value is less than 0,05, then the proposed model is a good fit for the data. If p is greater than 0,05 then the proposed model is a bad fit for the data. The t value indicates the size of the difference relative to the variation in the sample data and the greater the value of t, the greater the evidence against the null hypothesis (David & Sutton, 2016).

Hypothesis 2 is “Can distributional fairness, procedural fairness, and confidence in governance lead to trust in a company?” therefore this study seeks to determine what level of impact the independent variables, distributional fairness, procedural fairness, and confidence in governance have on the dependent variable trust. Multiple regression analysis excels at finding out if independent variables impact a dependent variable and to what extent. Therefore, multiple regression was used to test Hypothesis 2.

Chi-Square

Chi-square test determines if there is enough evidence from the sample data to conclude that one, with statistical assurance, can assume there is or there is no relation between two variables (David &

Sutton, 2016). If the number of expected counts and the actual real counts are not the same there is a reason to believe they are associated (Saunders et al., 2006). When trying to establish if the test is statistically significant or not both Pearson Chi-square and Fishers exact test can be used however, they suit different crosstabulations. Pearson Chi-Square can be used when the crosstabulation is bigger than 2x2 while Fishers exact test can be used when the crosstabulation is no bigger than 2x2, both tells whether the chi-square test is statistically significant (Saunders et al., 2006). A p value lower than 0.05 in the “Asymptotic Significance (2-sided)” square tells whether the test is

statistically significant or not (David & Sutton, 2016). Phi and Cramer’s V shows the strength of the association. Cramer’s V is applicable when the table is bigger than 2x2 while the Phi is only suitable

17

for 2x2 tables (Saunders et al., 2006). Cramer’s V tells how strong the association between the variables are. Cramer’s V is a number between 0 and 1, 1 means there is a strong correlation while 0 means there is no correlation between the variables (Saunders et al., 2006).

Hypothesis 3 is “Does trust in a company result in a SLO?” hence this study seeks to prove if there is, or there is not, a relation between two variables, trust and acceptance. Chi-square proves with statistical assurance if there is a relation or not between two variables. Therefore, Chi-square was used to test Hypothesis 3. Due to this study’s variables having more than two categories, a crosstabulation bigger than 2x2, the Pearson Chi-Square was used instead of Fisher’s exact test.

Cramer’s V was the measurement of choice instead of Phi due to the crosstabulation being larger than 2x2.

3.7 Quality standards

3.7.1 Reliability

The reliability of research concerns to what extent a research will give consistent results, in other words, if a similar research can be replicated by other researchers and yield alike results. To be able to replicate the study, the number of errors and biases needs to be kept at a minimum level (David & Sutton, 2016). To make sure others can repeat the study, an appendix of the questionnaire will be available, both in English and Swedish. Using the same questions in a future study increases the probability of obtaining equivalent results.

To test the reliability of the questionnaire, a retest reliability can be performed. It would mean, in this case, the same people who responded to the questionnaire will do it again in the near future. The problem with this method is that firstly, the respondents were anonymous, so it will be difficult to reconstruct the original population sample. Secondly, the respondents might still remember what they answered the last time, and the retest will then measure the respondent’s memory rather than the reliability (David & Sutton, 2016). The translation of the original English questions to Swedish had a chance to affect the reliability of the study but, the English proficiency of both authors were great, and feedback was provided from test sample groups about the translation so it would match the English questions.

18

3.7.2 Validity

The validity of research is concerned about whether the study measures what it is supposed to, if one can trust the findings to be about what they state. Validity can be divided into various aspects where the most suitable for quantitative studies are construct validity, internal validity, and external validity (David &

Sutton, 2016; Riege, 2003).

Construct validity concerns how well the scale measures what is intended, if it accounts for the different dimensions of what it seeks to explain (David & Sutton, 2016). According to Riege (2003) there are three different techniques to increase construct validity, the use of multiple sources in the data collection phase which will help against research bias, a well linked data collection phase, and lastly letting colleagues review the studies draft and changing it according to their feedback. Internal validity concerns whether the studies variables affect one another as it states, or if there are other, non-accounted for or accounted for, variables that are responsible (Riege, 2003; David & Sutton, 2016). External validity refers to what extent the findings in the study can be generalized and applicable to the broader population in different

environments. It is heavily affected by the sample size used (David & Sutton, 2016).

This study’s conceptual framework is built upon previous research such as Moffat and Zhang (2014) and Zhang et al (2015) that has been used and tested before. The literature review consists of several sources of information with secondary data that can be used along the gathered primary data. The drafts have been reviewed by our colleagues and supervisor as well as the survey that has been sent out to gather the data. A proper sample size has been chosen and the considerable number of respondents helps generalize the findings of this study.

19

4. Data Result

In this chapter the results are presented. The data has been gathered with a survey distributed online through Facebook groups to people living in, or near, mining towns in the north of Sweden. Firstly, a profile of the respondents will be presented, followed by statistical tests which have been made to determine relationships, associations and predicting variables.

4.1 Profile of respondents

The total number of respondents gathered in this study is 190 and all will be used in the data analysis.

Alongside the questions which will help us answer our research question, demographic data was collected by asking their gender, age, and location of living. The gender distribution is presented in figure 4.1, 109 (57.4%) males and 81 (42.6%) females choose to participate in the online survey.

Figure 4. 1 Gender distribution of respondents

Table 4.1 shows the age distribution among the respondents and the number of younger participants 18-24 (4.2%) and 25-34 (14.2%) is a clear minority among the respondents. While it is evenly distributed among the older age groups 35-44 (17.9%), 45-54 (24.7%), 55-64 (20.5), and 65+ (18.4%).

20 Table 4. 1 Age distribution of respondents

Figure 4.2 presents the location the respondents live. The respondents are evenly distributed between Kiruna (43.2%) and Gällivare (55.8%) while Svappavaara (1.1%) only has a couple participants. The survey was published in two Facebook groups which are connected to Kiruna and Gällivare which can be the answer to the lack of respondents from Svappavaara.

21 Figure 4. 2 Geographical location of respondents

4.2 Cronbach’s alpha

Appendix III displays the Cronbach alpha for the questions regarding dialogue (0.773), distributional fairness (0.877), procedural fairness (0.906), confidence in governance (0.790), and trust (0.968). The Cronbach alpha for the categories ranges from 0.773 to 0.968 implying there is a high validity of the measurements. The high Cronbach alpha means the questions in the survey measure their intended variable without being redundant.

4.3 Pearson’s test for correlation

Pearson’s test for correlation is presented in table 4.2 which displays dialogue and distributional fairness, dialogue and procedural fairness, and dialogue and confidence in governance. Dialogue has, with all the other variables, a p-value < 0.01 which indicates the values are statistically significant.

Dialogue and distributional fairness

Pearson’s correlation coefficient is 0.404 which indicates there is a medium, on the border to strong, positive correlation between the two variables. In other words, this means that a higher value in dialogue is related to the community perceiving a higher level of distributional fairness.

22 Dialogue and procedural fairness

Pearson’s correlation coefficient is 0.513 which indicates there is a strong positive correlation between the two variables. In other words, this means that a higher value in dialogue is related to the community perceiving a higher level of procedural fairness.

Dialogue and confidence in governance

Pearson’s correlation coefficient is 0.373 which indicates there is a medium positive correlation between the two variables. In other words, this means that a higher value in dialogue is related to the community

perceiving a higher level of confidence in governance.

Table 4. 2 Pearson’s correlation

4.4 Multiple Regression

The multiple regression analysis is represented below in table 4.3 and 4.4 and consists of the dependent variable trust and the predictor variables of confidence in governance, distributional fairness, and procedural fairness. An observed R value of 0.907 implies there is a solid level of prediction.

The adjusted R Square value is 0.819, which indicates that the independent variables explain 81,9 % of the variability of the dependent variable, trust.

The ANOVA table (see appendix I) shows a p value that is less than 0,05 and thus the proposed model is a good fit for the data.

23

The model consisting of trust, distributional fairness (DF), procedural fairness (PF), and confidence in governance (CG) and coefficients from table 4.4 would look like this:

Trust = - 0,148 + 0,094(DF) + 0,533(PF) + 0,757(CG)

The t value from distributional fairness is 1.885. This means distributional fairness shows that there is likely no significant difference. Procedural fairness and confidence in governance show high t values, 8,958 and 9,612 respectively. This indicates that there is a significant difference and implies a greater evidence against the null hypothesis.

Looking at the p values, the conclusion is that distributional fairness has a p value greater than 0,05, thus it is not statistically significant, and the proposed hypothesis is not strong enough to reject the null hypothesis.

The p value of procedural fairness and confidence in governance are less than 0,05, which implies the rejection of the null hypothesis and that the values are statistically significant.

The conclusion drawn from this is that distributional fairness is not a significant predictor of trust whereas procedural fairness and confidence in governance are significant predictors of trust.

Table 4. 3 Adjusted R square

Table 4. 4 Coefficients and model summary

24

4.5 Chi-squared test

The crosstabulation (see appendix II) shows that the number of expected counts and the actual real counts do not add up; it leads to believe there is an association between the two.

Table 4.5 displays the Phi and Cramer’s V measurement. Cramer’s V is 0.435 which indicates that there is a moderate relationship between trust and acceptance. A high level of trust indicates that there is a high level of acceptance as well.

Table 4.6 Displays the Chi square test. Pearson Chi-Square is 0.000 which is lower than 0.05 and infers that the test has statistically significant results, trust in the mining industry and acceptance of the mining

company has a statistically significant association with each other.

Table 4. 5 Cramer’s V

Table 4. 6 Chi-Square Tests

25

5. Data analysis

In this chapter the results presented in chapter 4 will be analyzed. The studies hypothesis will be discussed, and the results will be explained.

5.1 Is dialogue an efficient communication tool?

The Pearson’s test for correlation yielded following results:

• Dialogue and distributional fairness, a medium to strong positive correlation between the two variables.

• Dialogue and procedural fairness, a strong positive correlation between the two variables

• Dialogue and confidence in governance, a medium positive correlation between the two variables

The medium to strong positive correlation between dialogue and distributional fairness indicate that the communities in Kiruna, Gällivare, and Svappavaara taking part in dialogue with mining companies perceive a higher degree of company resources being distributed fairly.

The strong positive correlation between dialogue and procedural fairness implies that the community in Kiruna, Gällivare, and Svappavaara taking part in dialogue with mining companies strongly perceive that they are part of the decision-making processes.

The medium correlation between dialogue and confidence in governance shows that the community in Kiruna, Gällivare, and Svappavaara taking part in dialogue with mining companies perceive that the government will be able to manage the social and environmental impact from company operations.

This proves Hypothesis 1 “Is dialogue an efficient way for companies withing the M&E sector to

communicate with local communities?”, since the positive correlations mean a higher value of dialogue will lead to a higher value in all the three variables. The results are supported by Mercer-Mapstone et al (2017) which states that the quality of contact is important to acquire a SLO and that building a relationship with stakeholders need to focus on “doing with” them. Moffat and Zhang (2014) validate these results by stating that the trust communities have towards companies is affected by how companies engage with them. The reason dialogue is an efficient way to communicate procedural fairness can be explained by the fact that people who feel like they have been heard and participated in the procedural processes will regard the procedure as fair (Cremer et al., 2005). So even though the communities do not get what they want, they

26

still perceive the procedural fairness as high and will accept the outcome if companies have a dialogue with them. An explanation as for why dialogue was an acceptable tool to communicate confidence in governance to local communities, can be found in an article published by Zhang et al (2018). Zhang et al (2018) points out that companies who mention in their early dialogue with local communities that they will adhere to government regulations and laws when applying for a SLO, will significantly increase the local

community’s confidence in governance. A plausible reason as to why dialogue was not as good of a tool in northern Sweden compared to what Zhang et al (2018) showed, could be that the mining companies failed to highlight early in their dialogue how they would adhere to regulations and laws set out by the government.

5.2 What leads to trust?

The multiple regression analysis uncovered the following results:

• Distributional fairness, not a significant predictor of trust

• Procedural fairness, a significant predictor of trust

• Confidence in governance, a significant predictor of trust

With two out of three factors being significant predictors of trust, Hypothesis 2 “Can distributional fairness, procedural fairness, and confidence in governance lead to trust in a company?” is partially proven true.

Procedural fairness and trust in governance were strong predictors of trust and this is backed up by the results from Zhang et al (2015) that states a higher level of trust in governance resulted in more trust in companies and a prominent level of procedural fairness resulted in an elevated level of trust (Cremer et al., 2005). For this study, trust in governance was shown to be the strongest predictor of trust for Swedish people living in Kiruna, Gällivare, and Svappavaara. Data analysis from the multiple regression pointed out that distributional fairness was not a significant predictor of trust, which goes against the hypothesis

presented by Zhang et al (2015) where they found that distributional fairness positively predicted trust in the mining industry in China, Chile, and Australia. These different results can be due to societal differences among the countries. Roy et al (2018) points out individualistic societies perceive fair distribution as proportional distribution whereas collectivistic societies view fair distribution as equal distribution among the group. As such, one can argue that people living in an individualistic society prioritize their own

wellbeing over the group and are more content if they are pleased rather than the whole group. Therefore, an individual living in Kiruna or Gällivare might accept an unfair distribution of the wealth from mining due to them not seeing themself as a part of the mining process. And because they do not see themselves as part of

27

the mining process, distributional fairness does not have the same effect on trust of the mining industry.

This would be a stark contrast to the China and Chile example, where the society is based on the community and the group, and people would not be so willing to accept mining unless the benefits of it was distributed as evenly as possible among the stakeholders being affected by it.

The constant of the proposed equation of trust was observed to be negative. This is an interesting

observation as it means that trust, when the other three factors are zero, is negative. Negative trust implies there is mistrust from the start when looking to establish trust. Another interesting observation is that to overcome this negative trust, one only needs to affect the procedural fairness with one and the trust result will be positive. It would still be a low overall trust score but nonetheless a positive trust score.

5.3 Does trust in a company result in a SLO?

The Chi-square results indicated there is a relation between trust and acquiring a SLO. A company building trust with the community will make the community have a higher level of acceptance which will result in acquiring a SLO. This proves Hypothesis 3 “Does trust in a company result in a SLO?”. This is backed up by Moffat and Zhang (2014) that states trust to be a significant predictor of company acceptance from local communities. Zhang et al (2015) also concludes that trust had a positive relationship with acceptance of the company. Why trust leads to a SLO can be explained by the community having knowledge-based trust, where knowledge-based trust was trust based upon knowing the decision maker’s actions, thoughts, and intentions well enough to predict their actions (Lewicki & Bunker, 1994). In Kiruna, Svappavaara, and Gällivare, the community “trusts” that the mining company will behave and honor its agreements because they expect the mining company to avoid conflicts as they are a waste of time and company resources.

Therefore, the community’s trust in the company makes them think they can predict the company’s actions which leads them to granting it a SLO.

Trust can also be seen as a psychological state where a company and the community rely on each other making them both believe the other parts intentions and behavior to be positive (Rousseau et al., 1998).

Why trust leads to acceptance can therefore be explained by trust putting the community in a psychological state where they expect the mining company’s behavior to be positive and their intentions to be helpful.

When the community believes the companies intentions and behavior to be positive, they will accept the company and grant them a SLO.

28

6. Conclusion

In this chapter the conclusions of the study will be presented, and the research question will be answered with the help of the three hypotheses. This will be followed by implications and recommendations for further research.

The purpose of this study was sought to be explained with the help of three hypotheses:

Hypothesis 1: Is dialogue an efficient way for companies within the M&E sector to communicate with local communities?

The finding of this study suggests dialogue to be an efficient tool to communicate with local communities since there is a medium to strong positive correlation between dialogue and distributional fairness,

procedural fairness, and confidence in governance. The analysis had a significant p value at 0,01.

Early emphasis on following laws and regulations when having a dialogue with a local community, strengthens the community’s confidence in governance.

Dialogue is an efficient way to communicate distributional fairness with local communities, making them feel involved in the process, which will lead to them accepting the outcome.

Hypothesis 2: Can distributional fairness, procedural fairness, and confidence in governance lead to trust in a company?

The findings indicates that the level of trust has two strong predictors, which are procedural fairness and confidence in governance. Both factors have a significant p value below 0,05. Distributional fairness was found to not be a strong predictor of trust and the reasoning as to why can be explained by cultural

differences such as Sweden being a more individualistic society and therefore, they do not see themselves as entitled to the benefits if they are not a part of the company.

Hypothesis 3: Does trust in a company result in a SLO?

According to the findings of this study, a relation exists between trust and the acquisition of a SLO. The findings have a statistically significant p value at 0,01. This suggests that trusting a company impacts the

29

view of the company, meaning the community expect their behavior to act in their best interests and therefore they will grant them a SLO.

RQ1: What does companies within the M&E sector need to communicate with local communities to acquire a SLO?

According to findings of this study procedural fairness and confidence in governance needs to be

communicated by companies towards communities in order to build trust and acquire a SLO. However, the analysis points towards cultural differences contributing to what factors are the most crucial to establish trust between community and company in a SLO circumstance. Early dialogue and making the local community feel involved in the decision making processes will increase the effectiveness of dialogue.

6.1 Implications for theory

This study contributes to theory via the results obtained from combining two previous research articles, Moffat & Zhang (2014) and Zhang et al (2015) and applying them in a new context. Dialogue has previously been used in other studies (Moffat & Zhang, 2014; Mercer-Mapstone et al, 2017; Prno, 2013) and applying dialogue in another context will help expand current theory. By basing the study in Sweden, we provide new significant data to current research which will contribute to theory.

The findings of this study suggests that companies need to communicate procedural fairness and confidence in governance with local communities to acquire a SLO. Whilst distributional fairness does not lead to a SLO. Therefore, it is to be confirmed that cultural differences affect which variables companies need to communicate with the communities in order to build trust and acquire a SLO. This takes theory a step further to understand why different factors affect communities differently when a company is looking for a SLO.

The findings in this study also suggest that dialogue is an efficient way for companies to communicate with local communities. It complements earlier theories about people who feel like they have been heard will accept the outcome no matter the result. We would like to add to these theories that a company entering dialogue with local communities directly after a change has occurred will receive more trust than a company waiting before engaging in dialogue with a local community. This is especially relevant concerning changes to laws and regulation, regarding confidence in governance.