An Explanatory Model of Materiality in Sustainability Accounting: Integrating Accountability and Stakeholder Heterogeneity

Abstract

This study aims to advance the conceptualisation of materiality in sustainability accounting by tackling theoretical deficiencies that remain unresolved in the current understanding of materiality. In doing so, we integrate accountability and stakeholder heterogeneity into a new conceptual model of materiality that is used to explain how an organisation sets priorities in reporting and in managing different sustainability issues in a heterogeneous stakeholder environment. The model is illustrated with reference to events involving the multinational organisation Exxon. The empirical accounts of the Exxon case presented herein fortify our knowledge claims for this model and, moreover, confirm its explanatory potential with respect to the materiality phenomenon, including how unaccountable behaviour is dealt with in materiality practices.

1. Introduction

Materiality is a cornerstone concept in accountancy [1,2,3]. It refers to a rationale that resources, efforts, and time should be allocated to what really matter [4,5,6,7]. In a financial context, materiality signifies a requirement for the firm to report all financial issues that substantively have an impact on shareholders’ investment decisions [8,9]. Since the 1990s, the concept of materiality has been generalised into a sustainability context that is intended to function with respect to assessing, managing, and reporting environmental and social issues that generate significant impacts for stakeholders [10,11].

Although materiality is emphasised as a foundational principle in sustainability accounting standards [12,13], the current state of understanding is deficient in terms of constructing an effective explanation of the materiality phenomenon if certain aspects are taken into consideration, which include stakeholder heterogeneity, reporting–management association, and consistency, as well as clarification regarding to whom (that is, stakeholders, the organisation, or both) sustainability issues matter. In particular, the current conception of materiality is limited with respect to explicating the cases that appear to contradict the rationality of the materiality principle (that is, taking action on what matters). This is because, in such cases, firms do not provide, or even intentionally refuse to provide, adequate actions in response to issues that might have significant impacts.

The aforementioned deficiencies not only confine the ability of researchers and practitioners to understand the complex materiality phenomenon emerging from sustainability practices, but they also militate against maximising the potential of this concept in reporting and managing sustainability issues. The primary objective of this study, therefore, is to provide a conceptualisation that resolves these deficiencies. The outcome of this study is a new explanatory model called the heterogeneity–accountability materiality model (HAM model), which provides a rational basis for the reporting and management of material issues involving many different stakeholders. A more thorough understanding of materiality is developed herein using an illustrative example based on Exxon, which demonstrates a practical application of the HAM model, as well as the importance of transparency of information flows to stakeholders to ensure that reporting is not limited to simply those events regarded as material by the organisation as the account giver.

This study proceeds with a discussion of materiality, accountability, and stakeholder heterogeneity, followed by the derivation of the HAM model, which specifies the underlying process of materiality within a heterogeneous stakeholder environment. This leads into the illustrative example of Exxon, which is followed by consideration of the implications of this study for sustainability accounting theory and practice into the future.

2. Materiality and Its Sustainability Applications

2.1. Materiality and Its Basic Logic

In essence, materiality is a social phenomenon used to distinguish what is significant from that which is trivial, thereby guiding action and resource allocation to the resolution of issues that matter [5,6]. Materiality is observable in daily life, regardless of whether the social actors are aware of the accounting version of the concept. As Bernstein (1973, p. 68) [5] states, “[t]he concept of materiality is part of the wisdom of life. Its basic meaning is that there is no need to be concerned with what is not important or with what does not matter”. Likewise, Hicks (1964, p. 158) [6] explains materiality as “if it doesn’t matter, don’t bother with it”, and refers to this as “the elementary proposition of materiality”. Similar to Hicks, APB (1995, para. 3) [4] defines materiality as “an expression of the relative significance or importance of a particular matter”, while [7] defines materiality as “an index of time and trouble in relation to the amount in question”. As a cornerstone concept in accountancy, materiality refers to the rationale that resources should be allocated to that which really matters [1,2,3].

Materiality applies to both financial and sustainability contexts. Within a financial accounting context, the application of the materiality concept requires the firm to accurately report all financial issues that would substantively impact a current or potential shareholder’s investment decision [8,9,14,15]. Hicks (1964, p. 159) [6] locates the application of materiality to managerial decision making, beyond its traditional reporting scope:

… [M]ateriality is not solely an accounting consideration, as some of the literature on the subject would seem to suggest. In fact, the concept is widely and frequently used. For example, when a business executive, applying the technique of “management by exception”, cuts through to the matters of significance, he is recognizing materiality.

2.2. Materiality in Sustainability Accounting

Materiality constitutes a foundational principle in sustainability accounting standards (GRI, AA1000APS, IIRC and SASB). For example, in AA1000APS (AccountAbility, 2008, p. 12) [16], a material issue is defined as “an issue that will influence the decisions, actions, and performance of an organisation or its stakeholders”. In GRI G4 (2013, p. 17) [12], material issues are defined as those that “the report should cover … [issues which] reflect the organisation’s economic, environmental and social impacts, or substantively influence the assessment and decisions of stakeholders”.

GRI’s materiality assessment method is a two-variable model, where one variable measures influence on stakeholder assessment and decision, which corresponds to external factors, while the other variable measures the significance of economic, environmental, and social impacts, which refers to internal factors or influence on the organisation. The GRI framework provides an example of three materiality levels in response to three reporting priority levels. The least material (immaterial) issues that correspond to low reporting priority should not be included in the report. The moderate material issues that are assigned a medium-level of reporting priority may be considered for inclusion in the report, but are given less attention. The highly material issues are deemed to be high-level reporting priorities, and should be reported in great detail, be given more attention, or even form a theme for the report itself [12,17].

The application of materiality in sustainability accounting ranges beyond the reporting function to include the management of environmental, social, and economic issues regarded as significant by stakeholders. This managerial function is evident in AccountAbility (2006, p. 19) [10], which states that materiality enables firms to “seek to manage their performance, and look to communicate effectively to their stakeholders”, and emphasises that materiality is used to “direct, rather than (merely) reflect performance” (2006, p. 51). Refs. [18,19] connect the materiality decision process to both management strategy as well as the reporting system, while Lydenberg (2012, p. 63) [20] states that the effective identification of material sustainability issues “will create opportunities for increased operating efficiencies and cost savings as managers monitor sustainability challenges that address inefficient operations”. Therefore, traditional financial materiality assessment focuses, for the most part, on the significance of financial items to shareholders. In contrast, refs. [10,12] recommend materiality assessment from a combined assessment of significance to stakeholders and significance to the organisation. This is because stakeholders, such as workers, shareholders, and suppliers, are invested specifically in the success of the organisation, and, hence, their interests and expectations are significant not only for stakeholders, but also the organisation [12].

2.3. The Theoretical Deficiencies

There are clear deficiencies that remain unresolved with reference to the current state of conceptualising materiality practices pertaining to sustainability issues. First, previous work on materiality has not addressed the fact that there will be various stakeholders with heterogeneous interests that require appropriate responses. Instead, the various different stakeholders will be taken as a whole when it comes to assessing the impacts of issues. This view of non-differentiation, however, reinforces the concerns regarding subjectivity that are inherent in sustainability applications of materiality [10,21], with this outcome being due to the account giver’s bias and partiality, which is inevitably embodied in any consideration regarding which stakeholders should be included in the assessment, and how accounts are prioritised for different stakeholders.

The results of Geerts and Dooms’s research (2020) [22] show that the expected content of a sustainability report is viewed differently by various stakeholder groups in terms of the relative importance of the dimensions of the triple bottom line (TBL), as well as in terms of the specific indicators representing material issues. Furthermore, the concept of boundary setting with respect to the different dimensions of the TBL and the desired level of inclusion by stakeholders during the development of a sustainability report may vary owing to differing subjective interpretations. Garst, Mass and Suijs (2022, p. 80) [23] criticise the traditional materiality model as “an oversimplified picture of reality”, given that all stakeholders are grouped on one axis, which means that “the heterogeneity of stakeholder views is lost”. Accordingly, the same authors recognise that, in the field of contemporary sustainability accounting, “materiality assessment is an art, not a science”, especially given that “representing the complex reality” remains “a challenge” for those providing accounts of sustainability outcomes (p. 84).

Second, previous works have not established a logic to underpin the association of reporting and management functions of materiality. Materiality has been solely conceptualised through a reporting focus (e.g., [12,13] and has been traditionally applied to determine what should be reported across different industrial settings (Jorgensen, Mjos, and Pedersen, 2022) [24]. Scholars have also extended materiality to a managerial function [11,18,25], which has resulted in improved sustainability performance in the wind power industry (Whitehead, 2017) [26] and better sustainability management strategies in the mining industry (Saenz, 2019) [27]. However, the relationship between reporting and managerial functions remains very much underexplored in the materiality literature, with no clear explanation as to how a change in the reporting of material sustainability issues might impact the management of those issues.

The third deficiency concerns clarification of “what matters to whom”. Previous materiality works have not achieved consensus on the subjects impacted by sustainability issues, but are open to three different ways of dealing with this, including “stakeholders or the firm” [10,12], “stakeholders and the firm” [18], and “only stakeholders” [25,28]. CGA-Canada argues that the two indices (impacts on stakeholders and the firm) should not be accorded equal status; instead, material issues “should be defined in the context of what is of importance to stakeholders and not the organization” (CGA-Canada, 2006, p. 4) [28]. In addition, GRI G3 (2000–2011) [17] states that the two assessment indices overlap to some extent, because stakeholder interests and expectations are also significant for the organisation itself.

Wu et al. (2018) [29] argue that proliferation of sustainability reporting does not ensure the consistent and accurate reporting of material issues. Their study reviews definitions of materiality and proposes methods for screening or identifying material sustainability issues using publicly available resources. They found that most acknowledged standards and initiatives diverge in their definitions regarding an approach towards considering materiality. An associated argument has developed in the recent literature concerning conflicts between “double materiality” and “single materiality” with respect to selecting reporting content (Baumuller & Sopp, 2022; Delgado-Ceballos et al., 2023) [30,31]. The double materiality principle privileges the notion of “matters to either the firm or stakeholders”, thereby allowing reference to issues that generate either financial values for the firm’s development, or environmental and social impacts on stakeholders. In contrast, the single materiality principle requires issues to matter to both the firm and to its stakeholders. Given the co-existence of these two conflicting views in theory and practice, scholars have called for further exploration of the connection between what matters to the firm and what matters to stakeholders in order to achieve a consistent, inclusive, and useful sustainability report (Baumuller & Sopp, 2022) [30]. Therefore, a deeper understanding of the linkage between ‘matters to stakeholders’ and ‘matters to the firm’ is required to advance the overall conceptualisation of materiality.

Fourth, although the materiality logic is regarded as an application of rationality with respect to “matching” resources allocated to what matters [6], the current conception of materiality has not achieved a convincing understanding of the “mismatch” phenomenon, in which the rationality of materiality appears to be contradicted. With this phenomenon, an organisation fails to (or intentionally refuses to) make sufficient reporting and managerial responses to an environmental and social issue that is material to stakeholders. For example, scholars have observed that, even though firms might acknowledge their reporting obligations, they sometimes attempt to limit disclosure, especially when information is significant to stakeholders, but has the potential to impact negatively on the relationship of the firm with its stakeholders (Captuo et al., 2021; Pizzi et al., 2021) [32,33]. According to Slacik and Greiling (2020) [34], materiality does not consistently guide reporting practice and is not taken seriously, while the mediocre quality of coverage and communication in sustainability reports shows that stakeholders’ information needs are not being considered adequately. Accordingly, the integration of sustainability into non-financial reports is not simply a matter of course, even for firms claiming to have adopted sustainability as a core principle (Pizzi et al. 2021) [33]. From this perspective, the mismatch phenomenon can provide an opportunity to renew our understanding of the challenges presented by existing models of sustainability.

To address the aforementioned theoretical deficiencies, we integrate accountability and stakeholder heterogeneity into a new model, in which we develop a more effective explanation of the materiality phenomenon in sustainability accounting.

3. Accountability and Stakeholder Heterogeneity

In this section, we connect materiality to accountability and stakeholder heterogeneity, which leads to the conceptual foundation of the explanatory model proposed in the following section.

3.1. Materiality and Accountability

The connection between materiality and accountability is addressed in ([35] p. 8), which defines materiality as “the potential to adversely affect the discharge of accountability by the management or governing body of the entity”. While financial accounting concerns the discharge of accountability to capital providers, the objective of sustainability accounting is to discharge accountability for environmental and social impacts to stakeholders [36,37]. This perspective, which embraces the notion of materiality as a form of sustainability accounting practice, is necessary in order to discharge a firm’s accountability to its stakeholders; however, there is minimal discussion connecting materiality to accountability in the previous literature.

An accountability-related issue pertains to the principle of “inclusivity”, which concerns identifying the stakeholders to whom an issue is (or should be) significant. The classical stakeholder approach, which classifies primary and secondary stakeholders according to the strength of their influence on the organisation [38,39], identifies “key” stakeholders as those that should matter to the firm. Such an approach is supported by materiality research, such as that of [40]. In contrast, refs. [41,42] contend that it is problematic for accountability practices to concentrate on primary stakeholders, especially given that “inclusiveness” is a precondition of accountability according to [12,43]. Inclusivity means that the firm is accountable to all stakeholders, rather than just some [12,43]. As a result, this study develops a model of materiality that supports the discharge of the firm’s accountability as it pertains to sustainability to all, rather than just key stakeholders.

3.2. Stakeholder Heterogeneity

For our model, we incorporate the well-established belief within the stakeholder management literature that stakeholders are heterogeneous in their expectations, interests, and identities [44,45,46,47]. According to stakeholder mobilisation theory, a stake is regarded as a particular end state that stakeholders will protect, where necessary, by taking action in order to influence the firm (Rowley & Moldoveanu 2003, p. 207) [47]. The distinction of specificity in stakes informs the categorisation of groups of persons as separate stakeholder groups with their own specific expectations, interests, values, and resources, all of which determines what mobilisation action(s) they take to achieve their goals.

Given their divergent interests and differing identities, stakeholders evaluate the firm’s performance and exert influence on the firm to ensure that it meets their own specific interests and expectations [46,47,48]. These are not exclusively economic interests, but extend to the firm’s sustainability performance. The inherent heterogeneity of stakeholders is emphasised as they mobilise differently in response to the same issue to protect their diverse interests. Davenport and Leitch (2006) [49] contend that, because different stakeholders have different interests, the likelihood of taking action varies between stakeholders. Where the impact of an issue on the interests or social identity of a stakeholder group is high, it is more likely that the group will mobilise and take action against the firm. Mobilisation, therefore, is a consequence of an issue being material to a stakeholder.

The heterogeneity phenomenon is exemplified in O’Dwyer’s (2006) [50] study, where it is reported that it is difficult for different stakeholder groups to achieve consensus regarding certain issues. Although it may be assumed that shareholders hold a primary interest in common, that of maximising wealth, stakeholder heterogeneity suggests that this may matter to some stakeholders more than it does to others, with some stakeholders being potentially more interested in a firm’s social or environmental performance. This highlights the need to reconceptualise materiality as it is applied in sustainability accounting in order to recognise stakeholder heterogeneity, rather than erroneously substituting stakeholders for shareholders within standard definitions of materiality in the context of financial accounting. In such definitions, the core component of ‘influence on stakeholders’ is derived in a simplistic way and corresponds to its financial counterpart of ‘influence on shareholders’ (e.g., see [12,16,51]), an action which fails to account for the complexity that is inherent in the heterogeneous stakeholder environment.

4. The Heterogeneity–Accountability Materiality Model

In this section, we have connected materiality to accountability, sustainability accounting, and stakeholder heterogeneity, which forms the theoretical foundation for this study. The heterogeneity–accountability materiality model (HAM model) reveals the underlying structure of the process, through which firms and their stakeholders interact to resolve sustainability issues, and which leads to the discharge of the firm’s accountability to many heterogeneous stakeholders.

4.1. The Elementary Materiality Process

Here, we construct the materiality process based on the process of accountability. As is consistent with other accountability scholars (e.g., [52,53,54,55,56], we argue that the logic of accountability flows from “a principal-agent view”. In this context, an agent is required to justify its actions or intended actions to a principal, with failure to do so resulting in negative consequences for the agent. This conception of accountability identifies interaction between (a) a principal with enforcement power, and (b) an agent answerable to the principal [57,58], together with the existence of an “action” and an “account”. Here, the agent is required to (a) act (or not to act) to resolve the issue, and (b) give an account to the principal. Action is, therefore, a product of the internal management system, and the account is produced by the external accounting system, where the principal’s expectations guide the agent’s actions [59,60,61,62].

Lindberg (2013) [63] translates a course of practices, as presented in accountability definitions, into a stylised timeline continuum, as is depicted in Figure 1.

Gray et al. (2014, p. 50) [57] propose a similar model (Figure 2), which defines accountability as “the duty to provide an account or reckoning of those actions for which one is held responsible”.

According to [52,57,63], the discharge of accountability encompasses four successive steps:

(1)

The agent understands the interests and expectations of the principal who has enforcement power over the agent;

(2)

The agent takes action or refrains from taking action in response to an event;

(3)

The agent gives an account about their action or non-action to the principal;

(4)

Giving a false or inaccurate account would lead to the agent being sanctioned by the principal.

In contrast, the process by which the application of the materiality principle enables the firm to discharge its accountability to stakeholders can be summarised in the following five steps:

• An issue matters to the principal, so the principal should make an effort to resolve the issue according to the basic logic of materiality;
• The principal is unable to resolve the issue directly and therefore engages the agent to do so, which triggers an accountability relationship where the agent is expected to resolve the issue and provide an account to the principal regarding the action taken;
• If this account is not forthcoming, or if the account does not satisfy the principal, the agent would be sanctioned by the principal; hence, this issue matters to the agent;
• As the issue matters to the agent, the agent should take action to resolve the issue and, as required by the accountability relationship, an account must be given to the principal to explain the action taken;
• If the agent’s account fails to satisfy the principal and the issue remains unresolved, this issue will still matter to the principal, who would continue to sanction the agent until satisfied with the action taken and the account given.

Through this five-step process, four objects (principal, agent, issue, and account) are structured to realise the discharge of accountability, as is illustrated in Figure 3.

The dual dimensions of materiality are depicted in this model, which differentiates the managerial system from the informational system and conforms to the accountability conception in which both an “account” and “action” are required in response to material issues [59,61,64]. In this materiality system, managerial action aims to resolve, control, or mitigate the negative impacts of each material issue, including the decision to take no action if the event is not regarded as material. The informational system includes the principal, agent, and account, where the agent provides an account to the principal of action taken/not taken to resolve material/immaterial issues. In Figure 3, we identify the connection between management and reporting, which necessarily requires both management and reporting actions in response to stakeholders regarding material sustainability issues.

This elementary materiality process transfers that which matters to the principal to that which matters to the agent, where the chain of actions begins with an issue that matters to the principal (stakeholders) and, given the principal’s enforcement power, also matters to the agent (organisation or firm). Significance to the firm, therefore, results from significance to stakeholders, which is measured by the strength of action taken by stakeholders in response to an issue [12]. It is this causal relationship that justifies the significance to stakeholders as the primary indicator of materiality. From the materiality process, we form a new perspective regarding the debate over “matters to whom”, which is discussed in Section 2.3 as a deficiency in the current conception of materiality. We conclude that, if an issue matters to stakeholders, this causes the issue to also matter to the firm. This conclusion supports the view of stakeholders being the primary gauge with respect to measuring the materiality of an issue [17,28], but rejects the “only stakeholders” view that excludes the firm from the materiality scope.

4.2. The Transparency Condition for the Materiality Process

Accountability relationships are activated by the condition of transparency, which “requires allowing people to see into systems and to understand the reasons for decisions taken” (Osborne, 2004, p. 292) [65], thereby providing visibility of the agent’s actions to the principal [66]. Accountability does not function effectively in an opaque situation, as the principal would not know what action an agent takes [66]. In the view of [67,68], transparency typically occurs in situations before and after action is taken, and, therefore, the condition of transparency enables the principal to observe change resulting from the agent’s action and to benchmark the agent’s performance against their standards and expectations. Similarly, Ebrahim and Weisband (2007) [52] contend that accountability relies on the presence of transparency, as the principal is able to evaluate the agent’s performance and decide whether to sanction the agent only under conditions of sufficient transparency.

The condition of transparency is achieved by monitoring control methods, such as auditing and assurance practices, that enable the principal to verify actions taken by the agent [69,70,71]. This prevents the agent from deceiving the principal in pursuit of self-interest, rather than prioritising the interests of the principal [72,73]. The essential role of transparency within accountability [52] is addressed in the elementary materiality process described by the five-step process (Figure 3). Only when the principal knows of an issue and its significance can they take appropriate action, including an evaluation of whether the agent has responded effectively to the issue and provided an accurate account, or whether enforcement action against the agent is needed.

4.3. The Heterogeneity–Accountability Materiality Process

The next stage in the development of the HAM model is to adapt the elementary materiality process illustrated in Figure 3 to the heterogeneous stakeholder environment where stakeholders take action (mobilise) in response to an issue they regard as significant—which requires the firm to give an account of its performance regarding this issue. As Rowley (1997, p. 890) [74] states: “[f]irms do not simply respond to each stakeholder individually; they respond, rather, to the interaction of multiple influences from the entire stakeholder set”. Figure 4 depicts the HAM process, where N stakeholders identified as S1, S2, …, Sn, hold heterogeneous expectations of the firm.

In Figure 4, Step ① depicts an issue that stakeholders cannot resolve that impacts differently on N stakeholders, owing to their heterogeneous interests and identities. Step ② depicts varying levels of strength of stakeholder action, where, collectively, stakeholders hold the firm accountable with respect to the issue. In Step ③, the firm takes managerial action in response to the issue, while, in Step ④, the firm gives an account to stakeholders of both the issue and action taken to resolve the issue. In Step ⑤, stakeholders assess whether the managerial action taken has resolved the issue. If the issue is not resolved, stakeholders continue to mobilise against the firm. If the issue is resolved, mobilisation ceases.

The heterogeneity paradigm is assimilated into this process without discriminating between powerful or weak stakeholders by applying the principle of inclusivity, where the material concerns of all stakeholders are treated as legitimate [21,75,76,77] and taken into consideration in Steps ①, ②, and ⑤ of the five-step process depicted in Figure 4.

4.4. The Materiality Web and Rules

Figure 5 illustrates heterogeneous stakeholder materiality in the form of a “materiality web”.

The firm is represented at the intersection of all axes, each of which symbolises the strength of the action exerted by each stakeholder, S1, S2 … Sn, against the firm. The significance of a particular issue to any one stakeholder is designated at one point along the stakeholder action axis (continuum). This point symbolises the action taken by the stakeholder in response to the issue. Therefore, positions S1 (θ), S2 (θ), …, Sn (θ) symbolise the actions taken by S1, S2 … Sn, respectively. These positions on each continuum are connected to form a web, which denotes the overall materiality of the issue to the firm.

Although there is concern regarding the subjectivity inherent in applying the materiality principle to sustainability accounting [10,21], our HAM model reveals a reality pertaining to how issues are prioritised according to their level of materiality. By considering the condition where one stakeholder only is provoked into taking different action in response to two issues, with all other stakeholders remaining unchanged in their action positions, we are able to derive Rule 1.

Rule 1.

Suppose two issues, α and β, occur. Issue α is more significant to one stakeholder than issue β, while the significance of issue α is the same as β to all other stakeholders. Thus, issue α mobilises this one stakeholder to a higher action position against the firm than for issue β, while other stakeholders remain unchanged in their action positions with regard to both issues. (The phrase “stakeholders remain unchanged in their action positions with regard to both issues” does not mean that these stakeholders take no action to both issues; rather, it means that they may take action, but that their actions are identical in response to both issues.) This leads to issue α being more material to the firm than issue β. Therefore, the firm would prioritise issue α over β in both its management and reporting practices.

Rule 1 is presented in Figure 6 using two materiality webs, where the web for issue α encloses the web for issue β, which is symbolic of issue α being more material than issue β to the firm; hence, the firm prioritises issue α over issue β.

Rule 1 results from this condition of singular stakeholder movement and reflects a rationale that guides the resource allocation activities of the firm. Given the unlikely scenario of just one stakeholder changing their mobilisation position, this rule can be expanded to Rule 2.

Rule 2.

If all stakeholders are consistently mobilised in equal or higher action positions for issue α relative to issue β, issue α is more material than issue β, and the firm should prioritise issue α relative to issue β in managerial and reporting practices.

Rule 2 is presented in Figure 7 showing two materiality webs, where the web for issue α encloses the web for issue β.

The condition invoking Rule 2 is consistent change, which occurs when each stakeholder is in an equal or higher action position for issue α relative to issue β. The operation of Rule 2 is demonstrated in the next section using the illustrative example of Exxon.

5. An Illustrative Case

In this section, the explanatory HAM model, including the HAM process, rules, and transparency condition, is illustrated using selected sustainability events involving the multinational corporation Exxon.

Exxon was chosen for this analysis because of its leading role in the oil industry and its direct involvement in events that have had significant environmental and social impacts. This study draws on secondary data generated by the selected events available in the public domain and which are accessible online, with such information identifying the actions taken by various stakeholders, and the response to these actions by Exxon. The four events selected are: the Exxon Valdez oil spill; the Montana oil pipeline leak; the funding of weapons purchased for the civil war in Angola; and the sponsorship of global warming sceptic groups. As Exxon was exposed to external monitoring from media, government investigations, and NGO and community monitoring, its actions with respect to these issues were transparent at varying levels to stakeholders, thereby enabling the observation of the impact of transparency on the activation of the HAM process.

There are two stages to this empirical analysis. First, we assess the materiality of each of the four events (using the assessment criteria shown in

Table 1. Assessment of materiality to stakeholders.

Criteria of Assessing Stakeholder Actions
Rating Action Level	The media (S1)	The public (S2)	Government (S3)	NGOs (S4)	Community (S5)
1 Inaction	No report	No rejoinder	No sanction	No social movement	No engagement
2 Concern	Neutral media reporting	Some concern; relatively neutral attitude	Warning; letters from government officials to Exxon	Press releases; Website reporting	Official complaints; continual monitoring
3 Moderate action	Exposed and criticised in some influential media	Moderate criticism	Official governmental critique; moderate punishment	Localised campaign	Target blame at and/or demonstrate against Exxon
4 Strong action	Extensive long-term negative reporting	Extensive unified concern; public blame	Significant punitive measures	Large-scale campaigns; organise consumer boycott; litigation	Community intervention; litigation

and

Table 2. Assessment of materiality to Exxon.

Criteria for Quantifying the Firm’s Managerial Action
Managerial action level	Strength of managerial action
Decisive action (level 4)	Large investment in solutions
Moderate action (level 3)	Moderate investment in solutions
Minimal action (level 2)	Minimal action or expressed intention to act
Inaction (level 1)	No evidence of action; indifference
Criteria for Quantifying the Firm’s Reporting Practice
Reporting level	Reporting practice
Extensive (level 4)	Special purpose themed report
Moderate (level 3)	Included in standard reports
Minimal (level 2)	Public statement by executive
None (level 1)	No formal reporting or information releases

) and present relative materiality in both account and graphical form. Second, we test the HAM model and identify the operational situations of the HAM process.

5.1. Materiality Accounts

This analysis captures fewer than the complete range of stakeholders. However, if a stakeholder group is significantly mobilised in response to any of the four events, we assumed that the stakeholder actions were reported in the media. Hence, stakeholders not included in the following analysis were assumed to be in a state of constant indifference towards the selected events. From the available data, we observed five broad stakeholder groups, which were assumed to act as single entities: S1, the media; S2, the public; S3, the government; S4, NGOs (non-government organisations); and S5, local community. Note that shareholders and investors were not included, as there was no strong evidence to suggest that these stakeholders mobilised against Exxon in response to these four events.

Table 1 and Table 2 show the assessment criteria used to measure the strength of action taken by stakeholders and Exxon management, and, hence, the materiality of each event and the rationale underpinning our rating for each action level.

The Exxon Valdez oil spill, regarded as one of the most devastating human-caused environmental disasters, occurred in Alaska on 24 March 1989. A large oil tanker, the Exxon Valdez, struck Bligh Reef and spilled approximately 11 million gallons of crude oil. The oil eventually covered 2100 km of coastline and threatened the habitat of multiple species, including salmon, sea otters, seals, seabirds, and other wildlife.

Table 3. Exxon Valdez oil spill.

Stakeholder	Actions of Exxon Stakeholders and Management	Assessment
1 Media	Frequent, extensive, and critical reporting overwas conducted over a long period by major news networks directly blaming Exxon.	strong (4)
2 Public	Widespread public concern was unified as strong negative opinion against Exxon.	strong (4)
3 Government	Exxon Valdez Oil Spill Trustee Council established to manage environmental damage and restoration. Court ruled that Exxon was to pay punitive damages of USD 507.5 million.	strong (4)
4 NGOs	NGOs voiced extreme concern regarding the impacts of the oil spill; located offices in Alaska; and engaged the media to criticise and campaign internationally against Exxon.	strong (4)
5 Local community	Alaskan residents worked throughout the region to restore the ecosystem. Local resident groups were formed to protect local fisheries.	strong (4)
Managerial	Exxon acted promptly by paying large fines and funding clean up and restoration costs; it engaged in active public relations through media announcements.	decisive (4)
Reporting	Exxon responded promptly to the oil spill event by releasing information and advertising through the media. Exxon maintains a section of its website dedicated to the Exxon Valdez event.	extensive (4)

summarises the issue of the Exxon Valdez oil spill.

On 1 July 2011, Exxon’s Montana pipeline burst and leaked an estimated 42,000 gallons of oil into the Yellowstone River. Exxon removed water contaminated by oil to a refinery in Billings for storage, thereby prompting the US EPA to advise that the source of the leak had been repaired by 18 July 2011.

Table 4. Montana pipeline oil leak.

Stakeholder	Actions of Exxon Stakeholders and Management	Assessment
1 Media	Extensive reports by major US media outlets (CNN, Fox, New York Times, Wall Street Journal), which all took a strong critical stance against Exxon on the grounds of public safety and environmental protection.	strong (4)
2 Public	No evidence of social campaigns or boycotts.	concern (2)
3 Government	Montana’s governor declared a state of emergency and criticised Exxon for failing to respond quickly and effectively. The US EPA requested immediate clean up measures and safety improvement.	moderate (3)
4 NGOs	NGOs (Greenpeace, Sierra Club, Friends of the Earth) criticised Exxon’s slow response and demanded accountability, full disclosure, and improved environmental management practices. There was no evidence of consumer campaigns or boycotts, nor litigation involving large compensation.	moderate (3)
5 Local community	Although the risk level was uncertain, local communities reliant on the river for farm irrigation and drinking water were concerned about pollution. They blamed Exxon and requested an immediate clean up.	moderate (3)
Managerial	Exxon agreed to pay USD 1.6 million in compensation and agreed to clean up contaminated area.	moderate (3)
Reporting	Exxon disclosed relevant information to the public through official media statements and in the Exxon 2011 corporate citizen report.	moderate (3)

summarises the issue of the Montana pipeline oil leak.

During the Angolan civil war from 1975 to 2002, Exxon, BP, and Elf made signature bonus payments to obtain drilling licenses for offshore oil concessions. A total of USD 870 million from this fund was used by the Angolan government to purchase weapons. The Angolan foreign minister acknowledged that these oil funds were used to fund the war effort. The connection between the drilling license purchases and the funding of the Angolan war is not definitive, although the possibility of this link has been acknowledged in media reports. The relevant US legislation is the Foreign Corrupt Practices Act (FCPA).

Table 5. Angolagate scandal.

Stakeholder	Actions of Exxon Stakeholders and Management	Assessment
1 Media	Media were critical of scandal without directly blaming Exxon.	moderate (3)
2 Public	No evidence as to the public’s response.	inaction (1)
3 Government	The FCPA exempted payments that facilitated routine government actions; the US Department of Justice attorney ruled that the exemption covered Exxon’s signature bonus payments, and took no action against Exxon. The US oil industry became concerned about the competitive impact of anti-bribery legislation.	concern (2)
4 NGOs	The Human Rights Watch and Global Witness expressed dissatisfaction concerning the lack of transparency regarding the payments. The International Consortium of Investigative Journalists’ investigation resulted in strong and clear criticism of Exxon for engaging in bribery for commercial gain.	moderate (3)
5 Local community	There was no observable connection of this event to relevant local communities	inaction (1)
Managerial	Lee Raymond (the then head of Exxon) explained that, although he was unsure whether the governance record of the Angolan leaders was satisfactory, Exxon had scrupulously observed its contract confidentiality with the government, thereby suggesting that Exxon recognised the potential significance to stakeholders of the Angola scandal. No evidence of direct action taken by Exxon is available.	minimal (2)
Reporting	No reporting or disclosure of this event. Lee Raymond made a public statement that it was not the company’s role to disclose how its foreign investment funds were spent.	minimal (2)

summarises the issue of the Angolagate scandal.

Between 1998 and 2006, Exxon was accused of providing approximately USD 23 million to fund organisations that were sceptical of the science linking global warming to the burning of fossil fuels.

Table 6. Funding of organisations promoting global warming scepticism.

Stakeholder	Actions of Exxon Stakeholders and Management	Assessment
1 Media	There was no extensive media coverage of this event, although there was some criticism of Exxon in The Guardian and in the Wall Street Journal.	concern (2)
2 Public	There is no evidence of general public concern regarding this event.	inaction (1)
3 Government	Senators Rockefeller and Snowe wrote to Exxon expressing concern on moral and scientific grounds. There is no evidence that the US government required Exxon to cease or disclose this funding.	concern (2)
4 NGOs	The UK Royal Society released an open letter in 2007 asking Exxon to stop this funding. The Union of Concerned Scientists compared Exxon’s tactics to those used by the tobacco industry, and criticised Exxon for misrepresenting and denying climate change science. There is no evidence of consumer campaigns, boycotts, or litigation.	concern (2)
5 Local community	There is no specific local community relevant to this issue.	inaction (1)
Managerial	Exxon continued to sponsor 41 climate sceptic organisations.	inaction (1)
Reporting	Exxon’s manager for public affairs, Kenneth Cohen, stated in 2006 that funding of the Competitive Enterprise Institute and “a handful” of similar groups had ceased, but he refused to reveal names of these organisations. However, given the continued funding of denial groups, the minimal information disclosed by Exxon was vague and misleading.	none (1)

summarises the issue of Exxon funding organisations promoting global warming scepticism.

5.2. Materiality Practices in a Heterogeneous Stakeholder Environment

Table 7. Summary of materiality assessment for the two Exxon events.

Event	S1	S2	S3	S4	S5	U1	U2	…	Ug	M	R
1	4	4	4	4	4	c1	c2	…	Cg	4	4
2	4	2	3	3	3	c1	c2	…	Cg	3	3
3	3	1	2	3	1	c1	c2	…	Cg	2	2
4	2	1	2	1	1	c1	c2	…	Cg	1	1

presents an assessment of the materiality of the four Exxon events that applies the assessment criteria summarised in Table 1 and Table 2. This table includes action positions of the five recognised stakeholders (S1 to S5) and the assumption that all unrecognised stakeholders (U1 to Ug) remained in neutral action positions with regard to these events. Exxon’s managerial (M) and reporting (R) actions are also included in Table 7.

As is displayed in Table 7, for each observed stakeholder, a diminishing level of stakeholder action was evident sequentially from Event 1 to Event 2 to Event 3 to Event 4. This table is depicted in Figure 8 as a materiality web.

Both Table 7 and Figure 8 demonstrate that the conditions specified for Rule 2 to operate were satisfied with regard to these events or issues; that is, it can be observed both numerically and graphically that the materiality of Event 1 > Event 2 > Event 3 > Event 4. According to Rule 2, the HAM model would, therefore, predict that Exxon would prioritise the allocation of resources to its response to Event 1 ahead of Events 2, 3 and 4; prioritise Event 2 ahead of Events 3 and 4; and prioritise Event 3 ahead of Event 4. This predicted prioritisation was, in fact, evident in our observations made of Exxon’s actions, which were measured at a level of 4, 3, 2, and 1, respectively, for Events 1, 2, 3, and 4 for both Exxon’s managerial and reporting actions in response to each event. That is, the data from the four Exxon events empirically illustrated the HAM model, which presents the prioritising activities in a heterogeneous stakeholder environment, with an association with sustainability reporting and management, as well as consistency between reporting level and management action.

5.3. Explaining the “Mismatch” Phenomenon

In this section, we use the HAM model to explain the “mismatch” phenomenon that appears to contradict the rationality of materiality; that is, where a firm’s reporting and/or managerial actions do not match the materiality of a sustainability issue. We examine the effect of transparency in each of the four events. Events 1 and 2 are used to illustrate the routine operation of the HAM process under transparency conditions, whereas Events 3 and 4 (both relating to the mismatch phenomenon) are used to demonstrate the absence of transparency and the resulting dysfunction of the HAM process.

The HAM process, as is depicted in Figure 4, was described as a five-step process that includes the issue, account, materiality assessment, and action taken by stakeholders and the firm, and was predicted to operate according to Rules 1 and 2. Where the conditions of transparency and accountability are met, we can observe managerial reporting and action following the process described by the HAM model. Both the Exxon Valdez and Montana oil spills illustrated this routine operation of the HAM process. During the Exxon Valdez event, widespread media reporting activated the transparency condition. This triggered stakeholders to mobilise and apply pressure on Exxon to resolve the crisis. Given that the impact of this event mattered to stakeholders, it also mattered to Exxon; hence, the event was assessed as material. Since Exxon would be held accountable, it allocated resources according to the principle of Rules 1 and 2. Central to the routine operation of the HAM process is the activation of the transparency. This is further emphasised in Events 3 and 4, where the HAM process was dysfunctional when failing to meet the transparency condition.

The Angolagate event provides an example where transparency was exempted within the legal jurisdiction. Given that the US Foreign Corrupt Practices Act (1977) (FCPA) exempts payments to facilitate or expedite the performance of routine government actions, and the US Department of Justice attorney ruled that the exemption covered Exxon’s signature bonus payments, Exxon did not disclose these payments. However, BP fully disclosed its signature bonus payment of USD 111 million to the Angolan government in exchange for drilling licenses. Both Exxon and BP were extremely similar in many aspects, including business operation and size, and both endured similar criticism from the public, media, and NGOs. However, US law protected Exxon from potential accountability pressure from stakeholders by enabling the non-disclosure of the signature bonus payments, whereas the UK government promoted a publish-what-you-pay campaign [78]. Hence, the type of payments regarded as material by BP when operating in the UK are not disclosed by Exxon, given that it operates within a different legal jurisdiction.

That Exxon’s non-disclosure appears to contradict the basic logic of materiality is a typical “mismatch” phenomenon, where Exxon intentionally refused to report and manage the event that seemed significant. This non-disclosure event can be explained by dysfunction of the HAM process owing to a lack of the transparency condition, as transparency was exempted by US law as interpreted by the US Department of Justice. In contrast, the HAM process was operational within the UK, thereby prompting BP to report the signature bonus payments to stakeholders. Hence, the routine operation of the HAM process was impeded by the opacity of information concerning signature bonus payments made by Exxon. This is because the HAM process is fully operational only when the information flow concerning the event is fully transparent.

The significance of the transparency condition to the activation of the HAM process was again illustrated in Event 4. Exxon management decided not to disclose the funding of climate change sceptic organisations to stakeholders, and, therefore, experienced no external pressure to change this practice. The public did not become aware of this funding until action was taken by the UK Royal Society in 2007 and reinforced by other stakeholders (refer Table 6). Finally, in May 2008, ExxonMobil pledged in its annual corporate citizenship report that it would cut funding to several public policy research groups whose views on climate change could divert attention away from the urgent need to address anthropogenic climate change. According to a 2012 Frontline interview, Exxon ceased funding climate change sceptics by 2009.

The HAM process did not lead to the disclosure of material payment transactions, as the opacity of the information flow enabled these payments to remain a commercial secret until 2007. However, once these funding activities became transparent, the HAM process became operational, and Exxon was held to account by its stakeholders, with management being forced to report on and cease funding climate change sceptics. This event also illustrates the application of the essential principle underlying the accountability relationship inherent in the HAM process; that is, ‘what matters to stakeholders matters to the firm’. Indeed, once stakeholders expressed their opposition to the funding, Exxon responded by discontinuing this action.

We can observe in all four events the pressure being exerted by stakeholders that forced the disclosure and management of negative social and environmental impacts for which Exxon was held accountable. These events suggest that progress towards sustainability requires the activation of the transparency condition within the HAM process, as this leads to the disclosure of material information, which then results in negative environmental and social problems being resolved.

6. Discussion and Concluding Remarks

In this section, we discuss the explanatory capacity of the HAM model and how this model advances our understanding of materiality, all of which then informs future research possibilities.

6.1. Discussion

Table 8. Comparison of three states of materiality conception.

	Materiality in Financial Accounting	Current state of Materiality in Sustainability Accounting	Heterogeneity–Accountability Materiality Model (New Model Developed in This Study)
1. Applying materiality	Binary; Magnitude of the issue.	Continuum; Quantification; Significance to stakeholders and firm indexed by their actions.	Current sustainability accounting method is synthesised into the new construct of materiality.
2. Heterogeneity	Assumed Homogeneity.	Report to all stakeholders as a homogeneous whole.	Stakeholder heterogeneity paradigm is integrated into the HAM model and depicted as a materiality web.
3. Accountability–materiality connection	Not defined	Unresolved.	The accountability process [57] and the materiality logic [6] are integrated into the HAM model.
4. Matters to whom?	Matters to shareholders.	Materiality refers to an issue that matters to stakeholders, or the firm, or both.	An issue that matters to stakeholders must also matter to the firm through the principle–agent logic in the accountability principle.
5. Reporting and management	Materiality only applied to reporting practices.	Materiality relevant to either managerial or reporting practices.	Materiality’s management and reporting functions are connected, thus enabling the discharge of accountability to stakeholders.
6. Mismatch phenomenon	No explanation.	No explanation.	“Mismatch” occurs when the transparency condition is not met, and it is resolved by the activation of transparency mechanisms.

tracks the evolution of the materiality concept from financial accounting, to sustainability accounting, through to the HAM model. In particular, this table shows how the HAM model tackles the conceptual deficiencies discussed in Section 2.3.

First, we note that materiality in sustainability accounting that is modelled from the standards (e.g., [10,12,18]) and exemplified in the empirical studies (e.g., De Cristofaro & Raucci 2022; Garst, Maas & Juijs, 2022) [23,79] is operationalised by quantifying multiple levels of materiality pertaining to stakeholder and firm actions, all of which differs from the standard method of applying the materiality concept in financial accounting. This method, which is based on quantifying the effort, resources, and time invested to resolve sustainability issues, was integrated into the HAM framework and illustrated using selected events involving Exxon. The precision of the measurement of stakeholder action levels and, hence, materiality was likely enhanced by the extensive monitoring of stakeholder action and engagement with the firm, which was intended to provide a more sophisticated method of measurement compared to the illustrative examples provided in this study.

Second, stakeholder heterogeneity, which is not adequately captured in the current materiality literature, was depicted as a materiality web. Together with the HAM process (Figure 4), this was used to model materiality in a complex heterogeneous stakeholder environment where stakeholders hold diverse interests and take different action positions in response to the same event. This provides a solution to the problematic sustainability accounting practice that draws on the traditional standardised models (see, e.g., [12,13,16]), whereby stakeholders are simplistically identified as one undifferentiated homogenous group when defining and assessing materiality (Carst et al., 2022) [23]. In this way, we have overcome the challenge depicted by Carst et al. (2022, p. 80) [23] relating to replacing the “oversimplified picture of reality” with the complex heterogeneous stakeholder reality found in materiality practices.

Third, the accountability–materiality relationship was described by the HAM model as a five-step principal-agency process that shows how materiality practices enable the discharge of the firm’s accountability to its stakeholders, thereby supporting the primary stance of sustainability accounting as a means of fulfilling accountability to stakeholders [36,37,80]. The HAM model elucidates the accountability–materiality relationship, which has been rarely studied in the previous literature, but is required if we privilege the general view that accountability is enacted by accounting practices [81]. Furthermore, this new explanatory model supports the endeavour of implementing “accountability-based accounting” to improve both the very nature of accounting and also the disclosure of environmental and social information (Dillard & Vinnari, 2019) [82].

Fourth, the debate as to whether materiality is determined by what matters to the firm, or what matters to stakeholders, was resolved. This is because the HAM model clarified that an event that matters to stakeholders must also matter to the firm according to the principal–agent logic inherent in the basic accountability concept. Current sustainability accounting standards have been divergent in their ways of assessing materiality. Some focus on issues that matter to either the firm or to stakeholders [10,12], and some focus on issues that matter to both [18]. As was described earlier, some scholars have encapsulated this debate as a conflict between “double materiality” (the event matters to either) and “single materiality” (the event matters to both) (Jorgensen et al., 2022; Baumuller & Sopp, 2022) [24,30]. Such studies have highlighted problems relating to the coexistence of different understandings of materiality and how they may lead the users of sustainability reporting information astray. The need for clarity for scholars and practitioners alike regarding the “matters to whom” debate poses a significant challenge for institutions that shape the practice of sustainability reporting (Jorgensen et al., 2022) [24]. The view that “what matters to stakeholders must matter to the firm”, which is derived from the HAM model, encompasses a new way to analyse the materiality process from an accountability perspective, thereby informing a new direction to develop and implement materiality standards.

Fifth, the HAM model explicated the relevance of the materiality concept to both the firm’s management and reporting of sustainability issues, an aspect which connects the dual application of materiality logic to the discharge of accountability. Thus, our model reinforces the connection between the reporting and management of sustainability issues in materiality practices [10,83], and also provides a theoretical foundation for extending materiality research from its traditional reporting context to a managerial context (Whitehead 2017; Saenz 2019) [26,27].

Sixth, we have seen that there is a mismatch between the level of information currently provided in sustainability accounting reports and the information required by stakeholders to fully assess organisational performance. This lack of transparency is unresolved by reporting obligations that arise from materiality standards, as these obligations are unable to enforce the disclosure of environmental and social information that is material to stakeholders (Captuo et al. 2021; Slacik & Greiling 2020) [32,34]. With the HAM model, however, we are able to elucidate the nature of this “mismatch” as a failure to discharge accountability. Indeed, we argue that practical solutions lie in the establishment and activation of transparency mechanisms, as exemplified in the Exxon case, including stakeholder engagement and audits (Events 1 and 2) and law enforcement (Event 3), together with stakeholder intercommunications and the existence of whistleblowing stakeholders (Event 4).

6.2. Concluding Remarks and Future Research

In sum, this study addresses current theoretical deficiencies by integrating accountability and stakeholder heterogeneity into a new conceptualisation of materiality. In the Exxon case, we used this new model to explain materiality practices of sustainability issues in a heterogenous stakeholder environment by investigating a range of aspects that have not been sufficiently explained in the previous literature. These aspects include the firm’s varying responses to different ranges of stakeholder pressures; how reporting and managerial actions are connected; the consistency between reporting priority and managerial priority; and how the “mismatch” phenomenon occurs when the firm is unaccountable with respect to material issues, owing to a lack of transparency.

Future research on this theme includes three directions, all of which focus on the generalisation of the HAM model to a broader context that encompasses organisations of various sizes and types. First, the model can be applied to entities other than large multi-national corporations, such as small businesses or NGOs. By drawing on analysis of materiality practices across different entities, we may be able to construct a more comprehensive account of the successes and failures of discharging accountability with respect to environmental and social issues, all of which could enrich practical knowledge about how to improve sustainability outcomes. The second future research direction relates to the methodological aspect of materiality. Herein, we have provided a measurement method that is specific to the Exxon case by quantifying stakeholder actions and the firm’s response across five levels. Context-specific materiality measurement methods can likely be developed by investigation into quantifying the diverse materiality practices evident within different organisations or industries, which will enrich the array of calculative tools that are available to operationalise sustainability accounting tools. Third, future research could focus on the discovery and exploration of new transparency mechanisms that activate the discharge of accountability to prevent occurrences of the mismatch materiality phenomenon in different social settings. This direction would potentially inspire a deeper understanding of the very nature of accountability, and, in a practical sense, how accountability systems can be activated, depressed, and revived to report and manage sustainability issues.