Stakeholder engagement in sustainability reporting: A classification model

2.1 Stakeholder engagement and communication strategy

Critical stakeholder attention is not restricted to a company's decisions and actions but also focuses on the decisions and actions of suppliers, consumers, and politicians, which may spur criticism toward a company (e.g., Nike and Cheminova) (Brotons & Sansalvador, 2020; Morsing & Schultz, 2006). However, because no manager or organization makes sense in total isolation (Craig-Lees, 2001), the objective of this study is to identify ways in which companies try to engage with their stakeholders. This process is what Gioia and Chittipeddi (1991) refer to as interpretive work under the label “sensemaking” (i.e., trying to figure out what others want and ascribe meaning to it).

The extent to which individuals or organizations are able to integrate the sensemaking of others influences their ability to strategically enact a productive relationship (Gioia & Chittipeddi, 1991). In line with the works of Craig-Lees (2001), Cramer, Jonker, and Van Der Heijden (2004), Morsing and Schultz (2006), and Johnson, Redlbacher, and Schaltegger (2018), we use the sensemaking method for a better understanding of the communication processes and engagement of companies with internal and external stakeholders.

Gioia and Chittipeddi (1991) expand the notion of sensemaking by introducing the concept of sensegiving, putting a special focus on the managerial processes facilitating sensemaking in organizations. According to them, sensemaking is followed by actions in terms of articulating an abstract vision that is then widespread and championed by corporate management toward stakeholders in a process labeled “sensegiving” (i.e., attempting to influence the way another party understands or makes sense).

It is important to also point out that Gioia and Chittipeddi's (1991) theory has an internal focus on sensegiving and sensemaking processes among managers and employees. Craig-Lees (2001), Cramer et al. (2004), and Morsing and Schultz (2006) expanded this notion to an external focus by involving external stakeholders in corporate CSR efforts. Based on stakeholder theory, we have chosen to investigate and analyze communication processes used to engage with internal and external stakeholders concomitantly and under equal criteria.

Stakeholder engagement is understood as practices the organization undertakes to positively involve stakeholders in organizational activities (Greenwood, 2007). Corporate responsibility refers to the corporation's obligation to act in the interests of legitimate organizational stakeholders (Herremans et al., 2016). To integrate such views, Lim and Greenwood (2017) presents a model that reflects the multifaceted relationship between the two following constructs: stakeholder engagement and corporate responsibility.

Under Greenwood's model, stakeholder engagement is a process of consultation, communication, dialogue, and exchange. High engagement is when these activities are numerous and/or of high quality; low engagement is the opposite. Additionally, Lim and Greenwood (2017) presents stakeholder agency as a proxy for the responsible treatment of stakeholders, which represents the number and breadth of stakeholder groups in whose interest the company acts.

This study focuses on a deeper investigation into the nature and level of stakeholder engagement in actions, which, according to Greenwood, takes place when an organization responds to the needs of stakeholders with the aim of furthering its goals. Therefore, the management of stakeholders is understood as strategic in nature. Rather than acting with the intention of fulfilling the interests of stakeholders, the organization acts in its own interests, and the stakeholders are merely a vehicle for doing so (Greenwood, 2007).

Greenwood (2007) also applied a Habermasian view and some criticism to the notion of strategic engagement under her model. Based on Grunig and Hunt's (1984) characterization of models of public relations, Morsing and Schultz (2006) unfold three types of stakeholder relations in terms of how companies strategically engage in CSR communication vis-à-vis their stakeholders: the information strategy, response strategy, and involvement strategy.

According to Morsing and Schultz (2006) and similar to Grunig and Hunt's (1984) public information model as it concerns the stakeholder information strategy, communication always occurs in one direction, from the organization toward its stakeholders. Communication is basically viewed as “telling, not listening” (Grunig & Hunt, 1984), and, therefore, the one-way communication of the stakeholder information strategy has the purpose of disseminating information, not necessarily with a persuasive intent but rather to inform the public as objectively as possible about the organization (Morsing & Schultz, 2006). In our investigation for this study, we classified the information strategy as a Level 1 stakeholder engagement communication strategy, which is further explained later.

Building on Gioia and Chittipeddi's (1991) terminology, Morsing and Schultz's (2006) work provides a framework that analyzes communication between companies and internal stakeholders, as well as engages companies in progressive iterations of sensemaking and sensegiving processes with external stakeholders, enhancing awareness of mutual expectations.

Additionally, for Morsing and Schultz (2006), the stakeholder response strategy is based on a two-way asymmetric communication model, as opposed to the two-way symmetric model of the stakeholder involvement strategy. In other words, the main difference between the models of the stakeholder response strategy and stakeholder involvement strategy is that companies also change as a result of interactions with stakeholders in the latter case, even though both consist of a two-way communication strategy (or sensemaking and sensegiving, as previously noted). In this sense, stakeholder engagement also involves communication for mutual consensus to achieve organizational goals (Lim & Greenwood, 2017). Concerning the purpose of our investigation, the models of the stakeholder response strategy and stakeholder involvement strategy have been respectively classified as Level 2 and Level 3 stakeholder engagement communication strategies.

Morsing and Schultz (2006), Greenwood (2007), Herremans et al. (2016), Grushina (2017), and Lane and Devin (2018) all explain the various stakeholder engagement processes and their definitions and dialogue with communication and CSR practices, but none of these works advanced knowledge about engagement strategies differences and how to implement the strategies in sustainability reports. These researchers also no longer attempt to predict when organizations will choose an engagement strategy to the detriment of others. These are topics that this study intends to advance.

3.1 Methodological procedures

Data collection was carried out through a content analysis of the GRI sustainability reports. A content analysis is a research method used to encode text (i.e., communications) and generate inferences (Krippendorff, 2018), assuming that the language used by the sender reflects its context and communication objectives. In this work, we analyze the content of the reports, generating inferences about how companies express themselves, and record communications with different groups of stakeholders regarding company activities and corporate social responsibility practices. We present a methodological matrix in Figure 1 to explain the choices that define our bibliographic review, data collection, and the synthesis and analysis of the results so that the conclusions and contributions of this study are scientifically consistent.

As presented in the methodological matrix, the database was comprised of 119 reports downloaded from the GRI platform. For the purpose of analysis and systematization of all content, data was collected from all the reports separated by company, country, and region. The data analysis was performed by analyzing the content of the reports in relation to the engagement actions declared by each of the companies and classifying them according to the levels proposed by this work (Level 1, Level 2, and Level 3), as presented in Table 1.

We carried out the content analysis in a “Stakeholder Identification & Engagement” session, available through the GRI reports, by considering the engagement actions reported by each company. This is a mandatory session during the preparation and dissemination of reports and focuses on the following components: G4-24, list of identified stakeholders; G4-25, basis for identifying stakeholders; G4-26, approach to stakeholder engagement; and G4-27, key topics and concerns by stakeholder.

In the database created for the systematization and classification of the engagement actions, columns were inserted for each stakeholder identified in the reports and the number of actions mentioned for each stakeholder. The frequency of the stakeholders for each report/company, the total number of actions per level of engagement, and the division of the engagement actions for each stakeholder were then calculated, thus segregating the engagement actions by level and by stakeholder.

The process of coding the data, reading the reports, and analyzing the engagement section of the reports was done manually, following three steps for its completion and validity of collection and analysis, as suggested by Unerman (2000) and Krippendorff (2018): (a) preparation prior to collection and analysis that involves consulting dictionaries in Portuguese, English, and Spanish for synonyms of “stakeholders” (e.g., employees, community, customers, and shareholders) and for various engagement initiatives (e.g., reports, newsletters, forums, customer service, and working groups); (b) internal validation by comparing the results of different analysts (a sample of reports was analyzed by two different analysts and the results compared so that there was a calibration of the collection method and the coding spreadsheet); and (c) external validation with the verification of the coded material and a sample of reports, with the corroboration of two specialists of sustainability reports and international social and environmental disclosure guidelines, such as the GRI.

To categorize the stakeholders' engagement strategies, we employed a few analysis steps in terms of quality, focus, and extent. First, the quality is evaluated by the three levels of engagement developed in the proposed model. Second, the focus is analyzed by identifying the stakeholders to whom the engagement actions are directed. Finally, the extent is examined through the number of stakeholders involved by the companies in their engagement actions. One example extracted from a GRI report used in our sample for Duke Energy identified 11 key stakeholders. Among the many engagement actions reported, an example for each level follows: Level 1, plant visits and tours for the media; Level 2, meetings with local authorities and organizations within the community; and Level 3, the employees' continuous improvement program.

3.2 Categorization of the engagement level

As previously mentioned, we defined engagement quality according to three levels of communication strategy conducted to engage the firm with internal and external stakeholders, based on Morsing and Schultz's (2006) classification: information strategy (Level 1), response strategy (Level 2), and involvement strategy (Level 3). We complemented Morsing and Schultz's (2006) classification with Gioia and Chittipeddi's (1991) and Gable and Shireman's (2005) definitions. Based on these studies, we proposed a model of classification of stakeholder engagement actions that includes the communication strategy level, type of engagement, and the process of social interactions. Table 1 exhibits the classification and engagement actions adopted by the firms.

The classification model in Table 1 was the methodological tool used to analyze the reports. We identified engagement actions in the reports and classified them according to the level of communication strategy. Each company could simultaneously show actions at the three levels directed to different stakeholders. We calculated the frequency of actions at each level for each analyzed report.

3.3 Analysis of engagement quality, focus, and extent

The stakeholder engagement quality involves the number of actions, the most cited actions, and the most cited stakeholders at each level. It is differentiated by three levels of quality: Level 1 (stakeholder information strategy) includes actions aimed at identifying and informing stakeholders, Level 2 (stakeholder response strategy) involves consulting the interests and supporting the demands of stakeholders, and Level 3 (stakeholder involvement strategy) is aimed at establishing partnerships and collaborations with stakeholders in projects.

For the engagement focus, we calculated the frequency of the stakeholders for each report/company and the total number of actions per level of engagement and per stakeholder. The most cited stakeholder groups at each level were taken as clues to analyze the focus of the engagement, and the number of stakeholders cited was used as a method to analyze the extent of the engagement. To analyze the stakeholder engagement extent, we consider three levels of engagement strategies and categorize stakeholder extent into three groups: high (three or more stakeholder groups), intermediate (two groups of stakeholders), and low (just one stakeholder). To analyze the engagement extent impact, we also propose a stakeholder engagement matrix that categorizes companies in terms of the level and extent of engagement. The matrix helps determine whether the companies involved a large number of stakeholders with a high quality of engagement.

4.1 Stakeholder engagement quality: The engagement level

We classified the engagement actions within the sample of 119 GRI reports disclosed by companies operating in the energy sector. Table 2 summarizes the findings and presents the number of actions, the most cited actions, and the most cited stakeholders at each level.

Levels 1 and 2 concentrate the highest number of engagement actions identified. Therefore, the engagement strategies are consistent with actions that inform (Level 1) and consult and support the stakeholders (Level 2). We observed that the sample companies have a strong tendency to focus on Level 1 stakeholder engagement actions as the primary communication strategy. We also highlight the most cited stakeholders in the actions. Employees are still the most favored stakeholders in engagement actions at Levels 1 and 2, whereas communities and the government represent the majority of the efforts at Level 3.

The classification of engagement levels, as illustrated in Table 1, goes beyond the categorization of the levels of communication strategies proposed by Morsing and Schultz (2006), also incorporating the different types of collaboration and involvement with stakeholder groups, supported by the work of Gable and Shireman (2005), and Gioia and Chittipeddi's (1991) process of the interaction of sensemaking and sensegiving. Thus, the understanding of the quality of engagement levels reflect the process not only of the organization's communication with its stakeholders but the interaction and construction of the relationship, being at certain times focused on the organization and its internal stakeholders, as usually happens in sensemaking processes, and at other times through sensegiving (high-quality engagement actions), with the involvement of stakeholders external to the organization.

In this sense, the results show that, in general, external stakeholders to the organization, such as the government and community, demand a more structured engagement process with a higher level of interaction, which represents a Level 3 involvement strategy, which is an interactive sensegiving process in which the organization seeks to develop collaborative actions and initiatives that for stakeholders within the organization is inherent in the practices already developed internally.

Thus, in the dissemination of the organization's actions, such as the case explored here of sustainability reports, companies in the energy context have developed more strategic actions of engagement of greater quality with external stakeholders, which are mainly linked to their external impact and dependence and are mandatory, depending on their operating context in society. It is noteworthy that most companies in the energy sector operate in highly regulated markets and have a social, economic, and environmental obligation of great interest to external stakeholders.

4.2 Stakeholder engagement focus

The stakeholders cited in the 119 reports include employees (108 reports), the community (100), shareholders (95), customers (93), suppliers (86), and the government (80). In addition to those stakeholders, we highlight several others that relate to and are important to companies belonging to the energy sector. These stakeholders include industry regulators, media, nongovernmental organizations (NGOs), business partners, academia (universities and research centers), and finance institutions.

We observe that when stakeholder engagement actions are divided according to Levels 1, 2, and 3, the stakeholders´ focus changes. As already highlighted in Table 2, different stakeholders received more attention according to each level of engagement in relation to the total number of engagement actions. It can thus be concluded that, as previously mentioned, employees are generally the stakeholders most favored in Level 1 and 2 actions in most reports/companies, whereas communities and governments are the primary receivers of Level 3 actions. It is also observed that, although most of the actions of engagement classified at Levels 1 and 2 focus on the same stakeholder groups (employees, shareholders, and customers), Level 3 of engagement actions shows a wider distribution, with efforts and engagement strategies more evidently focused on stakeholders such as the community, government, employees, and NGOs.

As already defended by authors such as Noland and Phillips (2010), Johnson-Cramer et al. (2004), and Greenwood (2007), it is important to differentiate the nature of firms' engagement with their stakeholders and the quality of those relationships. In this sense, there is an opportunity for firms to improve the quality of engagement by investing in involvement strategy actions and to gain a competitive advantage by engaging its stakeholders.

4.3 Stakeholder engagement extent

As previously mentioned, to analyze the stakeholder engagement extent, we consider three levels of engagement strategies and categorize the stakeholder extent into three groups: high, intermediate, and low. We thereby developed the following stakeholder engagement matrix, presenting nine categories of engagement strategies, as shown in Figure 2.

The matrix presented in Figure 2 shows all the possible combinations of a high, intermediate, and low number of identified stakeholders (Y-axis) and the concentration of actions classified as engagement Levels 1, 2, and 3. Such combinations were then divided into nine quadrants. At the lower-left bottom, for instance, there are companies reporting a low number of identified stakeholders and a high number of engagement actions, thus falling into the Level 1 category, which we have called “Insulated.” At the extreme opposite quadrant at the upper-right corner, we have “Generous” companies—those that involve a high number of stakeholders in their reports and concentrate most of their engagement actions at Level 3 in their strategy to engage and influence those stakeholders.

6.1 Limitations and future studies

As in all studies, there are certain limitations that should be noted. The sample of this study represents only the energy sector, although the choice of the sector has been justified due to the higher level of commitment and development of CSR practices and risks involved in its operations, and a certain homogeneity has been found in the reports analyzed.

A methodological weakness of this work is that it is difficult to replicate due to the non-systematization of the content analysis used in the sustainability reports. As the data coding and analysis was carried out in three languages and for a specific section of the report with a method of association and the frequency of words, if we were to use qualitative data analysis software (such as NVivo or Atlas.ti), we would not be able to capture what was sought and generate the inferences described here. However, there is an avenue of possibilities for future work, applying the content analysis method to sustainability reports, as well as other data sources, using the engagement matrix and levels proposed in this study.

For future research, we believe that an investigation and analysis of influences on companies from other sectors under different contexts, such as their capital structures, corporate responsibility practices, supply chains, or competition networks, can provide an important contribution to identify which variables drive stakeholder engagement strategies. Finally, to justify why companies should invest in different or more robust stakeholder engagement strategies, an investigation of companies' performance over the years for each quadrant of the stakeholder engagement strategy matrix can provide an important contribution to the field.