De-densifying the black hole – what if professional discretion strongly influences corporate social responsibility?

Corporate social responsibility (CSR) democratises economic activities and can potentially step in where legislation is insufficient (Scherer et al., 2016). It offers a mechanism for transnational regulation in a society that is more vulnerable to corporate harm because of its global connectedness (Donaghey and Reinecke, 2018; Girschik et al., 2022; Shaw, 2007). Multinational corporations use intercontinental structures that can loophole them out of regulation because they exist “with no loyalty to any particular place or people” (Shaw, 2007, p. 11). Technological advancement creates similar risks. Companies have come under scrutiny for strategies that clash with privacy rights, worker rights and national policies (Donaghey and Reinecke, 2018; Girschik et al., 2022). Alternative democratic systems are needed to deal with social risks created by these advances. CSR offers this alternative (Girschik et al., 2022; Shaw, 2007).

In a CSR democracy, governance can be enacted through philanthropy, public policy, strategy, social pressure, social unrest or advocacy (Frederick, 2018). In academia, advocacy typically occurs through inquiry. Academics within the CSR field have reasoned that scholars within the discipline are suitably positioned to use academic inquisition to engage in intellectual activism to drive the meaningful transformation that was initially envisioned for the social responsibility agenda (Girschik et al., 2022). The call to use CSR principles for intellectual activism is driven by a growing sentiment that CSR’s democratisation of economic activities is failing to deliver true social justice (Girschik et al., 2022; Sarkar and Searcy, 2016). The evidence for this failure is twofold. Firstly, the development of ineffective remedies for the negative impact of corporate action (Girschik et al., 2022; Sarkar and Searcy, 2016). Secondly, socially irresponsible acts have not stopped as corporate social irresponsibility (CSiR) – the dark side of CSR – continues to grow (Abbasi et al., 2024; Riera and Iborra, 2023). Some suggest that the failure is because of the highly contested nature of CSR (Sarkar and Searcy, 2016) while others argue that it is a result of a shift from socially responsible CSR to strategic CSR (Girschik et al., 2022; Mitnick et al., 2023). The true roots are still being debated.

Petrylaite et al. (2023, p. 2) have posed multiple questions to drive a focused inquisition into this failure including: “Is reliance on managerial discretions within corporations sufficient to meet the rising public expectations in reducing irresponsible business practices?”. We respond to this question by advocating for a continued focus on discretion. We, therefore, contribute to this ongoing CSR failure investigation by offering a discretion-centric perspective. We unpack discretion to determine the factors that influence how professionals drive responsibility or prevent irresponsibility so that the relationship between discretion and CSR is fully understood before it is questioned.

Numerous studies explore the CSR-through-discretion phenomenon (Alias et al., 2019; Key, 1996; Luth and Schepker, 2017; Xu, 2023). Most focus on executives and exclude recent advances in discretion theory. Discretion theory has been extended to include activity-level discretion, discretion domains, middle management and professional discretion (Caza, 2011, 2012; Finkelstein and Peteraf, 2007; Wajszczyk, 2016; Wallander and Molander, 2014). Furthermore, studies that explore CSR through discretion typically focus on the amount of discretion (whether high or low) or factors that influence discretion (Alias et al., 2019; Key, 1996). Few studies examine discretion constituents. Caza’s (2012) work is one of few that dissects discretion into parts but focuses on organisations and excludes CSR. We, therefore, posit that scholarly understanding of CSR failure is constrained by an underdeveloped CSR-through-discretion theory. To address this, we separate discretion into its multiple constituents and explore the influence of discretion anatomy and multiplicity on CSR/CSiR.

We examine this phenomenon within the engineering profession in South Africa. Engineering plays a significant role in the socioeconomic dynamics associated with globalisation and technological advances. Many global firms are led by engineering professionals and technology companies are typically engineering dependent (Børsen et al., 2021; Isaksson and Kiessling, 2023). However, CSR is currently not viewed as a core engineering matter (Isaksson and Kiessling, 2023). This study places CSR concepts within core engineering practice guidelines to offer terminology and a process for scrutinising the impact of engineering decision-making on CSR/CSiR.

We establish a model that combines Wood’s (1991) CSR perspective and Frederick’s (2018) pragmatic and futuristic mechanisms and replaces Wood’s (1991) managerial discretion principle with a professional discretion principle. This replacement enables the linking of discretion theory developments such as activity-level discretion, middle management, professional discretion and discretion domains with CSR theory. Our expansion of the discretion principle reflects the corporate social conditions of our time and ensures that it is suitable for application in the engineering context. We use a discretion-centric umbrella approach to link CSR and CSiR while managing the tension that stems from the contested nature of these concepts.

CSR has a long history. There is evidence of socially responsible practices as early as the 16th century (Latapí Agudelo et al., 2019). Social initiatives originally existed as practices that were informed by ancient laws, religion, charitable endeavours, welfare schemes and philanthropic values (Gond and Moon, 2011; Latapí Agudelo et al., 2019). In the 1920s, these practices took a more solid shape as managers were seen as trustees and stewards of social factors that were external to the company (Gond and Moon, 2011; Latapí Agudelo et al., 2019). Between the 1950s and 1960s, these social acts moved from practice to academia and the social responsibility term was formalised and spread. The shift to CSR occurred in the late 1960s as this label became increasingly popular (Latapí Agudelo et al., 2019).

To date, CSR has maintained its popularity, however, the concept continues to evolve. This evolution has resulted in the development of conflicting and complementary terminology (Gond and Moon, 2011; Masoud, 2017; Mitnick et al., 2021). Mohan (2003) developed a timeline depicting the escalation of various terms between 1920 and 2000. These include corporate philanthropy, public stewardship, businessmen’s social responsibilities, CSR, corporate social responsiveness, corporate social performance (CSP), stakeholder theory, sustainable development, triple bottom line and corporate citizenship. Later, Gond and Moon (2011) added two more labels to this timeline: corporate stakeholder’s responsibility and political-CSR. Mohan’s (2003) original timeline, and Gond and Moon’s (2011) updated timeline both exclude CSiR. This is not surprising as some CSR scholars view CSR and CSiR as separate and independent constructs (Abbasi et al., 2024; Riera and Iborra, 2017). Recent developments have resulted in more specific CSR descriptions such as strategic-CSR (Porter and Kramer, 2006), implicit-CSR and explicit-CSR (Matten and Moon, 2008), micro-CSR (Girschik et al., 2022) and moral-CSR (Mitnick et al., 2023).

CSR remains a continually evolving and contested concept (Gond and Moon, 2011; Mitnick et al., 2021). Scholars acknowledge the contestation, attribute it to differences in underlying worldviews and encourage continued debate (Gond and Moon, 2011; Mitnick et al., 2021; Windsor, 2022). CSiR is also contested and associated with varying conceptualisations (Clark et al., 2022). The origins of CSiR are associated with Armstrong’s (1977) conceptualisation that examined social responsibility dynamics from the perspective of what should not be done as opposed to what should be done. The relationship between CSR and CSiR is also contested. Some authors assert that CSR and CSiR are connected concepts that sit on opposite ends of a continuum (Windsor, 2022), while others argue that they are distinct and independent constructs (Abbasi et al., 2024). To manage the tension stemming from contestation, scholars propose the application of CSR as an umbrella term through broad concepts relevant to large communities, or the use of narrow CSR definitions to preserve validity (Gond and Moon, 2011).

This study uses CSR as an umbrella term for the following expectations (Gond and Moon, 2011, p. 2):

• business is responsible for and to society;

• business conducts itself responsibly; and

• business appropriately manages the corporation-society interface.

It defines CSR in the context of CSP and broadly applies Wood’s (1991) perspective. Wood (1991, p. 693) defines CSP with CSR as follows: “CSP is a business organisation’s configuration of principles of social responsibility, processes of social responsiveness, and policies, programs, and observable outcomes as they relate to the firm’s societal relationships”. This definition is viewed as an integrative perspective which is useful when dealing with contestable concepts as it counteracts fragmentation and manages over-theorisation (Gond and Moon, 2011; Windsor, 2022). The link between CSR and CSP is also encouraged because it connects responsibilities with performance (Carroll, 2018).

Armstrong (1977, p. 185) characterised CSiR by defining an irresponsible act: “A socially irresponsible act is a decision to accept an alternative that is thought by the decision maker to be inferior to another alternative when the effects upon all parties are considered”. Adding a second definition, Armstrong (1977, p. 185) outlined further that “an act was irresponsible if a vast majority of unbiased observers would agree that this was so”. Although Wood’s (1991) perspective briefly and implicitly links CSiR and CSP, several expositions and examples suggest that her conceptualisation accommodates CSR and CSiR as separate constructs. Plus, her 2010 review positions her 1991 model as based on principles that connect discretion and CSiR (Wood, 1991, 2010). We, therefore, view Wood’s (1991) perspective as an effective umbrella approach for exploring the role of professional discretion, managing contestation tension and integrating CSR, CSiR and CSP.

Several scholars propose CSR-related models that align with the umbrella approach. The most notable are Wartick and Cochran’s (1985) CSP model, Carroll’s (1991) pyramid of responsibilities, Windsor’s (2006) corporate citizenship metaphor and Frederick’s (2018) historical and speculative five-phase CSR framework.

Wartick and Cochran (1985) proposed a model based on economic responsibility, public responsibility and social responsiveness with limited detail on individual discretion. Carroll’s (1979, 1991, 2016) model involves four parts: economic, legal, ethical and philanthropic (discretionary) responsibilities. Carroll’s (1991) model limits discretion to philanthropic responsibilities (Wood, 2010). Windsor’s (2006) model proposes citizenship as an interpretative and unifying metaphor that involves discretion but omits CSP and responsiveness.

Wood’s (1991) model involves three fundamental principles (institutional, organisational and individual), three social responsiveness principles (environmental, stakeholder and issues) and three outcomes (impacts, programmes and policies). The individual principle is linked to managerial discretion. This model has limitations. “While Wood’s (1991) model represents a significant piece of scholarship, it nevertheless failed to address the needs of practising managers charged with implementing CSR/CSP programmes” (Meehan et al., 2006, p. 390). Consequently, authors who have extended Wood’s (1991) model typically incorporate practical aspects. This is evidenced by extensions to include normative aspects (Swanson, 1995), stakeholder dynamics (Wood and Jones, 1995), normative and descriptive elements (Swanson, 1999), metrics (Mitnick, 2000) and pragmatism (Blindheim and Langhelle, 2010). We acknowledge the shortcomings and various extensions. We propose that the pragmatism gaps can also be addressed by providing more detailed responsiveness processes. Wood (1991) limited these to environmental assessment, stakeholder management and issue management excluding the associated instruments and mechanisms. We, therefore, address this by replacing Wood’s (1991) social responsiveness processes with Frederick’s (2018) more pragmatic CSR perspective that links historical and speculative CSR doctrine to practical drivers and policy implementation instruments. Frederick (2018) presents an evolutionary framework by exploring the mechanisms that enable shifts in socially responsible thinking. Frederick (2018) identifies five phases based on the CSR doctrine that dominated at a specific historical time or the doctrine that is speculated to dominate in the future:

• CSR-1: corporate social stewardship (1950s–1960s);

• CSR-2: corporate social responsiveness (1960s–1970s);

• CSR-3: corporate ethics (1980s–1990s);

• CSR-4: corporate/global citizenship (1990s–2000s); and

• CSR-5: millennial future (2000–3000).

Each phase is underpinned by CSR drivers and accompanied by policy instruments. We incorporate the generalised descriptions of Frederick’s (2018) phase, driver and instrument factors as corporate social response mechanisms. This is because the factors are more detailed and action-oriented and can produce Wood’s (1991) measurable outcomes. Furthermore, combining Wood’s (1991) and Frederick’s (2018) perspectives brings CSR doctrine to the present and positions it for the future while integrating well-established CSR/CSiR/CSP relationships.

CSR-through-discretion is not a new concept. Between the 1950s and 1990s, authors who formalised CSR theory positioned discretion as an integral part of CSR (Carroll, 1979, 1991; Latapí Agudelo et al., 2019; Wood, 1991). Even Armstrong (1977, p. 185), who operationalised CSR by defining irresponsibility, emphasised decision-making. Towards the end of the 20th century, authors continued to cultivate this relationship through models. The models of Carroll (1979) and Wood (1991) applied the discretion principle as an umbrella term for decision, judgement and choice. Carroll (1979, p. 500) proposed discretionary responsibilities that are “left to individual judgement and choice”. Wood (1991, p. 699) proposed an individual discretion principle based on the proposition that “a company’s social responsibilities are not met by some abstract organisational actor; they are met by individual human actors who constantly make decisions and choices, some big and some small, some minor and others of great consequence”.

Relinquishing dependency on discretion is also not new to CSR. Carroll (1979) – who initially proposed a definition that included discretionary responsibilities as a standalone element – later conducted a review which replaced discretionary responsibilities with a philanthropic responsibility element (Carroll, 1991). This change limited discretion to the boundaries of philanthropy. In a subsequent review, Schwartz and Carroll (2003) removed the standalone philanthropic responsibilities dimension and positioned it under ethical responsibilities – further deemphasising the role of discretion. However, in a more recent review, Caroll (2016) presented philanthropic responsibilities as a standalone element but continued to limit discretion to philanthropy. Sarkar and Searcy’s (2016) review of various CSR definitions indicates that CSR scholarship typically positions discretion as a core dimension of CSR, however, several scholars limit it to voluntary and philanthropic activities.

Some scholars have maintained a widespread dependency on discretion and advocate for continued reliance on it. Wood’s (1991) model is based on a far-reaching dependency on discretion. In her 2010 review, she reiterated that her application of the discretion principle applies to all decisions made in every domain of CSR where moral actors are obliged to exercise the discretion available to them towards socially responsible outcomes (Wood, 1991, 2010). She clarified that her model is a major reinterpretation of Carroll’s (1979) discretionary principle to assert that “managers have choices in everything they do” to not choose socially irresponsible acts, and this choice is not limited to philanthropy (Wood, 2010, p. 56).

Wood’s (1991, 2010) conceptualisation offers a broad and consistent discretion principle. We build on her perspective to advance CSR as a potential transnational democratic mechanism. “Discretion is the black hole of democracy” (Rothstein, 1998, p. 80), and this black hole is at the heart of CSR (Wood, 1991). The fundamental principles of discretion involve structural concepts such as latitude, degree and space (Hambrick and Finkelstein, 1987; Wallander and Molander, 2014). Consequently, most studies that explore discretion tend to use a structural lens. This lens provides necessary insights into the characteristics of what Ronald Dworkin described as the “hole in the donut” (Dworkin, 1978, p. 32). These characteristics include its location (Caza, 2011, 2011; Finkelstein and Peteraf, 2007), how it forms (Wangrow et al., 2015), how it varies in size (Iyiola and Trafford, 2023), how it changes over time (Hutzschenreuter and Kleindienst, 2013), its homogeneity (Abrahamson and Hambrick, 1997) and its impact (Wangrow et al., 2015). However, this structural perspective does not address the elements found inside the black hole. It reveals broad cause-and-effect dynamics, but it does not say enough about discretion anatomy and multiplicity (Wallander and Molander, 2014).

Discretion occurs at four levels: individual, activity, organisational and environmental levels (Abrahamson and Hambrick, 1997; Hambrick and Finkelstein, 1987). Studies that explore the relationship between discretion and CSR often focus on a specific level (Luth and Schepker, 2017). It is understandable for researchers to isolate one variable when investigating cause and effect dynamics, but it is impossible for only one level of discretion to influence CSR in isolation. Practically, someone who exercises individual-level discretion does so for specific activities (activity level) in an organisation (organisational level) that operates in a certain context (environmental level). Thus, multiple types of discretion influence CSR/CSiR simultaneously. Key (1996) explored the influence of individual and organisational-level discretion, but few researchers have extended this work.

Wood’s (1991, p. 699) discretion principle relates to the exercise of discretion in every CSR/CSiR domain based on the following ideas:

• managers exist in an organisational and societal environment full of choices;

• managers’ actions are not completely prescribed by corporate procedures, formal job definitions, resource availabilities or technologies;

• managers are moral actors on the job as well as in other domains of their lives. Despite the existence of certain CSRs prescribed in various domains, managers have choices about how to fulfil many of these responsibilities; and

• the principle implies that because managers possess discretion, they are personally responsible for exercising it and cannot avoid this responsibility through reference to rules, policies or procedures.

We interpret these ideas as applicable to all individuals with decision-making capability based on Wood’s (2010, p. 54) assertion that her discretion principle applies to “managers and other employees as moral actors”.

Figure 1 presents the conceptual framework on which our study is based. It shows that CSR involves three fundamental principles (institutional, organisational and managerial discretion) that represent CSR expectations for organisations and individuals (Wood, 1991). We then extend this work by replacing Wood’s (1991) social responsiveness processes with Frederick’s (2018) phases, drivers and instruments. We also dissect professional discretion using components, range, levels and goal elements to extend Wood’s (1991) model downwards and examine the discretion anatomy and multiplicity factors that influence CSR, CSiR and CSP.

Professionals exercise discretion according to their expertise and can be authorised to act as employees within organisations and as independents/contractors/consultants who collaborate with organisations from outside. Thus, replacing Wood’s (1991) managerial discretion principle with a professional discretion principle extends the scope of decision-making to non-employees who act outside the organisation. Our motivation for this is threefold. Firstly, CSR is a “chameleon concept” that evolves based on conceptualisations that reflect the social dynamics of its time (Gond and Moon, 2011, p. 6). Secondly, CSR umbrella concepts are typically adjusted and expanded to suit the context in which they are applied. For example, Visser’s (2009) rearrangement of Carroll’s (1991) pyramid to reflect the role of the economy and philanthropy in developing countries. Similarly, we add to Wood’s (1991) characteristics to reflect that decision-making scope can rely more on expertise than employment status. Thirdly, representatives who exist outside of structured employment can contribute to the total decisions associated with an organisation. Some organisations, for example, Anglo American, include contractor social management as part of their CSP strategy to mitigate the negative impacts of economic activities and create socially beneficial outcomes through external parties (Anglo American, 2023). Thus, these additions align with theory and are supported by practice. Furthermore, there are non-human algorithms that contribute to the total number of decisions of an organisation and, thus, influence the social impact of economic activities (Friedman, 2014). Therefore, our proposed professional discretion principle applies to human and non-human actors who conduct work as internal or external agents associated with an organisation.

These additions align with Wood’s (1991, p. 699) conceptualisation that endeavoured to shift the focus of CSR-related discretion away from the “abstract organisational actor” towards the specific sources of decisions. Furthermore, our use of professional discretion updates Wood’s (1991) discretion principle based on more recent developments in discretion theory (Caza, 2011, 2012; Finkelstein and Peteraf, 2007; Hutzschenreuter and Kleindienst, 2013; Petersen et al., 2020; Wajszczyk, 2016; Wallander and Molander, 2014).

We use professional discretion to analyse the South African engineering profession’s relationship to the CSR-through-discretion phenomenon. South Africa is a developing country that has undergone social reordering and economic transition (Siyobi, 2015). Doś and Pattarin (2021) assert that countries with these characteristics have a distinct history, social nuance and neoliberal transitions that provide a unique example for CSR research. South Africa also offers a useful case study because its post-apartheid context uses socioeconomic reforms to correct historical CSiR (Siyobi, 2015).

In this study, discretion describes the latitude of decision or action that is exercised by engineers who have the following characteristics (Freidson, 2001; Ponnert and Svensson, 2016):

• belong to a profession;

• have specialised skills and knowledge that influence their judgement during decision-making; and

• practise within a discretionary specialisation, where all tasks demand professional judgement.

Based on these characteristics, we propose that engineers exercise professional discretion.

Professional discretion is influenced by occupational and organisational standards (Ponnert and Svensson, 2016). The study is limited to the occupational standards set by a professional body, the Engineering Council of South Africa (ECSA). ECSA governs the profession according to the standards set by the International Engineering Alliance (IEA) which describes engineering as “an activity that is essential to meeting the needs of people, economic development, and the provision of services to society” (IEA, 2021, p. 2). This description links economic activities with society making this profession an effective lens for exploring CSR.

This investigation is based on two main questions:

These questions form the problem statement used for conceptual theorising (Weick, 1989). To develop propositions (which we call principles), we use four sub-questions to establish the theory-building selection criteria described by Weick (1989):

The sub-questions are represented in the conceptual framework in Figure 1. What-if analyses are used by decision makers to evaluate impact before action is taken (Rizzi, 2009). This study proposes theoretical principles that can be used to conduct what-if analyses by both practitioners and researchers to further examine the CSR-through-discretion dynamic and study the impact of reliance on discretion at a detailed level.

We apply a methodology typical for theory-building conceptual papers (Gilson and Goldberg, 2015; Weick, 1989) using the model approach (Jaakkola, 2020) to identify previously unexplored connections between discretion constituents and CSR through Wood’s (1991, 2010) and Frederick’s (2018) conceptualisation. An engineering lens is used to introduce a new construct that links professional discretion components and CSR/CSiR/CSP principles.

This paper uses theoretical and discursive methods to collect and analyse literature to develop arguments for association (Gilson and Goldberg, 2015). The data collection and analysis sequence and approach are detailed in Table 1.

The findings are presented in four sections: discretion components and CSR are detailed in Finding I, discretion levels and CSR are described in Finding II, discretion range and CSR are examined in Finding III and goal discretion and CSR are discussed in Finding IV.

Discretion is the latitude to decide or act (Hambrick and Finkelstein, 1987). It has two sides: discretionary space (structural dimension) and discretionary reasoning (cognitive dimension) (Molander et al., 2012; Petersen et al., 2020; Wallander and Molander, 2014). Discretionary space is the area assigned to professionals to make decisions (Petersen et al., 2020). ECSA (2023) outlines that professional engineers (PEs) operate in an area that consists of complex engineering activities (CEAs) and complex engineering problems (CEPs) that must be amenable to engineering science. This specification defines the discretionary space in which PEs have the opportunity to act and decide (Molander et al., 2012).

Principle 1: PEs have a latitude to act or decide within a specific discretionary space. The discretionary space assigned to them consists of CEAs and CEPs that are amenable to engineering science.

Discretionary reasoning is the cognitive activities undertaken during decision-making (Molander et al., 2012; Petersen et al., 2020) and how decisions are justified (Wallander and Molander, 2014). The ability to undertake these cognitive activities is developed through formative dimensions associated with a specific profession and includes specific knowledge, modes of thinking, values and norms (Molander et al., 2012). PEs develop this dimension by obtaining specific knowledge and experience (ECSA, 2023). The knowledge is obtained through four-year engineering bachelor’s degrees and experience over a minimum three-year working period that involves CEPs and CEAs (ECSA, 2023). It enables PEs to perform cognitive activities within a team but without supervision while exercising sound judgement and exhibiting the required modes and norms of competent thinking (ECSA, 2023).

Principle 2: PEs have a latitude to act or decide using specific discretionary reasoning. The ability to reason is developed through training over a minimum seven-year education and experience period.

Discretion can be classified as formal or informal. Formal discretion exists within rules and standards and results from broad policies (Petersen et al., 2020; Vega et al., 2013). Informal discretion exists outside of rules and standards and results from subjective evaluation (Petersen et al., 2020; Vega et al., 2013). PEs are differentiated from other engineering professionals because they solve CEPs (ECSA, 2023). CEPs can be examined using engineering science which has rules and standards and, thus, CEPs can be associated with formal discretion. However, the complex characteristics of CEPs also link them with informal discretion. ECSA (2023) describes CEPs as complex because they are typically underspecified, abstract and fall outside of standards and rules necessitating the application of contextual professional judgement. Contextual judgement can result in subjective evaluation and, thus, can be linked with informal discretion.

Principle 3: PEs exercise formal discretion through the application of engineering sciences. PEs exercise informal discretion through the resolution of CEPs that are underspecified, abstract and fall outside standards and rules.

Although the space to act assigned to PEs is limited to the engineering discipline, ECSA (2023) instructs PEs to manage the negative effects that engineering may have on economic, social, cultural and environmental areas. This instruction extends the discretionary space of PEs to other areas to minimise CSiR in CEAs (Popa and Salanță, 2014). Furthermore, the discretionary reasoning applied by PEs must involve recognition and suitable consideration of the foreseeable economic, social, cultural and environmental impact of CEAs (ECSA, 2023). In this way, PEs exercise discretionary reasoning that foresees, recognises and considers links between CSR and CEAs.

Principle 4: PE’s discretionary space extends to economic, social, cultural and environmental areas to minimise CSiR in CEAs.

Principle 5: PEs exercise discretionary reasoning in a manner that foresees, recognises and considers the links between CSR and CEAs.

ECSA (2023) does not offer an explicit way to reasonably connect the management of informal and formal discretion principles with CSR. However, Wajszczyk (2016) proposes that engineering professionals operate within and outside of rules and standards when managing factors linked to ethical and moral principles where the use of rules is offset with discretionary actions in the form of exceptions depending on scale and the availability of contextual information. Literature suggests that engineers exercise formal and informal discretion in an interlinked manner when managing CSR-related aspects (Key, 1996; Wajszczyk, 2016). Moreover, Wood’s (1991) discretion principle is based on the notion that managers’ actions are not completely prescribed by corporate procedures or formal job definitions and they cannot avoid responsibility through reference to rules or procedures.

This thinking provides principles and perspectives that engineering bodies like ECSA and the IEA can use to revise their standards and establish guidelines that explicitly link informal/formal discretion and CSR/CSiR.

Discretion occurs at four levels (Hambrick and Finkelstein, 1987, p. 379; Finkelstein and Peteraf, 2007):

1. Individual level: the degree to which one can envision or create multiple courses of action.

2. Activity level: the degree to which activities permit change based on complexity, uncertainty and observability.

3. Organisational level: the degree to which the organisation is responsive to a list of possible actions.

4. Environmental level: the degree to which an environment allows change.

PEs solve CEPs by using individual judgement in uncertain contexts (ECSA, 2023). This indicates that PEs exercise individual-level discretion. PEs also exercise activity-level discretion because they handle underspecified and ambiguous problems within their specific activities (CEAs) (ECSA, 2023). To do this, they “exercise sound judgement by evaluating the outcomes, impacts and alternatives in the course of CEAs” ECSA (2023, p. 15). Engineers compliant with ECSA’s (2023) standards practise as employees within organisations or independent consultants who work with organisations and are, thus, subject to organisation-level discretion. The organisations in which PEs practise exist within various industries and disciplines, and ECSA (2023) governs PEs from the institutional level, thus, making PEs subject to environmental-level discretion. These four discretion levels occur concurrently as practising PEs exercise professional judgement to solve CEPs (individual-level discretion) and manage CEAs (activity-level discretion) while working with/for organisations (organisation-level discretion) in specific industries subject to ECSA’s (2023) institution-level authority (environmental-level discretion).

Principle 6: PEs are subject to four levels of discretion that manifest concurrently, thus, making PEs subject to discretion multiplicity.

At the individual level, ECSA (2023, p. 16) specifies that PEs are expected to apply sound judgement “in considering the interactions between conflicting engineering and social issues”. CSR is typically used to describe efforts made to manage the interactions between economic activities (i.e. engineering) and social issues (Popa and Salanță, 2014). Based on this, PE’s experience of individual-level discretion involves sound judgement that aligns with CSR principles. Furthermore, at the activity level, ECSA (2023) stipulates that PEs take measures to mitigate the negative impact of CEAs whether social, cultural or environmental. Thus, PE’s experience of activity-level discretion aligns with the minimisation of CSiR. Organisational and environmental discretion’s connection with CSR cannot be inferred using the professional standard as organisations and industries use contextual drivers and instruments to achieve CSR objectives (Frederick, 2018).

Principle 7: PE’s individual-level discretion involves the application of sound judgement in considering interactions between engineering and social issues (i.e. CSR).

Principle 8: PE’s experience of activity-level discretion involves action towards the minimisation of CSiR.

ECSA (2023) does not outline PE standards that can be reasonably related to the interaction between discretion multiplicity and CSR. However, Smith et al. (2021) indicate that when implementing CSR policies, engineering professionals negotiate how to balance conflicts between their ethics (individual-level), their role (activity-level), employee status (organisational-level) and professional practices (institutional-level). Set in the American context, their study revealed that engineers who work in the mining industry typically manage multiplicity by prioritising the CSR practices determined by their organisations (Smith et al., 2021). This choice is not typical in every context. For example, in the South African mining context, engineers are legally appointed to manage social risks based on the competencies and standards of their profession (Mine Health and Safety Council, 2018). This indicates that CSR practices are to be determined by professional requirements. ECSA (2023) does not indicate which aspects should be prioritised when managing multiplicity and broadly instructs PEs to balance all sides by using a holistic approach to manage sociocultural impacts while ensuring the sustainability of economic activities. Literature offers more specific perspectives but their generalisability is limited. More expansive research is required that explores prioritisation approaches for managing discretion multiplicity and CSR/CSiR. Furthermore, more specific practice standards are required to better guide practitioners on prioritisation approaches for managing discretion multiplicity in the CSR/CSiR context.

High complexity and uncertainty are associated with high levels of discretion (Finkelstein and Peteraf, 2007). Because PEs solve CEPs and manage CEAs that involve high discretion due to complexity and ambiguity (ECSA, 2023), they can be classified as “high-discretion managers” who experience a discretionary range that can be expanded (Finkelstein and Peteraf, 2007, p. 243).

Principle 9: PEs are “high-discretion managers” who can expand their discretionary range.

Hutzschenreuter and Kleindienst (2013, p. 265) described discretion as the meeting point of two independent dimensions:

1. The list of options that exist in the zone of acceptance.

2. The list of options that a manager is aware of. Awareness is a list of options that a manager can perceive and is influenced by stimuli that determine where attention is placed.

The first dimension can be classified as objective discretion, and the second as perceived discretion (Caza, 2011). Researchers propose that professional development can expand perceived discretion over time (Hutzschenreuter and Kleindienst, 2013; Caza, 2011). PEs are required to undertake professional development activities that enhance competence and enable them to adapt to the changing nature of work (ECSA, 2023). This mandatory extension of competence involves interventions that increase perceived discretion over time and expand the discretion range of PEs.

Principle 10: PEs expand their discretion range by undertaking professional development activities that increase their perceived discretion over time.

Discretionary specialisation occurs when all tasks require professional judgement and this specialisation requires situational and frequent developmental interventions to increase perceived discretion (Freidson, 2001). Owing to the ambiguity and complexity characteristics of CEPs, all CEPs occur within discretionary specialisation because they demand professional judgement (ECSA, 2023). Consequently, PEs operate within a discretionary specialisation, and, thus, situational and frequent developmental interventions are required to increase their perceived discretion.

Principle 11: Situational and frequent developmental interventions are required to increase the perceived discretion of PEs.

Developmental elements that increase perceived discretion do not necessarily mean that PEs will allocate their attention to the provided stimuli when faced with a decision. It is ultimately the individual who decides which stimuli they allocate attention to when exercising discretion (Hutzschenreuter and Kleindienst, 2013). This can be impacted by personal, relational or situational factors.

Principle 12: PEs’ perceived discretion may not increase if they decide not to allocate attention to the various stimuli provided by developmental interventions for increasing awareness of options.

ECSA (2023) does not specify the developmental interventions that can be directly linked with CSR. Literature also does not provide much guidance on this for engineering practitioners. CSR-related developmental interventions in engineering tend to focus on students in undergraduate engineering studies (Børsen et al., 2021; Brodeur, 2013). This is expected because practitioner interventions need to be situational and CSR needs tend to be field and dilemma-specific (Smith et al., 2021). The few studies that exist suggest that practitioner interventions can include mentorship, role-based interventions and postgraduate studies (Brodeur, 2013; Smith et al., 2021). More research is required in this area to develop literature that explicitly links discretion range and CSR/CSiR.

Goal discretion is the extent to which one can decide what they are trying to achieve (Caza, 2012). It is exercised over the desired outcome and the criteria used to evaluate it. Caza (2012) asserted that this type of discretion is low for general staff and supervisors, moderate for managers and high for executives. Since PEs manage CEAs and can work as independent consultants (ECSA, 2023; South African Government, 2000), they can operate at manager or executive levels with moderate or high levels of goal discretion. PEs who operate at the manager level have moderate goal discretion and, thus, can influence desired outputs and the criteria used to evaluate them (Caza, 2012). Alternatively, PEs who operate at executive levels have high goal discretion and can choose desired outputs and the criteria used to evaluate them (Caza, 2012).

Principle 13: PEs who operate at the manager level have moderate goal discretion and, thus, can use their latitude to influence desired outputs (i.e. advance CSR/minimise CSiR) and the criteria used to evaluate them.

Principle 14: PEs who operate at the executive level have high goal discretion and, thus, can use their latitude to choose desired outputs (i.e. advance CSR/minimise CSiR) and the criteria used to evaluate them.

The professional latitude to influence or choose desired outputs does not guarantee that the choice will favour the advancement of CSR or the minimisation of CSiR. The relationship between CSR, CSiR and CSP can be leveraged to achieve this. To establish this principle, this study makes the following assertions based on Carroll’s (2018), Ha et al.’s (2019), Popa and Salanță’s (2014) and Wood’s (1991) perspectives:

• CSR and CSiR exist as separate constructs;

• CSR means that a positive/desired social performance result is achieved. CSiR means that a negative/undesired social performance result is achieved;

• CSiR is broadly described as a failure to achieve the appropriate balance between economic, environmental and social requirements;

• an organisation can exhibit strong CSR attributes and a high level of CSiR attributes because an organisation can be socially responsible in one domain and socially irresponsible in another; and

• moral actors can exercise discretion toward socially responsible acts (CSR) and avoid socially irresponsible acts (CSiR).

PE’s goal discretion must, therefore, be directed to align with these assertions to ensure that an organisation does not behave irresponsibly in one domain but responsibly in another. Thus, this study proposes that to achieve the desired CSP output, the goal discretion exercised by professionals should be concurrently directed to advance CSR and minimise CSiR.

Principle 15: The desired CSP outputs can be achieved by directing PE’s goal discretion to concurrently (1) achieve an appropriate balance between economic, environmental and social requirements (advance CSR) and (2) prevent a failure to achieve an appropriate balance between economic, environmental and social requirements (prevent CSiR).

Specific instruments and drivers are used to achieve CSR objectives and, thus, the desired CSP result (Frederick, 2018). Three ECSA (2023) outcomes show that professional bodies can drive CSR and PEs can, therefore, act as instruments that facilitate the attainment of CSR goals. ECSA (2023, pp. 14–15) specifies three outcomes that are exhibited by PEs:

• outcome 6: “recognise the reasonably foreseeable economic, social, cultural, and environmental effects of CEAs seeking to achieve sustainability”;

• outcome 7: “meet all legal and regulatory requirements and protect the health and safety of persons during all CEAs”; and

• outcome 8: “conduct engineering activities ethically” range statement: “ethical behaviour involves the comprehension and application of professional ethics, responsibilities and norms of engineering practice within one’s limits of competence”.

Outcome 6 aligns with the CSR goal of achieving an appropriate balance between economic, social and environmental imperatives (Popa and Salanță, 2014). Outcome 7 aligns with CSR’s objective to ensure the well-being of society (Frederick, 2018). Outcome 8 aligns with the moral and ethical aspects used to mediate personal motivations that may conflict with CSR (Wajszczyk, 2016).

Principle 16: Professional bodies act as drivers of CSR by specifying competency outcomes that align with CSR principles. PEs act as instruments by complying with these standards.

An analysis of the standard set for South African engineers, CSR theory and discretion theory shows that discretion factors such as space, reasoning, levels, range and goals can be related to CSR/CSiR and ultimately linked to CSP. We propose 16 principles that integrate these concepts through a CSR-CSP-professional-discretion model (Figure 2). Principles 1–5 address discretion components (space, reasoning, formal and informal), principles 6–8 deal with discretion levels, principles 9–12 focus on discretion range and principles 13–16 centre goal discretion.

The proposed principles are based on what-if questions (Figure 1). The principles can be used to conduct what-if analyses and scenario forecasting to explore the impact of reliance on professional discretion on CSR/CSiR/CSP. Given the uncertain nature of business and advances in data-based technologies, simulations are widely used to support tactical and strategic business decisions (Ben Rabia and Bellabdaoui, 2022; Golfarelli et al., 2006). CSR scenarios are incorporated into decision-support planning systems because positive CSP remains a priority even in turbulent business conditions (Athanasopoulou, 2008; Cairns et al., 2016) and what-if analysis is a popular technique (Ben Rabia and Bellabdaoui, 2022). Simulation granularity is one of the major practical challenges associated with implementing this technique (Golfarelli et al., 2006). Simulators can struggle with inputs that are too high-level rendering the model unreliable, and there are instances where different events require different levels of granularity (Golfarelli et al., 2006). Our model enables the exploration of the CSR-through-discretion dynamic at various levels of granularity since it includes principles and their elements within a larger CSR/CSP environment.

A focus on engineers shifts the CSR-through-discretion conversation away from executives where research tends to focus (Manner, 2010). Professional engineers exist in higher numbers and tend to be located at the middle management level. This places engineers in a unique position to act as confluences for executive-level strategic mechanisms that establish CSR objectives, and frontline implementation mechanisms used to implement CSR policy (Lipsky, 1997). Lipsky (1997) asserted that those who implement policies are the true policymakers because they exercise discretion to build the living policy as they translate standards into action. In this way, living policies are formed by those who are closer to where execution happens as opposed to the executives who authorise policies (Vega et al., 2013).

Some industries, like mining, are historically associated with high levels of CSiR and are composed of organisations that are typically led by engineers (Smith et al., 2021). In South Africa, mining is regulated through reform policies to correct historical CSiR which resulted from decisions made by professionals in the execution of technical work during apartheid (Siyobi, 2015). We centre professional discretion within the South African context to enable the continuation of the rectification of historical CSiR and to provide insights for countries with similar histories.

Integrating Wood’s (1991) model with a professional standard counteracts the limitation identified by Meehan et al. (2006, p. 390) who argued that the model did not enable implementation. This limitation is further addressed by replacing Wood’s (1991) social responsiveness processes with Frederick’s (2018) more pragmatic drivers/instrument and extending Wood’s (1991) model downwards to separate discretion into its various practice-oriented constituents. Incorporating Frederick’s (2018) futuristic perspective offers a mechanism for future empirical research that explores the discretion-centric millennial future of CSR. This paper bridges the gap between CSR and recent developments in discretion theory. The proposed theoretical principles can be used as a foundation for empirical research to further examine the CSR-through-discretion dynamic. These principles can be used to study the impact of reliance on discretion at a detailed level – using a professional lens.

This inquiry revealed gaps in ECSA’s (2023) standard, discretion research and CSR literature. The proposed model provides a theoretical basis for future CSR research that addresses these gaps. Principle 3 can inform future CSR research that explores the interlinked nature of formal and informal discretion. Principles 6–8 can be used to develop a theory on CSR-centric prioritisation methods for discretion multiplicity. Principles 9–12 can inform future inquiries on perceived discretion interventions for advancing CSR and preventing CSiR.

By focusing on the influence of individual professional discretion, this study aligns with emerging CSR concepts (Girschik et al., 2022). Micro-CSR is proposed as a transformative new conceptualisation of CSR because it engages closely with how individuals work with and experience CSR. Girschik et al. (2022) argue that this transformation can be achieved by advancing micro-CSR research through unconventional empirical research paths that explore employee activism, insider/outsider roles and challenge managerial control. This study enables future transformative research in this area to explore the following: the employment of discretion space, reasoning, levels, range and goals to facilitate employee activism; the use of a professional standard to expand the discretion scope and enable the incorporation of insider/outsider roles; and the application of professional discretion to shift the focus away from executives and challenge managerial control.

Although the study is limited by its theoretical approach, the principles are developed in an explicit and observable manner to enable further analysis and empirical testing. We invite fellow engineering practitioners to create inputs for future empirical research by connecting their lived experiences with the proposed principles. We urge fellow researchers to conduct future empirical research to test, challenge and build on the proposed principles.

The generalisability of the study is limited by the scope of the professional body standard. Informal discretion, formal discretion, discretion multiplicity and discretion range are not directly linked with CSR concepts in the proposed model because the ECSA (2023) standard does not offer explicit guidelines related to these elements. However, since these factors exist within professional discretion which sits within the CSR/CSP model, they are indirectly and ultimately connected to CSR/CSiR and CSP. In addition, our review highlights literature that addresses these relationships.

Our model directly links professional activities and the goals used to direct them with CSR/CSiR and ultimately CSP. It shifts focus towards the frontline where CSR policy is actioned and brings discretion closer to real life to avoid “passing on the myth of discretion”(Petersen et al., 2020, p. 304). We outline the two sides of discretion (space and reasoning) which are often overlooked (Molander et al., 2012), expand Wood’s (1991) model and contribute to the ongoing investigation into CSR’s failure to deliver on its promise. By applying the proposed principles, professionals can develop effective CSR strategies and act as impartial observers who not only define socially irresponsible acts but also work towards preventing them. We offer principles for researchers to advance micro-CSR and engage in intellectual activism to make CSR democracy transformational by asking: “What if professional discretion strongly influences corporate social responsibility?”