Vulnerable Stakeholders’ Engagement: Advancing Stakeholder Theory with New Attribute and Salience Framework

Abstract

Stakeholder engagement is a central tenet for understanding and solving sustainability challenges. Given the existing stakeholder knowledge base and the fact that practitioners mostly focus on the engagement of high-power and salience stakeholders, the interests of low-power and vulnerable stakeholders are often manipulated. Therefore, this research is devoted to the engagement of low-power and vulnerable stakeholders. Grounded in the stakeholder theory and the results of two illustrations, we demonstrate how the physical proximity of vulnerable stakeholders influences salience patterns in a multi-stakeholder engagement context. The contribution of the study is the conceptualisation of proximity as a stakeholder attribute, in addition to power, legitimacy, and urgency, to help managers identify and appropriately engage with vulnerable stakeholders. Thus, we extend stakeholder typologies by incorporating proximity into the existing attribute model. The proposed model addresses the paradoxical nature of stakeholder salience and engagement theories and furthers the sustainability agenda.

1. Introduction

In the second half of the 20th century, concerns about sustainability and sustainable development (SD) became a global issue [1]. This policy discourse, following a massive stakeholder consultation process orchestrated under the aegis of the United Nations, developed 17 Sustainable Development Goals (SDGs), and committed to achieving these goals by 2030 [2,3]. Addressing sustainability at the city and community scales is at the heart of the SDG agenda [4]. Post-disaster reconstruction (PDR) is a unique context to further the SDG agenda as it aims at and provides opportunities to build back better [5]. The extensive body of knowledge emphasises that SDGs must be achieved without compromising the rights and autonomy of local communities [6,7]. Many researchers, therefore, insist that the only truly just or fair outcome is the one that is achieved through the engagement of local communities in the decision-making processes [8,9]. While researchers argue that the local communities affected by a development initiative, and who stand to benefit the most, should be recognised as key stakeholders, many recent reports have highlighted the treatment of the affected community as a low-power stakeholder [5,8,9,10]. Low-power stakeholders have also been referred to as ‘vulnerable’ stakeholders for their low capacity to influence decision-making patterns that often result in their interests’ manipulation [7]. We turn to stakeholder theory for a further integrative understanding of stakeholder identification and prioritization.

A common theory in the management literature used to identify stakeholders and prioritise their interests is stakeholder salience theory, which determines the existence and relative importance of stakeholders and their claims from a managerial perspective [11,12]. A cornerstone of the stakeholder salience theory comes from Mitchell et al. (1997), who defines stakeholder salience based on three attributes: power, legitimacy, and urgency. Salience is the degree to which managers must pay attention to specific stakeholders [13]. According to Mitchell and colleagues (1997), the most salient stakeholders are those who possess all three attributes. Contemporary disaster management studies find the stakeholder salience model relevant for the identification of salient stakeholders [14,15,16,17,18]. Studies have used the stakeholder salience model to explore Christchurch’s post-earthquake reconstruction in New Zealand [17], reconstruction after the Kashmir earthquake in Pakistan [14], reconstruction after a Tsunami in Sri Lanka [14,18], and flood recovery projects in Australia [16]. However, the salience framework does not support the prioritisation of vulnerable stakeholders such as disenfranchised local communities who, despite heavily depending upon the outcome of a development initiative, possess no or low levels of power and/or other attributes. While many recent empirical studies have reported compromised outcomes of PDR projects in developing countries due to the marginalisation of low-power stakeholders such as local communities, the inability to prioritise vulnerable stakeholders is a critical limitation of the stakeholder salience model [18,19,20,21]. Vulnerable stakeholders may have a valid claim and stand to benefit the most; however, they do not have the power to influence decision-making processes and other stakeholders. For instance, in a PDR context, the community affected by a natural disaster is a vulnerable stakeholder [20,22,23]. While a PDR project is initiated to build more resilient and sustainable housing facilities for the affected community, they often do not possess the power to influence other stakeholders or the related decision-making process [8,18,21]. Therefore, we argue that the stakeholder salience model does not appropriately acknowledge low-power stakeholders.

In this study, we have argued for ‘proximity’ as a fourth attribute for appropriately characterising vulnerable stakeholders who, despite their powerlessness, have a valid claim for recognition and inclusion in the implementation of a PDR project. This paper distils the findings from two case study projects to answer a guiding question: Could ‘proximity’ serve as fourth attribute for engagement of vulnerable stakeholders? We undertook a qualitative study of the stakeholders of two PDR housing projects in Sri Lanka and Pakistan in the aftermath of the 2004 tsunami and 2005 earthquake, respectively. Our data revealed a perceived under-representation of stakeholders’ attributes in the salience model. While the conceptualisation of new attributes—namely, proximity [24], social proximity [25], organisation [26], and status [27]—has been contested on several grounds in recent research [28,29,30,31], our findings support the proposition of ‘proximity’ as a fourth attribute on the following grounds: (1) the proximity attribute exists independently in the same way as the power, legitimacy, and urgency attributes exist; (2) proximity is distinct from the conceptual bounds of power, legitimacy, and urgency; (3) proximity has a similar effect on stakeholder salience as in the power, legitimacy, and urgency attributes; and (4) proximity can co-exist with power, legitimacy, and urgency without affecting their possession.

The unique contribution of this research is an extended stakeholder salience model and typology that includes proximity as an independent stakeholder attribute. In this way, we introduce the concept of identifying and prioritising engagement with vulnerable stakeholders toward the achievement of project objectives. We next discuss more nuanced concepts of the stakeholder salience theory followed by a brief summary of the case studies and methods used to analyse the primary qualitative data. The subsequent section discusses the rationale for the inclusion of proximity as a new stakeholder attribute and presents a new stakeholder typology. Finally, the contributions of the paper are summarised in the conclusion section.

2. The Stakeholder Salience Theory: Issues, Development, and Arguments for Proximity Attribute

The concept of a stakeholder has been embedded in the management literature since Freeman [32] published his landmark book, Strategic Management: A Stakeholder Approach [13,33,34]. Freeman [28] provided a schema to identify individuals and groups with an interest in an organisation and described them as stakeholders. The definition of a stakeholder suggested by Freeman is “any group or individual who can affect or is affected by the achievement of the organisation’s objective” [32] (p. 46). The stakeholder theory asserts that managers of organisations and projects should consider the interests of individuals or groups affected by the organisation [35,36]. Indeed, it is devised to answer two important questions: Who or what really counts, and to whom (or what) do managers pay attention? [13,32]. In this section, we describe the traditional and extant stakeholder salience position. We also present several critiques of this position, including those critiques advocating for additional attributes and moderating factors to be considered when gauging stakeholder salience. Finally, we conclude this section with our arguments for the inclusion of ‘proximity’ as a fourth stakeholder attribute.

2.1. Stakeholder Salience Theory: Strengths and Issues

Based on Freeman’s [32] seminal work, Mitchell et al.’s [13] stakeholder attribute and salience framework is considered the most substantial contribution to the academic understanding of stakeholder types and salience [9,29,37]. Mitchell and colleagues’ [13] stakeholder attribute and salience framework relies on three key assumptions: first, managers should pay attention to the stakeholders that are most salient to the achievement of their objectives; second, stakeholder salience is subjective and dependent upon managerial perception; and third, stakeholders and their salience can be determined through their possession of certain attributes [15,29]. They conceptualised three attributes—power, legitimacy, and urgency—to identify stakeholders and their salience. Power refers to “a relationship among social actors in which one social actor, A, can get another social actor, B, to do something that B would not have otherwise done” [13] (p. 865). Power reflects control over the physical, material, or financial resources [25]. Legitimacy is a generalised perception or assumption that the behaviour and actions of any stakeholder are desirable or appropriate within a socially constructed and accepted system of values, beliefs, and definitions [13,38]. Stakeholder legitimacy arises from a legal title or right, contract, exchange, at-risk status, and moral right or interest in the benefits or harms generated by the managers’ action [39]. It refers to the perceived validity of a stakeholder’s interests in a project or organisation. Urgency is the degree to which a stakeholder calls for immediate attention [12,13]. It refers to the perceived need for the stakeholder’s claims or grievances to be addressed as a priority. The stakeholder salience theory acknowledges that each attribute is socially constructed and is a matter of multiple perceptions rather than objectives [13] (p. 868).

The salience of a stakeholder reflects the extent to which decision-makers perceive the stakeholder’s claims as influential, legitimate, and urgent [40,41]. The most important or ‘salient’ stakeholder possesses all three attributes. Salience is used as a guide to help managers identify and prioritise stakeholders and their interests [7,30,42]. By problematising different combinations of perceived attributes, stakeholders are often divided into eight types: dormant, discretionary, demanding, dominant, dangerous, dependent, and definitive stakeholders, as well as non-stakeholders [13] (p. 874). Importantly, the premise of the extant stakeholder salience framework implies that while high-salience stakeholders’ interests or grievances cannot be ignored by managers, low-salience stakeholders’ claims or grievances can often easily be disregarded.

While stakeholder theory is widely accepted in the management literature, some issues require further scholarly attention [11,34,43]. The question ‘is stakeholder theory really a theory or is it merely a perspective?’ has plagued stakeholder theory since its inception [11] (p. 218). Despite its widespread acceptability and recognition in both academia and common parlance, the term ‘stakeholder’ is still a contested concept because it is considered variously describable, internally complex, and open in character [33,44]. Criticising the vagueness of the ‘can affect’ or ‘is affected’ parts of the definition of a ‘stakeholder’, Phillips and Reichart [45] mentioned: “why should we espouse a theory of stakeholder management if all living entities in as much as they can affect the firm must fall under the obligatory umbrella of managerial consideration?” [45] (p. 190). Thus, many have tried to define stakeholders in a more focused and explicit way. Still, a recent review of the scholarly endeavour in stakeholder theory revealed that 593 different stakeholder definitions exist in the academic literature [30]. One summative approach suggests that a stakeholder is any individual or group with a stake or claim to managerial consideration; that is, a stakeholder is anyone with an interest, right, or desire to be recognised by an organisation or project’s managers [30,46].

Moreover, the stakeholder salience framework relies on a limited set of three attributes, and stakeholders’ possession of these attributes is acknowledged from the managerial perspective. This is challenging because there is a potential for discrepancy between managers’ perception of the rights to recognition of individual stakeholders and stakeholders’ own perceived understanding of their rights and desire for their claims to be recognised by managers. Indeed, the existing salience framework relies on a set of subjective criteria for determining which stakeholders are important for managerial decision making. Thus, the existing perception of attributes is problematic. There could be stakeholders or groups who can have a significant influence on business, but that influence may not fall into the power, legitimacy, or urgency attributes [42]. It may be that those stakeholders that are, on the surface, deemed less important can significantly influence the activity outcome. In other words, managers can underestimate stakeholder attributes and the act of valuing some stakeholders as more important than others can have serious consequences [11,42]. While theorists argue that stakeholder relationships should be fair and reciprocal, focusing only on the managerial perspective is a major limitation and places the stakeholder salience model’s validity and practical relevance at risk [47].

The remedy is to either rely on the perceptions of a wider range of stakeholders (not only managers and high-salience stakeholders) or adjust the attributes to formulate more objective criteria of stakeholder salience. In this paper, we extend stakeholder theory by undertaking both solutions. That is, we explore stakeholder attributes from a 360° perspective by interviewing a range of stakeholders. Following this, after reviewing the attempt at the development of the salience model over time, we present our argument for the inclusion of proximity as an additional attribute.

2.2. Development of the Salience Model over Time

While Mitchell et al.’s [13] stakeholder attribute and salience framework is widely acknowledged and considered the most influential theory of stakeholder identification and prioritisation, the attempts to develop it further are limited [29,30,37]. Only few studies have attempted to extend the stakeholder salience model through the identification of a new stakeholder attribute or introducing a new typology.

Jonker and Foster [48] introduced a new stakeholder model using the power, criticality, and rationality attributes to analyse stakeholders’ involvement in strategic management. Their two-dimensional model explained the stake (what are the key issues?), parties (who is involved?), process (the process of managing the relationship), and connections (the form of the stakeholders’ and organisation’s relationship) as the four components of relationship, rationality, critically, and power as elements of stakeholder influence [48] (p. 194). However, this model reflects the analysts’ bias and does not incorporate the real perception of the stakeholders [49,50]. In addition, Haigh and Griffiths [51] developed a new stakeholder typology through incorporating the proximity attribute (based on Driscoll and Starik’s [24] conceptualisation of proximity). They grouped stakeholders into four categories: latent, expectant, definitive, and primary stakeholders [51] (p. 353). Their typology categorises stakeholders possessing different sets of attributes in one group, thereby contradicting the basic concepts of the stakeholder typology. Both models did not receive much attention with respect to research due to a lack of rigour, insufficient empirical evidence, and disregarding well-defined core concepts of stakeholder salience [29,31,52,53].

Focusing on the socially constructed nature of stakeholder salience, Ali [26] proposed ‘organisation’ as an attribute to determine stakeholder salience. The organisation is defined as the “body of persons organised for some purpose, as a club, union or society” [26] (p. 159). He argued that the concept of the organisation is integral to stakeholder agency, and it increases the ties between managers and stakeholders. Stakeholders who can mobilise themselves, initiate collective actions, create coalitions with other stakeholders, and improve their position in the social network can access more resources and power than others [26]. The organisation attribute also reflects upon collectiveness, as Ali [26] argued that stakeholders’ claims receive more attention from a manager if they develop a collective voice. Ali [26] concluded that urgency is irrelevant to determining stakeholderness. He replaced the ‘urgency’ attribute with the organisation and claimed that power, legitimacy, and organisation could exist together and/or independently. He also presented a revised stakeholder model, incorporating organisation instead of the urgency attribute [26] (p. 163).

Lähdesmäki, Siltaoja, and Spence [25] find a positive impact of the proximity and relationship attributes on stakeholder salience in a small business context. Instead of considering proximity as an independent attribute, they advocated for “social proximity” as a moderator that influences stakeholder salience. Social proximity is defined as the owner-managers’ involvement with stakeholders because of their sense of personal obligation, familiarity, and emotional closeness [25]. While social proximity between the business and stakeholder has a significant impact on the manager’s perception of the stakeholders’ salience, the authors also highlighted that social proximity could cause serious issues for the managers in the case of growing societal expectations and a higher sense of moral obligations [25].

Reflecting upon the lack of consideration of stakeholders’ behaviour and interests, in stakeholder theory, Perrault [27] introduced ‘status’ as a superior attribute for stakeholder identification and prioritisation. Based on Washington and Zajac’s [54] definition, Perrault [27] described status as “socially constructed, intersubjective agreed-upon and accepted ordering or ranking of individuals, groups, organisation, or activities in a social system” [27] (p. 26). The underlying assumptions for the conceptualisation of status as an attribute are: (i) stakeholders are real people with names and faces, (ii) every stakeholder is unique within the broad categories of stakeholders, and (iii) stakeholders have an inherent dual identity, i.e., social and economic. The status enables managers to account for stakeholders’ economic role towards the firm and their social identity simultaneously. Status can provide an individual-level ranking of all stakeholders; therefore, it is the considered superior attribute for stakeholder identification and prioritisation [27].

Management researchers, therefore, use Mitchell et al.’s (1997) stakeholder attribute and salience model for stakeholder identification and prioritisation and contest claims of new attributes on several grounds [28]. Since claims of organisation, social proximity, and status as attributes are relatively new, they are therefore subject to rigorous academic debate and empirical examination [29]. The main argument against these conceptualisations is that within a specific stakeholder group that possesses the same set of attributes, e.g., affecting the community in a PDR context, each individual may have a different status and organisation. The uniqueness of stakeholders’ status and organisation can affect an individual stakeholder’s salience but cannot affect the salience of the whole group. Thus, the conceptualisation of organization and status as attributes contradicts the widely accepted concept of the stakeholder group categorisation based on the attributes they possess.

Driscoll and Starik [24] proposed that in addition to power, legitimacy, and urgency, proximity should be included as an additional stakeholder criterion. They argued that stakeholders who are geographically close to a project or organisation are likely to be more influential than distant stakeholders. They defined proximity as “the state, quality, or fact of being near or next in space, time or order” [24] (p. 63). The rationale is that the characteristic of proximity—of being physically close to a particular project or organisation—gives stakeholders the ability to influence the project and those managing it. The goal of introducing the proximity attribute was to allow the natural environment to be a primary stakeholder [24]. In addition to physical closeness, organisations and stakeholders are also considered proximate if they share the same or similar ideas, actions, and occupy the same field [24]. In this way, proximity can be categorised into two groups: geographical and organised [55,56]. While geographical proximity is about the physical distance of stakeholders from an organisation, i.e., near and far, organised proximity is the ability to interact through rules, values, codes, beliefs, and shared representations [55]. Organised proximity is possessed when stakeholders and managers share the same system of representation, codes, norms, and values, which makes it easier for them to interact. Nevertheless, geographical closeness has been considered the most important state of proximity [56].

2.3. Physical Closeness as Stakeholder Attribute—Propositions for Validation

Some have considered that proximity is not itself an independent attribute; instead, it may be viewed as a moderator of stakeholder salience [31,37]. Scholars who espouse this stance argue that the impact of proximity on stakeholder salience is subsumed within the attribute of power [31], and that proximity alone does not afford one the status of a stakeholder [30]. However, there are clear arguments in favour of proximity as an attribute distinctive from the existing attributes of power, legitimacy, and urgency. First, the concept of proximity is distinct from the conceptual bounds of the power attribute because physical closeness does not necessarily mean a stakeholder can influence managerial decision making in their favour. In fact, especially in the context of sustainable development, many vulnerable communities may be physically close to a project and even benefit from it but lack the power to affect the outcome. Second, it is possible for stakeholders to be co-located within the physical operations of a project or organisation without managers perceiving the stakeholders’ claims and interests as valid or legitimate. Third, there is no inherent consideration of the importance of implementation timeframes (i.e., urgency) when one considers the physical closeness of a stakeholder. Yet, stakeholder proximity is an increasingly important consideration for managers because of growing societal expectations and a higher sense of moral obligations to engage with stakeholders within a proximal distance of managerial decision making [25]. Therefore, we argue that proximity could serve as a more objective attribute of stakeholder salience. We propose that proximity can compensate for the exclusion of stakeholders and help them to obtain an appropriate level of salience that they may lose due to not having the power, legitimacy, or urgency attributes. To support our argument, we present the following four propositions subject to validation through data analysis.

Proposition 1.

The proximity attribute exists independently in the same sense as the power, legitimacy, and urgency attributes.

Proposition 2

. Proximity is distinct from the conceptual bounds of the power, legitimacy, and urgency attributes.

Proposition 3

. Proximity has a similar effect on stakeholder salience as those of the power, legitimacy, and urgency attributes.

Proposition 4

. Proximity can co-exist with the power, legitimacy, and urgency attributes without affecting their possession.

3. Case Studies and Methods Used

3.1. Case Studies

We selected two post-disaster reconstruction (PDR) projects, one each from Pakistan and Sri Lanka. In these case studies, access to housing was provided in the wake of two natural disasters: the 2004 tsunami in the Indian Ocean that affected Sri Lanka, and the 2005 earthquake in Pakistan. Housing projects are the most common and highest-funded projects in PDR efforts. Chief amongst the aims of these efforts is to transition vulnerable communities from a state of living in a shelter to a state of access to safe housing and to protect them from future disasters. While shelters are spontaneous and temporary, housing is a long-term process and helps to re-establish a community after a disaster. The newly constructed housing tends to be safe, of high quality, and less vulnerable than in the pre-disaster situation. The expectations of involving the beneficiary communities in the reconstruction effort are also significantly higher because contemporary PDR housing projects also now incorporate sustainability goals [41]. Therefore, PDR projects are appropriate for the assessment of perceived stakeholder attributes due to the following reasons. First, the beneficiaries of housing in PDR contexts are largely among the poorest and most vulnerable communities, which provides appropriate context and opportunity to investigate the recognition-related and inclusive aspects of justice. Second, PDR projects involve the use of large amounts of resources and the management of a complex set of partnerships with multiple stakeholders often with contrasting perceived levels of salience. Finally, the level of stakeholder involvement at each stage of these reconstruction projects has a direct influence on the project’s success and sustainability [16,18].

3.1.1. Case Study 1: Sri Lanka

In Sri Lanka, the project titled ‘Relocation of Tsunami affected people’ was initiated to provide housing and facilities to the people of Galle who were affected by the Indian Ocean tsunami in 2004. The tsunami occurred on 26 December 2004, and caused the death of 226,226 (including 49,648 missing) people [57]. In the aftermath of the disaster, the government of Sri Lanka initiated a permanent-housing project to relocate tsunami-affected people from coastal areas of Galle city to safer areas. International donors provided more than USD 11 billion for reconstruction activities in the country [58]. The government allocated the land for housing construction. Donors constructed the houses through local contractors. After the completion of the project, the government of Sri Lanka allotted houses to affected communities. In the first quarter of 2017, four major project sites were visited to collect the data for this study. The selected project locations are in the Akmeemana (Katuguda, Nagudua, and Walahandua) and Hikkadua (Dadella) secretary divisions.

3.1.2. Case Study 2: Pakistan

In Pakistan, the ‘New Balakot city development (NBCD) project’ was initiated in the aftermath of the 2005 earthquake, which affected about a 30,000 km² area, and resulted in 86,000 deaths [59]. Post-disaster surveys revealed that Balakot city was one of the most affected cities and was therefore unsuitable for future construction because it is situated on the conjunction of three major fault lines [60]. After the emergency rescue operations, the government of Pakistan established the Earthquake Reconstruction and Rehabilitation Authority (ERRA) for post-disaster reconstruction [61]. In order to relocate 5000 affected families from the earthquake-prone zone to a safer place, the ERRA initiated the NBCD project, about 20 kilometres away from Balakot city. International donors committed to funding for the project [62]. The project was initiated in 2007 with an estimated cost of USD 200 million, and a target completion date of July 2010 [62,63]. However, at the time of data collection in December 2016, the project was still in the implementation phase.

3.2. Data Collection and Analysis

This study used a qualitative research method, face-to-face semi-structured interviews (n = 66—refer

Table 1. Number and roles of stakeholders interviewed.

Stakeholders	Role in the Project	Sri Lanka	Pakistan	Total
Affected community	Beneficiaries	14	34	48
Government officials	Decision makers	1	3	4
Contractors	Project executors	1	1	2
Political Leader	Elected public representatives	1	2	3
Community leader	Community spokesperson	1	2	3
Researchers	Observer	2	-	2
Social worker	Social activists	-	3	3
Journalist	Observer/activists	-	1	1
		20	46	66

), to explore a PDR project in Sri Lanka and Pakistan. Diverse perspectives were elicited through interviewing numerous highly knowledgeable informants with diverse backgrounds and perspectives [64]. A purposive sampling technique was used to recruit respondents with specific roles (such as government officials and contractors) and a snowball-sampling technique was used to engage respondents from the affected community. In this research study, a heterogeneous group of respondents needed to be selected. The inclusion and exclusion criteria determined who could be included or excluded from the study. The inclusion criteria restricted the selection of respondents to only adults (25+ years of age) who participated/performed a formal/informal role in project implementation and/or had a potential benefit/concern/interest in the case study project. Demographic data of the participants (sex, race, and ethnicity) were recorded to ensure diversity in the respondents, yet were not considered in the inclusion criteria. Local populations who were not affected by the disasters, and who did not have any direct or indirect benefit/concern with respect to the case study project, were excluded from the study. Clear, precise, and objective inclusion/exclusion criteria helped to select well-informed and experienced respondents who willingly took part in the research study. The recruitment of interviewees across diverse stakeholder groups also increased the reliability and validity of the research data [18,65].

Each participant was asked a set of open-ended questions. After the introductory briefing, the interview focused on the identification of the project stakeholders; their roles and interests; the pattern of their participation in project activities, particularly, in the decision-making process; and the achievement or exploitation (or lack thereof) of their potential interests from the project. The interviews themselves included several other planned and emergent lines of questioning that uncovered information about the stakeholders involved in each project, their roles and interests, and any observed patterns of involvement within decision-making processes. We used the underlying concepts and definitions within the stakeholder attribute and salience framework to assign attributes to each stakeholder. By assigning attributes and comparing their perceived levels of importance, we were able to infer which stakeholders were prioritised—that is, perceived as most salient—in the decision-making processes at different stages of both projects. Further probing questions were used to obtain a deeper insight into the understanding and experience of stakeholders regarding the project. All the interviews were conducted in the local languages. One researcher was native to Pakistan and could easily understand and speak the local languages. In total, the interviews lasted almost 2000 min. Subsequently, the interview transcripts were translated from the local languages to English.

The qualitative data analysis software package Nvivo was used to support the thematic analysis of the interviews. The thematic analysis followed a well-established procedure for deductive theory building from qualitative data as outlined by Boyatzis [66], Boje [67], and Braun and Clarke [68]. A theme is something that reflects a patterned response or meaning relevant to the key research concepts; in this study, these key concepts include patterns of the decision-making process, the attributes of the involved stakeholders and their interests, participation, and the achievement or exploitation of stakeholders’ expectations [68]. Each transcript was assigned a unique code. The transcripts were read and re-read several times and the sections of text—such as direct statements, quotes, stories, and references—that referred to a key idea of research were highlighted and encoded as a separate node. Multiple nodes pointing towards similar ideas were grouped as a theme. The same process was repeated for each interview. The guiding concepts grounded in relevant theoretical frameworks that were used to develop themes are summarised in

Table 2. Description of themes developed in the data analysis process.

Attribute/Theme	Description/Guiding Idea of Theme
Power	Stakeholders who made project-related decisions in any one or more processes of the project, managed or controlled project resources, or influenced other stakeholders to perform a specific role.
Legitimacy	Stakeholders with a defined and accepted role in the project, and/or whose claims and actions were accepted by other stakeholders.
Urgency	Stakeholders who required the immediate attention of other stakeholders and were interested in the earliest completion of the project/activity.
Proximity	Stakeholders who are physically close, living in same neighbourhood, and/or sharing same values, codes, beliefs, rules, and system of representation as decision-makers.

.

The analysis process resulted in more than 700 quotes that directly or indirectly related to our research objective.

Table 3. Illustrative quotes.

Quote from Interviews	Interviewee/Case Study	Perceived Attribute of Stakeholder
The government is the decision-maker in this project. They (government) decided to relocate us (affected community), they selected the land, established the ERRA, and provided the funds for the construction of houses.	Affected community member/Pakistan	Power & Legitimacy of Government of Pakistan
The donors were responsible for project implementation. The central (Sri Lankan) and local (Galle province) government provided all the necessary support to these donors. Local politicians and community leaders helped them to collect basic data from the community and local contractors were hired to implement the project. The affected community was happy that the donors were building houses for them.	Government official/Sri Lanka	Legitimacy of donors
We (respondent and his family) are living in this temporary shelter provided by the Saudi Arabian government. I lost four family members due to the earthquake and my children are still afraid of it. We cannot reconstruct our damaged house because the government has declared this area as unsuitable for construction. Our new houses are still under construction and we are not sure when it will be completed. We are left with no option but to live in this shelter.	Affected community member/Pakistan	Urgency of affected community
Few community members who belong to the same ethnic group as government official (Sinhala) took extra advantages. They got houses at a better location and got deeds (ownership) of the houses at the time of allotment, unlike other community members. We still do not have ownership of our house.	Affected community member/Sri Lanka	Proximity (organized)
The residents of Bakriyal (new project site) are getting benefits even though the project is not yet completed. Prices of their land have gone up because of improved infrastructure and increased economic activities in the area.	Affected community member/Pakistan	Proximity (physical)

presents illustrative quotes from the interviews to provide transparency to our analysis.

The possession of attributes influences the salience of stakeholders as per the existing stakeholder attribute and salience model. Therefore, the perceived attributes of each stakeholder from the individual interviews were aggregated into the perceived attributes of all the respondents of both housing projects.

Given that attribute possession is dynamic in nature, it was expected that stakeholders would possess different sets of attributes at specific stages of each project [13]. These phases were identified because each phase is distinct, represents a major milestone, and involves extensive decision making that directly influences the project outcome. For Sri Lanka, these phases were: (1) the selection and purchase of land, (2) the selection of an approach to implementation, (3) the establishment of project parameters, (4) the selection of a project implementation team, (5) the physical implementation of project, (6) the selection of community members for the allotment of houses, and (7) the allotment and acquisition of houses. For Pakistan, these phases were the same; however, we restricted our application of the salience framework to the first five phases because the project was undergoing implementation at the time of data collection.

4. Stakeholders’ Attributes and Salience in Case Study Projects

The summary of our data analysis in relation to the identification of stakeholder attributes and their salience is presented in

Table 4. Summary of perceived attributes and salience of stakeholders.

	Phase 1 (Land Selection)	Phase 2 (Implementation Approach)	Phase 3 (Parameters Selection)	Phase 4 (Implementation Team Selection)	Phase 5 (Project Implementation)	Phase 6 (Community Selection)	Phase 7 (House Allotment)
Sri Lankan project stakeholders
National Government	Power Legitimacy Urgency	Power Legitimacy Urgency	Legitimacy Urgency	Legitimacy Urgency	Legitimacy Urgency	Power Legitimacy Urgency	Power Legitimacy Urgency
Community	Urgency	Urgency	Urgency	Urgency	Urgency	Urgency	Urgency
Contractors				Legitimacy	Legitimacy
Community Leaders	Urgency	Urgency	Urgency	Urgency	Urgency	Legitimacy Urgency	Urgency
Local and provincial government	Legitimacy					Power Legitimacy	Power Legitimacy
Local Politicians	Urgency	Urgency	Urgency	Urgency	Urgency	Legitimacy Urgency	Urgency
Donors		Legitimacy	Power Legitimacy Urgency	Power Legitimacy Urgency	Power Legitimacy Urgency
Pakistani project stakeholders
National Government	Power Legitimacy Urgency	Power Legitimacy Urgency	Legitimacy Urgency	Legitimacy Urgency	Legitimacy
Community	Urgency	Urgency	Urgency	Urgency	Urgency
Contractors					Legitimacy
Community Leaders	Urgency	Urgency	Urgency	Urgency	Urgency
Local and provincial government	Legitimacy Urgency	Legitimacy Urgency	Legitimacy Urgency	Legitimacy Urgency	Legitimacy Urgency
Local Politicians	Urgency	Urgency	Urgency	Urgency	Urgency
Donors		Power Legitimacy Urgency	Legitimacy Urgency	Legitimacy	Legitimacy
Landowners	Legitimacy
ERRA	Power Legitimacy	Power Legitimacy	Power Legitimacy	Power Legitimacy	Power Legitimacy

. Acknowledging the dynamic nature of attribute possession, we analysed stakeholder salience at various distinct stages of project implementation. The data confirm that the changing roles in different phases of both projects led to a corresponding change in possessed attributes and salience. For instance, the government was the most salient stakeholder in the Sri Lankan case study with respect to land selection and designing the approach to implementation processes due to their exclusive decision-making role. However, the Sri Lankan government lost its power in the subsequent phases, with donors taking the lead and the perceived power in the latter phases of the project. The direct involvement of government officials in the phases involving the selection of community members and the allotment of houses helped to enhance their perceived power and legitimacy among the other stakeholders in phases 6 and 7. Likewise, in Pakistan, the national government was perceived by other stakeholders as the definitive stakeholder, possessing all three attributes in the selection of land and implementation approach phases, but seemingly losing perceived power later, due to the transfer of project-level decision-making power to the ERRA. Indeed, Table 4 confirms the dynamic nature of the possession of stakeholder attributes.

A key finding from our research is that none of the respondents in either case study mentioned that the affected community was involved in decision making at any phase from the project’s planning to its completion. The affected community was perceived as being in possession of neither the power nor legitimacy to influence the decision-making process or control resources. The role of the affected communities in both case projects was generally perceived as passive, and they were not consulted or considered. This suggests that the communities were not perceived as possessing power and legitimacy. Given that the affected community lost their loved ones and possessions to the tsunami and at the time of project planning and implementation were living in temporary shelters, they are perceived to be in possession of the urgency attribute. The affected community was in dire need of permanent housing, which as 54 out of the 66 respondents in both case studies suggested reflects the urgency of their claims. According to the existing stakeholder salience model, this would suggest that affected communities are low-salience stakeholders. This is a worrying observation because there is strong evidence that all the major donor agencies involved in PDR housing projects in developing countries, such as the United Nations and The World Bank, as well as researchers and experts on the ground, consider the affected community to be a significant stakeholder in these projects [8,69,70,71]. Indeed, for years, many have advocated for the greater involvement of affected communities in PDR housing projects, arguing that it is unfair that they are excluded from decision making [18,20,71]. Yet, if those tasked with implementing these projects in Sri Lanka and Pakistan relied on the contemporary stakeholder salience framework, the projects’ managers might conclude that the affected communities are low-salience stakeholders and can therefore be disregarded or ignored in the project’s implementation [13,72].

5. The Emergence of Proximity as Stakeholder Attribute

According to the existing stakeholder salience framework, a stakeholder who does not possess power, legitimacy, or urgency is considered a non-stakeholder [13]. However, our findings suggest that stakeholders’ identification using the existing attribute model is not comprehensive. Some individuals and groups did not possess power, legitimacy, and urgency attributes, yet fulfilled the criteria of being a stakeholder, whether affected or affected by the outcome of the activity [32]. This situation reflects that they simply possess another under-explored attribute instead. We found evidence that one crucial additional attribute is ‘proximity’ as defined and explained by Driscoll and Starik [24]. Specifically, 41 of our 66 respondents described characteristics of some stakeholders using words and phrases that suggested their physical and/or organized closeness to the project as a crucial determining factor for their stakeholderness and salience. We evaluate the emergence of proximity as a fourth attribute using four propositions:

Proposition 1.

The proximity attribute exists independently in the same sense as the power, legitimacy, and urgency attributes.

While stakeholders such as the local private and corporate sectors and business owners were not perceived by the respondents to possess the power, legitimacy, or urgency attributes regarding the PDR projects, they still gained direct benefits in terms of higher earnings due to substantial construction activities in the neighbourhood. The government official who was involved in the Sri Lankan project mentioned this aspect clearly (Quote 1;

Table 5. Selected illustrative quotes from interviews.

Quote No.	Project & Interviewee Role	Quotation (Translated from Native Language to English)	Identified Stakeholder
1	Pakistan/ Government Official Pakistan	The construction material was in high demand due to extensive construction activities after the tsunami. Therefore, local businesses flourished, and merchants earned high profits.	Local businesses
2	Pakistan/Journalist	Bakriyal (new project site) is far from the main road; however, after construction of wide access roads under the NBCD project, and the anticipated construction of new schools, a shopping area, and a hospital, land prices in this area have risen substantially.	Local residents
3	Pakistan/Contractor	We are facing a shortage of local labour because the local population is short in number and some of them already work in major cities. The project site is also far from the main road. So, we are paying higher wage rates to attract and retain labourers.	Local labourers
4	Sri Lanka/Affected community	A few community members received better houses than us because of their relationship with government officials who were involved in data collection and allotment of houses. They also received ownership deeds of their houses but we did not.	Affected community

).

The respondents also reported that the local population living adjacent to the project implementation sites benefitted in terms of higher market rates for their properties, improved infrastructure, and better access to a range of economic benefits (Quote 2). In addition, substantial construction and infrastructure development activities also provided significant employment opportunities to labourers located locally. A respondent from the project’s contractor group reported a lack of labour in the construction work in the Pakistani project but highlighted their benefits to locals (Quote 3). Similarly, local NGOs, the media, researchers, and social workers do not possess the power, legitimacy, or urgency attributes and have no influence on project. However, due to their physical closeness to the project implementation area, they were affected by the outcome of the project. However, Mitchell et al.’s [13] framework is unable to assign them any attribute to shift them from a “non-stakeholder” status to an appropriate stakeholder category. Our data highlight that the common aspect among local businesses, land owners, social workers, NGOs, the media, and researchers is their physical proximity to the project. This physical proximity helps them to qualify for an appropriate category of ‘stakeholder’. Hence, this validates proposition 1, namely, that the proximity attribute exists independently in the same way as the other attributes exist.

Proposition 2.

Proximity is distinct from conceptual bounds of power, legitimacy, and urgency attributes.

The characteristic of the proximity attribute—physical distance, shared values, and a shared system—is beyond the conceptual bounds of the existing attributes. The conceptualisation of the power, legitimacy, and urgency attributes’ identification does not include the core idea of the proximity attribute. To reiterate, power is about influencing others, legitimacy is about accepting roles, and urgency is about calling for urgent assistance [13]. However, as highlighted in the preceding paragraph, stakeholders who do not possess these attributes are still affected or affected by the project, which confirms their stakeholderness. Thus, there is another attribute that is distinct from the existing conceptualisation. The interviews of the stakeholders provide greater insight into the distinct characterisation of the proximity attribute. The data reflect that local landowners and businesses did not possess power, legitimacy, or urgency, but their physical closeness made them stakeholders of the project (Quote 1). Likewise, some stakeholders were also perceived to be in possession of organized proximity. The organized proximity exists when sharing the same values, codes, beliefs, rules, and system as the decision-makers. The stakeholders received additional benefits due to their possession of organized proximity (Quote 4). However, this aspect is also not theorised in the existing attributes. Therefore, both of proximity’s aspects, physical and organized closeness, are distinct from conceptual bounds of the existing attributes. Hence, the data analysis also validates proposition 2.

Proposition 3.

Proximity has a similar effect on stakeholder salience as those of the power, legitimacy, and urgency attributes.

The data reflect that the physical involvement of a stakeholder in any stage of the project enabled that stakeholder to attain a higher level of salience. For example, the physical involvement of contractors enabled them to possess the highest level of salience at the stage of the physical execution of the project. Likewise, in Pakistan, the ERRA’s physical involvement in the project allowed it to possess the highest level of salience. On the other hand, a stakeholder’s physical remoteness seemed to lead to a decrease in perceived legitimacy and power, and thus a decline in salience. For instance, the declining perceived salience of the government of Sri Lanka at the later stages of the project seemed to occur because of its remoteness from the project site (Table 4). A similar trend was observed in the donors’ attributes and salience: when they were physically present at the start of the project, they were perceived as salient stakeholders, but the donors’ salience declined as they became more physically remote during the latter phases of the project. Hence, the possession or loss of the proximity attribute has a corresponding effect on salience as those of the power, legitimacy, and urgency attributes, which validates proposition 3.

Proposition 4.

Proximity can co-exist with the power, legitimacy, and urgency attributes without affecting their possession.

The data analysis also highlighted that stakeholders possessed the proximity attribute in addition to any one or combination of the power, legitimacy, and urgency attributes, e.g., ERRA, the local and provincial government, and the contractors in the Pakistani case study, and the National government, the local and provincial government, and the local donors in the Sri Lankan case study. The proximity attribute did not cause them to lose any attribute they possessed; instead, they possessed proximity in combination with other attributes, which validates proposition 4.

Summary of Findings

This discourse suggests that there might be a missing attribute, which is not truly reflected within Mitchell et al.’s [13] existing stakeholder attribute and salience model. Given what our respondents explained in their interviews, we surmise that this new attribute must have certain key characteristics. First, the missing attribute seems to compensate for the lack of perceived ‘fairness’ in the current approach, which excludes members of vulnerable affected communities. Second, the missing attribute can be possessed in conjunction with other existing attributes, with the result that it moves some beneficiaries and/or influencers from a characterisation of a ‘non-stakeholder’ to a ‘stakeholder’. Finally, and because of the previous two characteristics of this missing attribute, the result of possessing this attribute should be an enhancement of the salience of the disregarded stakeholders from low to moderate or even a high level. Both case studies provided evidence that physical closeness to the project helped stakeholders to gain higher salience and remoteness caused the loss of the possessed attributes. This supports the argument that ‘physical closeness to the project’ is a feature that not only allows non-stakeholders to become stakeholders, but it also (positively) affects their salience. This relationship—that is, the closer to the project, the higher the stakeholder’s perceived importance—suggests that the influence of this feature on salience is similar to the attributes of power, legitimacy, and urgency. The result, we surmise, would be a stakeholder salience framework that achieves both consequential fairness (by more accurately identifying salient stakeholders) and deontological fairness by including these stakeholders in decision-making processes towards the achievement of the project outcomes.

6. Proximity as an Ethical Attribute: A New Framework

Indeed, physical closeness has a positive effect on the perceived salience of stakeholders. Therefore, we argue for the inclusion of proximity in the existing stakeholder salience framework as an attribute that enhances the fairness with which vulnerable stakeholders are not only recognised but also included in decision-making processes. Although Driscoll and Starik [24] recommended the inclusion of proximity to advocate for the natural environment as a primordial stakeholder, our inquiry points to the same solution and endorses the increasingly compelling case for the inclusion of physical closeness as a fourth attribute for stakeholder identification and prioritization (see Khurram, Pestre, and Charreire Petit [29]; Joos [37]; and Miles [30] for examples). After all, it is the persistent and cumulative presentation of compelling arguments and evidence from different standpoints that drives theory development.

In the context of sustainable project management targets for adequate, safe, and affordable housing, especially after natural disasters, we outline three arguments for a new stakeholder identification and salience framework that includes proximity. First, proximity is a fundamental ethical consideration for project donors and managers because it can deliver deontological justice for those most affected and vulnerable locally. Deontological justice aims for inclusiveness with respect to identifying/recognising and including people in the processes of decision making. Deontological justice suggests that an ethical or fair decision is one that includes the most (affected) people in all decision-making processes. In addition, implicit within the recognition-related aspects of deontological justice is a focus on marginalised or under-recognised groups of people [73]. Indeed, in many circumstances, although they may be considered ‘urgent’ stakeholders, local beneficiary communities in emerging markets may not be considered to possess the attributes of ‘power’ or ‘legitimacy’. Within the existing stakeholder salience framework, this does not allow for such communities to be identified and framed as important or salient stakeholders. Proximity as a fourth attribute has the potential to rectify the exclusion of these vulnerable communities by upholding the recognition-related values of deontological justice. In sum, proximity as a fourth criterion for stakeholder salience necessitates the inclusion of local communities by virtue of the physical proximity of their lives and livelihoods to the project site.

Second, proximity could produce consequential justice for those most affected and vulnerable locally. Consequentialism requires that all decision-making processes lead to an outcome that is agreed upon and desired by all those affected [73]. That is, an ethical decision is one that produces the best end result for the most (affected) people. The inclusion of proximity as an ethical stakeholder attribute acknowledges that stakeholders such as local communities are most affected by global (or, at least, multilateral) decisions about the SDGs. Therefore, consequentialism and the use of proximity as a criterion for the identification and prioritisation of stakeholders focuses on the most ethical ends or outcome. In other words, if locally affected people do not agree on the benefits of the outcomes achieved, then the outcome was not achieved with a fair consideration of their needs and rights. Thus, proximity as the fourth attribute of salience ensures that vulnerable communities reap the benefits envisioned by the housing project from which they are intended to benefit.

Finally, the United Nations has acknowledged the need to accelerate the pace of change towards the achievement of the SDGs. A more robust framework, which includes all those affected and in need of recognition, is a good way to discriminate among a complex network of heterogeneous stakeholders and prioritise their involvement in project implementation. This has the potential to reduce the time and other resources needed for the implementation of the SDG targets and contribute to a more seamless integration between the macro- and micro-level contexts of project implementation. Thus, the addition of proximity as a fourth criterion of stakeholder salience enhances the efficiency with which those involved in project implementation can work towards the achievement of the best possible outcomes for those most affected. Accordingly, building on the existing stakeholder salience framework, we support calls for a new model that incorporates all four attributes—power, legitimacy, urgency, and proximity. In response to these calls, and based on the findings from our own research, we propose a new framework (Figure 1).

Haigh and Griffiths [51] also previously developed a new stakeholder typology that incorporates proximity as an attribute. They grouped stakeholders into four categories: latent, expectant, definitive, and primary stakeholders [51] (p. 353). However, this model has been criticised within the academe for the insufficiency of the empirical evidence provided and for its purported disregard for the well-defined core concepts of stakeholder salience. Therefore, the model we present in Figure 1 is derived from a more robust foundation in empirical research, where physical closeness emerged as a clear differentiating and determining factor in stakeholder salience. The addition of proximity to the salience model as the result of empirical enquiry extends the number of stakeholder types from 8 to 16. The new framework illustrated in Figure 1 introduces eight new stakeholder types: recipient, claimant, influential, collaborative, independent, expectant, fiduciary, and crucial stakeholders. These new attributes are described in

Table 6. Description of newly developed attributes.

#	New Type (Attributes)	Description & Examples
4	Recipient (Proximity)	Recipient stakeholders receive some benefit from the project because they reside in or close to the project implementation area. Although not the project’s target beneficiaries, they benefit indirectly from the project by virtue of their physical closeness to the project’s beneficial outcomes (e.g., local business).
7	Claimant (Legitimacy & Proximity)	Claimant stakeholders have a perceived legitimate role and claim to the project and reside in or close to the project area. They receive a direct benefit from the project outcomes (e.g., landowners).
9	Influential (Power & Proximity)	Influential stakeholders are powerful and either reside in or have some form of control over the project implementation area (e.g., local politicians).
10	Collaborative (Urgency & Proximity)	Collaborative stakeholders do not possess power and legitimacy, but because they were affected by the disaster due to their closeness to affected areas, they have a high interest in the urgent completion of the reconstruction project. Their collaboration contributes to the success of the project (e.g., Pakistani provincial government).
12	Independent (Power, Legitimacy & Proximity)	Independent stakeholders are in a position to implement the project without the help of other stakeholders because of their ability to influence others and the official recognition of their role in implementation. These are usually local stakeholders with a physical presence in the project area (e.g., ERRA implementation team in Pakistan).
13	Expectant (Legitimacy, Urgency & Proximity)	Expectant stakeholders are not considered powerful, but because they possess urgency, proximity, and legitimacy attributes they expect to gain a direct benefit from the project. Other stakeholders (though not necessarily project implementers) recognise them as a stakeholder, which legitimises their role (e.g., affected communities and their leaders).
14	Fiduciary (Power, Urgency & Proximity)	If collaborative stakeholders acquire power over the project implementation area, they become fiduciary stakeholders. Project managers recognise their responsibility to report directly to these stakeholders on project outcomes (e.g., donors). Vulnerable affected communities might aspire to this role by demanding community-driven approaches to project implementation.
15	Crucial (Power, Legitimacy, Urgency & Proximity)	Crucial stakeholders are the decision-makers, implementers, and beneficiaries of the project. Possession of proximity attribute helps them to gain direct benefit from the project. For vulnerable affected communities, the role of crucial stakeholder is even better than that of a fiduciary stakeholder.

.

The new typology equips managers to systematically and fairly identify and engage a more comprehensive set of stakeholders as a priority, even including those stakeholders perceived as powerless or vulnerable. We argue that the true power of this new typology is its potential to support and frame ethical considerations within the decision-making processes that underscore stakeholders’ perceived levels of salience. In particular, individuals and groups previously disregarded as possessing low- or medium-level salience within Mitchell et al.’s [13] model, due to the absence of attributes such as perceived power, may be appropriately categorised as important stakeholders in the new typology if they are within close physical proximity to an organisation or project’s operations. The new framework has the potential to guide managers and practitioners on ‘who are the right stakeholders to involve?’ In cases of potentially conflicting interests of stakeholders, instead of following short-term individualistic patterns, managers could now follow a more systematic approach to deliver both deontological and consequential justice to the ‘right’ stakeholders of their activities.

7. Conclusions

Stakeholder engagement has been acknowledged as a necessary activity for sustainable development. However, it is difficult to determine which stakeholders are most important, or salient, when the complex global sustainability agenda requires scaling from the macro- and meso-contexts to the micro-contexts of individual communities and projects. Moreover, it is arguably even more challenging to take care of the vulnerable stakeholders, specifically when their salience is not appropriately perceived. Thus, we drew insights from our study of two post-disaster housing projects, where closeness to the project emerged as a crucial factor with respect to the perceived importance of stakeholders, and thus the perceived importance of their involvement in project implementation. Indeed, vulnerable beneficiary communities were considered not only appropriate but also important stakeholders to involve because of the urgency of their need for housing and because of their actual proximity to the project and stakeholders.

Thus, we surmise that Mitchell et al.’s [13] salience model requires an extension—one that fulfils the recognition and inclusion of not only powerful but vulnerable stakeholders as well. By problematising stakeholders’ attributes as a criterion for their identification and prioritisation, we endorse ‘proximity’ as a fourth attribute—a criterion that ensures the greater engagement of the vulnerable stakeholders. The resulting new framework extends the existing stakeholder typology to include the recognition and engagement of vulnerable stakeholders. We anticipate that this new framework will help decision-makers and disaster management practitioners to identify and engage vulnerable stakeholders and to safeguard and fulfil their interests. Future scholarly endeavours may extend the new model to additional stakeholder-related problematics and contexts.