Development of a Model and Measure

This outline traces the development of a conceptual model of the social license to operate. The development process was a conversation between theory and practice over a number of years. An original study of the levels of acceptance of a local mine across a 15 year period led to the conceptualization of the social license presented in the Thomson and Boutilier (2011) chapter. That was followed by attempts to measure the social license quantitatively in a survey of the stakeholder of the same mine. On a 5-point scale, representatives of stakeholder group for the same mine rated their agreement or disagreement with an initial pool of two dozen statements. Stakeholders’ verbal statements of support or opposition provided criterion validation for the whole set of statements as an additive scale of the social license. However, sub-scales meant to measure Thomson and Boutilier’s levels of social license did not conform with the hypothesized cumulative nature of the levels. Over the past three years, the pool of statements was refined in studies of stakeholder networks in Australia, Bolivia, and Mexico. The latest version consists of 15 statements, which were used earlier this year at the original mine in Bolivia. A factor analysis revealed four factors that did conform with the cumulative nature of the levels of social license, but which, in the process, suggested a modification to the model. Continuing the dialogue between theory and practice, we present the modified model and suggest ways it can focus mine management on a chronically neglected aspect of stakeholder relations, namely, the role of the corporation in fostering more equitable social contracts at local and regional levels in both developed and developing countries. Modelling and Measuring the SLO 2 ____________________________________________________________________________ Development of a Model and Measure According to Thomson and Boutilier (2011) a social license to operate (SLO) is a community’s perceptions of the acceptability of a company and its local operations. Based on extensive interviews with resettled villagers about the ups and downs of their relationships with a Bolivian mine over a 15 year period, Thomson and Boutilier identified four levels of the SLO. They claim that the level of SLO granted to a company is inversely related to the level of socio-political risk a company faces. A lower SLO indicates a higher risk. The lowest level of SLO is having the social license withheld or withdrawn. This implies that the project is in danger of restricted access to essential resources (e.g., financing, legal licenses, raw material, labour, markets, public infrastructure). Losing a social license represents extremely high socio-political risk. The next highest level of SLO is acceptance of the project. On Figure 1 this layer covers the greatest area in order to indicate that it is the common level of social license granted. If the company establishes its credibility, the social license rises to the level of approval. Over time, if trust is established, the social license could rise to the level of psychological identification, where the level of socio-political risk is very low. Figure 1: The “pyramid” model of the SLO proposed by Thomson & Boutilier (2011) Political Differences in Stakeholder Networks Speaking of a social license as granted by a community is a shorthand for a more complex situation. Thomson and Boutilier prefer to speak of stakeholder networks rather than communities. They adopt Freeman’s (1984) definition of stakeholders as those who could be affected by the actions of a company or who could have an effect on the company. The stakeholder network, therefore, could include many parties outside a geographic community, Modelling and Measuring the SLO 3 ____________________________________________________________________________ such as ethical investment funds, international human rights activists, international financial institutions, and national governments. The stakeholders may or may not agree on what level of SLO should be granted. Usually there are political differences of opinion within the network of stakeholders. An understanding of how the various levels of SLO are proportionally distributed throughout the network provides the basis for strategies for changing the overall SLO (e.g., alliance formation, issue reframing, etc.). Anti-mining activists of various stripes (e.g., anti-capitalism, anti-development, anti-globalization, anti-mining) will use such knowledge to lower the SLO while mining companies will use it to try to raise their SLO. In many villages of the developing world, local stakeholders use the opportunity to try to negotiate a better microsocial contract with the mining company than the one they currently have with their national government. Development of a Measure of the SLO The four level concept of the SLO proposed by Thomson and Boutilier was used to devise a pool of two dozen statements intended to measure the SLO in interviews with mine stakeholders in Bolivia in 2009. The whole set was validated against verbal comments by the stakeholders and contextual information about their political campaigning for or against aspects of the mine’s operations. However, the sub-sets of statements meant to measure the separate layers of the SLO did not display the cumulative nature hypothesized by Thomson and Boutilier. Technically speaking, they did not form a Guttman scale (Guttman, 1950). Moreover, although a factor analysis yielded four factors, the factors lacked some face validity in terms of matching them with the four levels of SLO. The pool was refined in 2010 for a study in Mexico and another in Australia. Standard psychometric item analysis techniques were used to refine the set of statements. The whole set continued to receive criterion validation against verbal comments from stakeholders. A refined set of 15 statements (see Appendix A) was used at the same Bolivian mine again in 2011. The agree/disagree ratings (5-point scale) were factor analyzed. The four factors shown in Table 1 emerged from a varimax rotation. Appendix A shows which statements measured which factor. The labels for the factors were derived from the content of the corresponding statements. The Modified “Arrowhead” Model of the SLO The modified model of the SLO is shown in Figure 2. The four factors can now be measured independently by the same statements that measure the overall level of SLO granted by the stakeholder. The conversion of the four levels into a continuum is meant to symbolize the idea that the overall level of SLO is a continuum. The average of the ratings on the 15 statements measures it. The division of the shape, which now looks more like an arrowhead than a pyramid, into four regions is meant to suggest the manner in which they accumulate as the level of SLO increases. Modelling and Measuring the SLO 4 ____________________________________________________________________________ Table 1: Four Factors Constituting Three Levels of SLO Level & Label Description Role in Determining SLO Levels as Described in Thomson & Boutilier Pyramid Model 1. Economic legitimacy The perception that the project/company offers a benefit to the perceiver. If lacking, most stakeholders will withhold or withdraw the SLO. If present, many will grant an acceptance level of SLO. 2a. Socio-political legitimacy The perception that the project/company contributes to the well-being of the region, respects the local way of life, meets expectations about its role in society, and acts according to stakeholders’ views of fairness. If lacking, approval level of SLO is less likely. If both this and interactional trust (2a & 2b) are lacking, approval level is rarely granted by any stakeholder. 2b. Interactional trust The perception that the company and its management listens, responds, keeps promises, engages in mutual dialogue, and exhibits reciprocity in its interactions. If lacking, approval level of SLO is less likely. If both this and sociopolitical legitimacy (2a & 2b) are lacking, approval level is rarely granted. 3. Institutionalized trust The perception that relations between the stakeholders’ institutions (e.g., the community’s representative organizations) and the project/company are based on an enduring regard for each other’s interests. If lacking, psychological identification is unlikely. If lacking but both sociopolitical legitimacy and interactional trust are present (2a & 2b), most stakeholders will grant approval level of SLO. Preliminary analyses indicated that 85percent of the cases conformed to a modified cumulative scale (i.e., Guttman scale) as described in Table 1. The factors are cumulative to the extent that stakeholders with a high score on perceptions of institutionalized trust (factor 3) will always have high scores on all the other factors. Stakeholders with low scores on perceptions of socio-political legitimacy (factor 2a) and interactional trust (factor 2b) will never have high scores on institutionalized trust (factor 3). Similarly, stakeholders with low scores on perceptions of economic legitimacy (factor 1) will never have high scores on any of the other factors (2a, 2b, or 3). Modelling and Measuring the SLO 5 ____________________________________________________________________________ Figure 2: Levels of Social License with the Four Factors that Determine the Proportions of Stakeholders at Each Level Figure 3 gives an example of how the proportions of stakeholders granting each level of SLO is altered with high or low scores on of each of the four factors. In the left panel of Figure 3 the project/company is perceived as having only economic legitimacy. The company basically offers nothing more than a financial transaction with the stakeholders. The range of the SLO that could possibly be granted is therefore confined to the region under both diagonal lines. It is a predominantly red and yellow region, indicating high risk and a SLO that vacillates between withdrawal and bare acceptance. Similarly, at any one time about a third of the stakeholders will withhold the SLO while two-thirds will grant an acceptance level. In the middle panel Figure 3 the project is not perceived to have socio-political legitimacy or institutionalized trust. The area of the pyramid corresponding to these two factors has been screened out to represent the low scores on these factors. The remaining area has a higher proportion of green than the left panel. This indicates that the SLO will likely fluctuate in and around the yellow area. At the best of times it will reach into the green approval range, but at the worst of times it will be withdrawn. At any one time, the largest block of stakeholders is likely to grant an acceptance level of SLO. In the right panel of Figure 3, the stakeholders see the project as having all three of economic legitimacy, socio-political legitimacy, and interactional trust. It is seen as lacking in only institutionalized trust. In this case the green area is nearly doubled. The chances of the project enjoying acceptance are high. 1 At any one time, the majority of stakeholders would be divided between granting an acceptance level and granting an approval level. 1 The model in Figure 2 can be applied at both the population level and the level of individual stakeholders. At the population level, the non-faded areas indicate proportions of stakeholders likely to grant each level of SLO. Accordingly the four factor scores are based on average perceptions in the whole stakeholder population. At the individual level, the four factor scores are based on the individual stakeholder’s perceptions. At the individual level the non-faded areas represent the possible range of SLO levels the stakeholder could grant. The level could Modelling and Measuring the SLO 6 ____________________________________________________________________________ Figure 3: How the Proportions of Stakeholders Granting Each Level of SLO Change with the Presence or Absence of Perceptions of Each Factor Figure 4 shows institutionalized trust being high without both socio-political legtimacy and interactional trust being high. This is a non-cumulative pattern that is hypothesized to be infrequently observed. Figure 4: A Non-Cumulative Configuration of Factors: Hypothesized to Be Rare vary across time but, given constant perception factor scores would only range in the non-faded area. Moreover, the likelihood of the stakeholder granting a specific SLO level at any given time is proportional to the percentage of the non-faded area dedicated to any one colour. Therefore, interpreting Figure 3 at the individual level, the stakeholder would likely grant an acceptance level (orange/yellow) most of the time, with frequent slips into withdrawal, and occasional sallies into approval. Modelling and Measuring the SLO 7 ____________________________________________________________________________