How corporate social responsibility and sustainable development functions impact the workplace: A review of the literature

This report sets out to analyse the emergence and distinctive impact of corporate social responsibility and sustainable development (CSR/SD) functions and professionals within organizations. By evaluating the literature on this topic, it seeks to clarify how leveraging the already established CSR/SD functions and professionals across organizations can contribute to the International Labour Organization’s (ILO) objective of achieving a future of work that provides decent and sustainable work opportunities for all.


Tackling four key research challenges 58
3.2.1. Challenge #1: Isolation -The CSR/SD boundary processes at the individual, organizational and institutional field level have been studied in isolation; the fragmentation of knowledge about them undermines the potential collective impact of CSR/SD functions and professionals 59 3.2.2. Challenge #2: Disempowerment -Although CSR/SD functions enhance the reactivity of organizations, they remain insufficiently resourced and fragile; CSR/SD professionals are continuously at risk of losing their capacity to influence organizations 60 3.2.3. Challenge #3: Glocalization -CSR/SD boundary processes are subject to tensions between global and local forces that can undermine the implementation of CSR/SD initiatives across organizations, supply chains and countries 60 3.2.4. Challenge #4: Marginalization -CSR/SD boundary processes follow a top-down logic, which may fail to take into account marginalized actors (e.g. workers, women and the global South) 61  Table 1. Academic and practice-based definitions of CSR and SD 15 Table 2. Related reviews: Areas of overlap and differences 17 Table 3. Boundary processes driving the emergence of CSR/SD functions and professionals 24 Table 4. Boundary processes linking CSR/SD functions to employees and trade unions: Key concepts and illustrative studies 44 Table 5. Knowledge gaps and possible questions for further research 55

Executive Summary
This report sets out to analyse the emergence and distinctive impact of corporate social responsibility and sustainable development (CSR/SD) functions and professionals within organizations. By evaluating the literature on this topic, it seeks to clarify how leveraging the already established CSR/SD functions and professionals across organizations can contribute to the International Labour Organization's (ILO) objective of achieving a future of work that provides decent and sustainable work opportunities for all. An extensive and integrative review of the academic literature was undertaken and an interview with a panel of academic experts conducted in order to highlight various aspects of CSR/SD functions and professionals. The focus was on three core topics: the embedding of CSR/SD functions and professionals (Topic #1); their role in managing stakeholder relations (Topic #2); and, more specifically, their contribution to the shaping of interactions with employees and trade unions (Topic #3).
While CSR and SD have different historical roots, the two concepts overlap significantly. The umbrella term "CSR/SD" is therefore used throughout the report. "CSR/SD" itself is defined as encompassing corporate interactions with society and in particular with the multiple stakeholder groups from the corporate environment. "CSR/SD professionals" are defined as organizational actors either working within CSR/SD functions or whose role and activities at least have to do with managing CSR/SD issues. These functions are not a new phenomenon: a first wave emerged in the late 1960s and early 1970s. Renewed managerial interest in CSR/SD matters triggered the second wave in the 1990s, which was, paradoxically perhaps, strengthened by the 2008 financial crisis.

X Introduction What do we know? A boundary-process framework
Recognizing that a core characteristic of CSR/SD functions and professionals is their location at the interface of organizations and their environments, this report adopts the notion of a "boundary process" -that is, a process linking internal and external political dynamics -in order to survey the academic literature and organize it into a coherent framework (as presented in Figure 1). This framework integrates analysis at the individual, organizational and institutional field levels. An institutional field is defined as sets of organizations that together constitute an institutional domain (in our case, the CSR/SD domain); these may involve not only corporations, but also their suppliers, consumer associations, labour unions, consultancy organizations and multiple types of regulatory agencies (see DiMaggio and Powell 1983).

Embedding CSR/SD functions and professionals
This framework was first used to study the mechanisms responsible for the embedding of CSR/SD functions and professionals across the different levels considered in this report. At the macro level, broad regulatory, market and professionalization dynamics foster the development of CSR/SD functions and create a need for CSR/SD professionals within organizations. Three trends may be identified in that respect: • Regulation: A global increase in legislation contributes to the expansion of CSR/SD functions and to the professionalization of this institutional field. This is supported by non-governmental standards, which encourage a culture of self-reflection within organizations and thus promote CSR/SD.
• Commodification: Environmental and social issues have been turned into commodities sold by consultancy firms on the CSR/SD market. Consultancies across the world raise awareness of and "translate" regulations and standards, often bringing together companies from different industries to share best practices. These consultancies reshape the meaning of CSR/SD and reinforce a business case rhetoric when it comes to the adoption of CSR/SD practices by organizations.
• Professionalization: The increasing professionalization of CSR/SD involves the establishment of informal and formal communities of practice, which may sometimes take the form of professional associations. In addition to serving as a platform for CSR/SD practitioners, these also draw up standards and provide certification for their members. Such networks of practitioners equip their members with shared knowledge and a common identity and purpose; they are largely characterized by passion for their work. However, the professionalization of the CSR/SD institutional field raises certain questions, such as the lack of a clearly defined CSR knowledge base, the multiple meanings attached to CSR, the absence of clear boundaries in the institutional field, and the marginalization of the work of CSR/SD practitioners.
At the meso level of analysis, it has been pointed out that the organizational-level deployment of CSR/SD functions takes place through structuring, translation and implementation/decoupling. Scholars are divided as to whether CSR/SD functions are condemned to marginalization or are a necessary step in the transition towards a sustainable future. The focus of research has been on: • Structuring: A number of studies have looked at the structuring of CSR/SD functions from "top-down", "bottom-up" or "middle-focused" perspectives. A top-down approach concentrates on how CSR/SD positions in senior management influence CSR/SD performance depending on the regulatory environment, previous performance, and an organization's choice of chief sustainability officer, where efforts are more substantive than symbolic. The bottom-up approach investigates employees' involvement in the successful implementation of CSR/SD across the organization. It is the middle manager who is often tasked with implementing CSR/SD strategies, despite not having sufficient authority or discretion and being in the midst of "turf wars" with other areas of the organization, such as human resources.
• Translation: CSR/SD functions are not a passive audience of social and environmental issues but actively interpret and translate these issues, as a result of which they also neutralize them. This may lead to routine solutions that are unlikely to address the issues successfully.
• Implementation and decoupling: How CSR/SD is implemented within an organization is also related to CSR/SD functions. Different stages of organizational "maturity" about CSR/SD will influence the size and expertise of these functions, as well as their relationships with other functions. The legitimacy of the functions is reflective of the ability of practitioners to achieve further maturity within their organization. CSR/SD functions play a role in managing potential gaps between organizational policies and discourses on the one hand, and practices on the other (as in the case of "greenwashing"). Studies have found that companies go through cycles of wider and narrower gaps, and that an initially wide gap can lead to discourses and practices that are more aligned. This is confirmed by studies which suggest that corporate hypocrisy can be helpful in triggering further action through aspirational talk, behind-the-scenes work and irony. Nowadays, studies documenting gaps are gradually being replaced by examinations of the actual impact achieved by CSR practices.
Lastly, at the micro level of analysis, research has documented how individuals shape CSR/SD functions through their roles and everyday tasks, as played out through: • Institutional work: The purposive actions undertaken by individuals within organizations to sustain CSR/SD-related institutions, such as activities aimed at making CSR or SD part of corporate strategy.
• Self-work: The reflexive work of CSR/SD employees to define their new professional identities and deal with the emotionally loaded aspects of their roles.
• Competency-building: The acquisition of the requisite skills, know-how and forms of behaviour that can help ensure that CSR/SD is implemented successfully within organizations.

Managing stakeholder relations through CSR/SD boundary processes
The boundary-process framework was next used to analyse the external stakeholder pressures on CSR/SD functions and the various relationships that these functions create with stakeholders. The present literature review suggests that both confrontation (activist targeting) and collaboration (partnership) are effective processes for CSR/SD adoption and change: • Confrontation -Dealing with external stakeholder pressure: Research shows that companies are more likely to adopt CSR/SD structures and practices if they are targeted by activist shareholders. Companies tend to be influenced more by their primary stakeholders (employees, investors, customers and governments) than by their secondary ones (such as non-governmental organizations (NGOs) and the media). Activist campaigns by secondary stakeholders tend to be more successful if they are stigmatizing and reputationally threatening, especially if the company has made previous CSR commitments and claims. This may lead to companies hiding sustainability credentials to avoid being labelled as hypocritical. However, research also suggests that, even if companies make only a small initial commitment, this can grow into more substantive change as they become more responsive to activist targeting.
• Collaboration -Managing stakeholder relations through multi-stakeholder partnerships: Studies have found that the involvement of third parties substantially improves the relationship between business and civil society in partnerships. Effectiveness is further improved by distributed decision-making, conflict resolution procedures and controlling the size of partnerships. However, as companies mature in their engagement and move towards a proactive sustainability strategy, they may become less responsive to issues that they deem to fall outside the scope of that strategy.

Shaping interactions with employees and trade unions through CSR/SD boundary processes
Lastly, the boundary-process framework was used to analyse the relationships between labour issues and employees and CSR/SD functions at the macro, meso and micro levels. At the macro level, research has highlighted the rise of private governance of work-and labour-specific CSR/SD and soft regulations initiated by transnational companies or transnational multi-stakeholder platforms, focusing in particular on: • The institutionalization of private governance throughout companies' supply chain networks and through the development of codes of conduct, international framework agreements and supplier programmes.
• The conditions shaping the effectiveness and social impact of soft regulations on companies' supply chains: Studies dealing with this aspect suggest that the social impact of soft regulations remains limited, owing to the competitive environment, the pressure to deliver, the lack of union and worker involvement in the development and implementation of standards, vague measurement indicators, and national constraints on democratic unionism. Nevertheless, studies have highlighted a few examples of trade unions being successfully involved in global production networks and the local implementation of standards.
• The outcome of private governance processes: Research shows that codes of conduct for supply chains increase demand for transparency and enhance global coordination across an industry. The impact of private governance is also influenced by the interdependence of stakeholders and the development of shared norms.
At the meso level, the present review shows that studies have focused on trade unions and, in particular, on: • The participation of workers and unions in the development of private governance: Although this participation is assumed to be limited, studies have showcased a number of relevant best practices. Unions could play a key role in developing these standards and monitoring their implementation.
• The interactions of trade unions with CSR/SD functions and professionals: There is evidence of trade unions driving the adoption of CSR/SD functions and practices through either collaboration or increased pressure. The degree of unionization influences the orientation of corporate CSR/SD policies. Unions may partner with NGOs to maximize their impact.
• The social responsibility of unions themselves: Some studies have focused on the positive impact of social responsibility on employees' engagement with trade unions. This impact is analysed in the context of the multi-stakeholder interactions between local governments, NGOs, companies and other unions.
Lastly, at the micro level of analysis, the literature reflects the following trends: • Studies tend to focus on employee perceptions of CSR/SD, rather than on employees' level of engagement with CSR/SD programmes. Where they focus on employee engagement, this process is described as encompassing transactional, relational and developmental components.
• A large number of boundary conditions which shape employees' reactions to CSR/SD have been identified, such as the ethical climate, the culture, the prevalent type of leadership and job-related factors (for example, job security).
• Studies have pointed to numerous employee-level outcomes of CSR/SD, showing, in particular, the positive behavioural, attitudinal and cognitive changes arising from enhanced perceptions of CSR/SD. However, some authors have also warned that more research needs to be undertaken on the potential dark side of CSR/SD policies and programmes, which could be turned into mechanisms of ideological control.

Where should we go from here? Four challenges and a research agenda
Using the insights gained from a systematic review of the literature based on the boundary-process framework, together with interviews with a panel of six academic experts about the key topics covered, this report identifies four current challenges: 1. Isolation: The CSR/SD boundary processes at the individual, organizational and institutional field levels have been studied in isolation; the fragmentation of knowledge about them undermines the collective action of CSR/SD professionals.

Disempowerment:
Although CSR/SD functions and the related boundary processes enhance the reactivity of organizations, they remain insufficiently resourced and fragile; CSR/SD professionals are continuously at risk of losing their capacity to influence organizations.
3. Glocalization: CSR/SD boundary processes are subject to tensions between global and local forces that can undermine the implementation of CSR/SD initiatives across organizations, supply chains and countries.

Marginalization:
CSR/SD boundary processes follow a top-down logic, which may fail to take into account marginalized actors (such as workers, women and the global South).
To tackle these challenges, a research agenda is proposed with four key priorities: • Research direction #1: Bridging CSR/SD boundary processes vertically -to clarify how CSR/SD functions and professionals operate at the institutional field (macro), organizational (meso) and individual (micro) levels of analysis -and horizontally -by clarifying how processes such as professionalization and the emergence of communities of practice interact. This could help consolidate and expand current knowledge. Specifying and leveraging these connections could make CSR/SD functions and professionals much more effective at helping organizations tackle CSR/SD issues.
• Research direction #2: Reconsidering temporal dynamics in the analysis of how CSR/SD boundary processes are formed within organizations and operate across multiple organizations. This could help make these processes better able to shape CSR/SD-related outcomes.
• Research direction #3: Exploring and harnessing the regulatory potential of CSR/SD boundary processes. Future research could focus on dynamics for mobilizing, empowering and/or developing the vast networks of CSR/SD experts and professionals with a view to regulating CSR/SD indirectly. Such studies could help evaluate the effectiveness of various forms of soft and hard regulation related to CSR/SD functions and experts.
• Research direction #4: Redesigning CSR/SD boundary processes to take better account of marginalized voices, for instance by developing research methodologies that are able to reach workers in long supply chains who may struggle to make their concerns heard. This also calls for attention to be refocused on the global South and on hidden aspects of supply chains.
Implementing such a research agenda will require multi-methods designs (for example, combining ethnography and quantitative analyses) and novel empirical methods (such as fuzzy-set qualitative comparative analysis). This endeavour is likely to benefit from innovative forms of cooperation between researchers and a variety of stakeholders, such as governments, private corporations, standard-setting organizations and trade unions.
To conclude, CSR/SD functions and professionals have considerable potential for resocializing organizations into their political, social and ecological environment, but this potential is far from fully realized. The research directions offered in this report seek to maximize the contribution of CSR/SD functions and professionals to the development of organizations' capacity to tackle major challenges such as climate change, the loss of biodiversity, and the protection of labour and human rights along extended transnational supply chains.

X 1 Research background, definitions and methodology
After presenting the background to the research (1.1), the definitions of corporate social responsibility (CSR) and sustainable development (SD) used in this report are specified and the research undertaken is put into context (1.2). Finally, the methodology underlying the literature review is outlined (1.3).

The continuous rise of interest in corporate social responsibility and sustainable development in the light of contemporary challenges
Although the notion that businesses should treat their constituents responsibly and show some consideration for the ecological environment on which they rely is probably as old as the capitalist system itself (Bowen 1953), over the past 30 years there has been an unprecedented upsurge in interest in CSR and SD across managerial and political spheres at local, national and transnational levels. Both concepts have to do with organizations' relationships with their key stakeholders and point to "context-specific organizational actions and policies that take into account stakeholders' expectations and the triple bottom line of economic, social, and environmental performance" (Aguinis 2011, 855). Several factors have driven the continuous rise of interest in CSR and SD, often referred to as the "mainstreaming" of these two concepts, in particular the following four trends: 1. CSR and SD "strategifying" (Gond, Cabantous and Krikorian 2018*) 1 , that is, the progressive recognition, in the academic field of strategy and in the managerial world, that CSR and SD should and do form an integral part of organizations' strategies. This trend is related to the embracing of CSR by influential scholars such as Michael Porter (Porter and Kramer 2006; and the emergence of a vast "business case" literature (Carroll and Shabana 2010) focused on establishing how CSR and SD initiatives can pay off for corporate actors by strengthening their competitiveness or reshaping the attitudes and behaviours of their stakeholders (for a meta-analysis, see Vishwanathan et al. 2020). The fact that CSR and SD have become frequent topics of discussion in boardrooms and are raised by institutional investors at general assemblies testifies to the strategic mainstreaming of these two concepts.
2. CSR and SD materialization, that is, the emergence and development of an extensive material infrastructure of standards, metrics, reporting, insurance procedures and business frameworks designed to support practitioners working in the domains of CSR and SD (Gond and Nyberg 2017;Waddock 2008). This material infrastructure is underpinned by regulations and by the institutionalization of CSR and SD as markets -the "markets for virtue" (Vogel 2005) that have been resilient to shocks as severe as the global financial crisis of 2007-08 and the coronavirus disease (COVID-19) pandemic of 2020.
3. CSR and SD globalization, which reflects the increased and continuous flows of workers and goods across borders and the outsourcing of multinational corporations' (MNCs) activities across long supply-chains that cut across geographical borders -at least until the COVID-19 pandemic. CSR and SD globalization is also related to financialization, as a growing number of institutional investors export CSR and SD standards from developed (mainly Western) countries to other parts of the globe through the pressure they exercise on their investee companies across multiple financial markets.
4. CSR and SD politicization, which is captured by frameworks such as Matten and Moon's (2008) distinction between "implicit" and "explicit" CSR, or Matten and Crane's (2005) analysis of "corporate citizenship". These frameworks show how institutional contexts shape local and political understanding of CSR. They also highlight the blurring of boundaries between the roles of governments and MNCs, turning de facto private organizations into political actors (Scherer and Palazzo 2011). CSR politicization has turned into a research field in its own right -referred to as "political CSR" -and has led to studies looking at the role of governments in relation to CSR and SD (see, for example, Knudsen and Moon 2017;Kourula et al. 2019).
All these trends have contributed to CSR and SD becoming firmly embedded within organizations. One of the most salient illustrations of this is the enduring institutionalization of CSR and SD functions and the sustained growth in the number of CSR and SD professionals. Tellingly, in the wake of the COVID-19 pandemic, there has been an explosion in CSR/SD initiatives by leading economic players and corporations all over the world with the aim of supporting local communities and hospitals, facilitating the production of essential products such as masks or hydroalcoholic gels, reconverting production lines and designing new components to help to manufacture ventilators, and protecting the health and safety of workers (Aguinis, Villamor and Gabriel 2020). Organizations have leveraged earlier CSR/SD initiatives or designed new ones from scratch in an attempt "to do the right thing". Admittedly, some of these initiatives have backfired; for example, Amazon's expansion of its online grocery delivery platform to provide people living in confinement with essential services and the donation of US$100 million to food banks in the United States of America by its chief executive (Amazon 2020) could not prevent criticism of the poor safety conditions provided for the company's logistics workers (Aguinis, Villamor and Gabriel 2020).
The example of Amazon illustrates the complexities, potentialities and limitations of the roles that CSR and SD functions and professionals are able to play. It serves as a reminder that the rise of CSR/SD functions and professionals also triggers legitimate questions about their actual impact in the workplace and within the broader environmental and social context of organizations. This report sets out, accordingly, to review studies of what impact CSR and SD functions and professionals have on the workplace.

Defining CSR/SD functions and contextualizing their emergence
The review undertaken for this report focuses on the development, within organizations, of functions and professionals dedicated to the management of CSR and SD, and on the impact of these functions and the associated practices on employees and working conditions. Although the concepts of CSR and SD have distinct historical roots, 2 they overlap significantly since they may both be defined broadly as encompassing corporate interactions with society and the ecological environment, channelled through multiple stakeholders. Table 1 provides a selection of academic and practice-based definitions of CSR and SD that illustrates the overlaps between the two concepts.
2 CSR studies are anchored in institutional and welfare economics and in normative ethical studies (Bowen 1953), whereas the later emergence of research on corporate sustainability was influenced by environmentalism as a social movement and by systemic thinking (Bansal and Song 2017).  Hillman and Keim (2001) A broad construct comprised of stakeholder management and social issue management. Hart (1995) A sustainable development strategy, however, also requires that efforts be made to counter the negative effects of economic activity on the environment in the developing countries of the global South.

McWilliams and
Siegel (2001) Actions that appear to further some social good, beyond the interests of the firm and that which is required by law. Russo (2003) An ecologically sustainable industry is a collection of organizations with a commitment to economic and environmental goals, whose members can exist and flourish (either unchanged or in adapted forms) for lengthy periods of time such that the existence and flourishing of other groups of entities can take place at related levels and in related systems.
Flammer (2013) While the original focus of CSR was on "social" responsibility (e.g. paying fair wages to employees, community-based programmes), a recent development is its expansion to cover environmental responsibility (e.g. the reduction of carbon dioxide emissions). Bansal (2005) Three conditions need to be fulfilled in order to achieve sustainable development: environmental integrity, economic prosperity and social equity.
BSR (2003) Achieving commercial success in ways that uphold ethical values and respect people, communities and the natural environment. CSR also means addressing the legal, ethical, commercial and other expectations that society has of businesses, and making decisions that fairly balance the claims of all key stakeholders.
WCED (1987) Development that meets the needs of the present without compromising the ability of future generations to meet their own needs.

ILO (n.d.)
CSR is a way in which enterprises take into consideration the impact of their operations on society and affirm their principles and values both in their own internal methods and processes and in their interaction with other actors. CSR is a voluntary, enterprise-driven process and refers to activities that are considered to go beyond mere compliance with the law.

ISO (2012)
The notion of sustainable development is based on the Rio Declaration on Environment and Development (1992), and it encompasses the concepts of intergenerational equity, social cohesion, human well-being and international responsibility; its implementation is intended to lead to sustainability. The International Organization for Standardization (ISO) uses the definition from the Brundtland Report (WCED 1987): "Sustainable development is development that meets the needs of the present without compromising the ability of future generations to meet their own needs." European Commission The responsibility of enterprises for their impacts on society. To fully meet their social responsibility, companies should have in place a process to integrate social, environmental, ethical, human rights and consumer concerns into their business operations and core strategy in close collaboration with their stakeholders, with the aim of maximizing the creation of shared value for their owners/ shareholders and civil society at large and identifying, preventing and mitigating possible adverse impacts.
European Commission (n.d.) Sustainable development means meeting the needs of the present while ensuring that future generations can meet their own needs.
It has three pillars: economic, environmental and social. To achieve sustainable development, policies in these three areas have to work together and support one another.
By means of a systematic review of articles published in mainstream journals between 1995 and 2014, Bansal and Song (2017) provide evidence of the considerable blurring of boundaries between CSR and SD research, at both the conceptual and measurement levels, and also in relation to the analysis of the antecedents and impacts of CSR and SD. Practitioner-focused journals such as Harvard Business Review, California Management Review or MIT Sloan Management Review refer alternatively, and sometimes interchangeably, to one or the other of the two concepts, suggesting a similar blurring of their definitions in practice. In the case of corporate reporting on non-financial information, it can be seen how these labels are used interchangeably (as in Royal Dutch Shell's Sustainability Report or JP Morgan's Corporate Responsibility Report). This report therefore adopts the umbrella term "CSR/SD" to refer to the functions and professionals in question, and considers articles published in both CSR-focused journals (such as Business & Society) and sustainability-focused journals (such as Organization & Environment), where authors may refer to one or the other notion.
The definition of CSR/SD functions used here is broad and includes any formal organizational unit in charge of CSR-and SD-related domains and/or responsible for more specific social and environmental issues. CSR/SD functions may involve only one actor, and their degree of formalization and structuring may vary widely, as may their prerogatives, levels of empowerment and authority, and positioning within organizations (for example, direct reporting to the board, or dedicated CSR/SD board members). Prior research has found that CSR/SD functions can arise from already established functions, such as public relations or human resource departments, and can contribute to the emergence of new coordination challenges by reshaping functional boundaries (Gond, Igalens, Swaen and El-Akremi 2011*). A similarly broad approach is taken in this report when defining CSR/SD professionals: these are considered to be any organizational actors working within such functions, or whose role and daily activities are at least partly taken up with the handling of CSR/SD issues. As with CSR/SD functions, these roles are subject to potential conflicts of task ownership and authority, since CSR/SD activities may significantly overlap with other occupations and professional jurisdictions (Abbott 1988).
Although the focus in this report is on the literature published in the past 20 years, the emergence of CSR/ SD functions in organizations began considerably earlier. Howard R. Bowen (1953, 155) could already foresee such a development in the early 1950s: […] perhaps, some day, a new official known as the "manager of the department of social responsibility" might be created to coordinate the activities of the various officials who represent various aspects of the public interest.
In fact, when synthetizing the findings of a two-year research project conducted at Harvard Business School (1972-74) that included case studies of how 40 US corporations were dealing with social issues, Ackerman and Bauer (1976) pointed to the emergence of new functional departments and professionals in charge of such issues. They identified some of the tensions inherent in corporate appropriation of this new domain and in the design of relevant functions (Ackerman 1973). Acquier (2016) interprets this first wave of institutionalization of CSR/SD functions as resulting from a dual inter-and intra-organizational movement of rationalization through the development of new management techniques to respond to external pressures on companies to take social and environmental issues into account (for example, life-cycle models of social issue management) and internal efforts to reconcile these techniques with other managerial objectives (such as profit maximization). In this regard, current research on CSR/SD functions is, to a large extent, linking up with discussions from the late 1960s and early 1970s about the management of social issues (Ackerman 1975;Ackerman and Bauer 1976), revitalizing the stream of research usually referred to as being focused on "corporate social responsiveness" (Frederick 1978).
Although this initial approach to CSR/SD functions has largely vanished over time (Acquier, Daudigeos and Valiorgue 2011*), there has been renewed managerial interest in the concepts of CSR and SD. This first occurred in the late 1990s with the development of a "business case" for CSR/SD and was subsequently further advanced by ideas such as "markets for virtue" (Vogel 2005). The maintenance of corporate focus on CSR and/or SD continued despite the global financial crisis of 2007-08. This institutionalization of CSR/ SD functions and the expansion of the role of dedicated CSR/SD professionals has created an upsurge in both academic and managerial interest in understanding the structuring of CSR/SD functions, the work performed by CSR/SD professionals within organizations, and the influence of CSR/SD functions and professionals on organizations' ability to tackle CSR/SD-related challenges. Accordingly, focusing on the institutional dynamics underlying CSR/SD functions, the practice of CSR/SD professionals and their impact within and beyond their organizations is an ideal approach for a literature review that is based on a partnership between academics and practitioners (Sharma and Bansal 2020).

Scope and methodology of the review
Given the profusion of CSR/SD literature in recent years - Lu and Liu (2014) found more than 400 papers published annually between 2009 and 2014 in the Web of Science database alone -and the lack of studies specifically focused on the functional and professional aspects of CSR/SD, a relatively broad scope was adopted for this review. This made it possible to include studies from various subfields and to begin to consolidate the growing yet fragmented body of research on the role of CSR/SD functions and professionals.
Although this report seeks to follow a distinctive approach and takes into account both academic and managerial interests (Sharma and Bansal 2020), its focus on the emergence of CSR/SD functions and professionals does mean that there are areas of overlap with earlier literature reviews, such as those dealing with the literature on CSR/SD embedding and implementation (see, for example, Bertels, Papania and Papania 2010*; Maon, Lindgreen and Swaen 2010*) and reviews of individual-level studies of CSR, that is, of "micro-CSR research" (such as Jones and Rupp 2018). The present review also overlaps with a few higher-level reviews of the entire CSR and SD research fields (such as Aguinis and Glavas 2012;Bansal and Song 2017). Table 2 presents the key takeaways and features of related reviews, indicating whether and how these overlap with the approach taken here.
X Consolidative and integrative review.
Integrative framework summarizing the key organizational stages and cultural phases of CSR implementation.
Overlapping areas: Covers extensively the process of CSR implementation (see the section 2.1.2.3. on "Implementation and/or (de)coupling" in the present review).
Differences: Because of its organizational focus, their review does not take into account the fact that implementing CSR usually involves CSR/ SD functions and professionals.
SLR with the involvement of practitioners.
Development of a framework showing all the practices required to embed sustainability (the "culture wheel").
Overlaps: Focus on "how-to" for individual professionals (competencies) and practice-based research.
Differences: Intra-organizational focus means that the authors neglect, to some extent, the boundary processes that connect actors to their environment.  Psychological micro-foundations of CSR.

SLR.
Behavioural framework covering the individual drivers of CSR engagement and employees' perceptions of and reactions to CSR.

Research agenda.
Overlaps: Partial overlap in relation to how employees can become engaged in CSR.
Differences: Does not consider how CSR/SD functions and professionals operate as an interface between employees and external stakeholders; does not take organizational dynamics into account. Jones and Rupp (2018) Psychological micro-foundations of CSR.
Notion that engagement in CSR is motivated by carebased, self-based and relationship-based concerns.
Consolidative framework to explain employees' reactions to CSR.
Overlaps: Partial overlap in relation to how employees can become engaged in CSR.
Differences: Focus on employees rather than CSR/SD professionals; organizational and extra-organizational dynamics are not taken into account in the analysis.

Authors
Focus and type of review* Key takeaways Areas of overlap with the present review and differences Aguinis and Glavas (2012) CSR field as a whole.

SLR.
Integrative behavioural framework spanning micro, meso and macro levels.
Research agenda focused on micro-CSR and multilevel approaches.
Overlaps: Similar willingness to embrace institutional, organizational and individual-level dynamics.
Differences: Lack of studies of CSR/SD professionals at the time; lack of multilevel frameworks; no consideration of boundaries between an organization and its environment. Bansal and Song (2017) Corporate responsibility and corporate sustainability.
SLR and historical genealogy.
Account of the overlapping approaches of CSR and SD.
Call for studies that differentiate between CSR and SD.
Overlaps: Similar willingness to embrace institutional, organizational and individual-level dynamics.
Differences: Differentiates between CSR and SD (in contrast to the present review's approach); no specific interest in understanding the organizational structuring of CSR/SD functions.
* CSR = Corporate social responsibility; SD = Sustainable Development; SLR = systematic literature review A formal literature review process was adopted for this report, because defining boundaries and offering new categorizations can help to establish emerging fields of research (Gond, Mena and Mosonyi 2020) and to consolidate fragmented areas (Jones and Gatrell 2014;Mosonyi, Empson and Gond 2020). The key steps involved in the production of systematic literature reviews (see, for example, Denyer and Tranfield 2009;Rousseau, Manning and Denyer 2008) have generally been followed. The practical relevance and potential managerial impact of the review have been enhanced by working together with the ILO to better understand the review's key parameters, as recommended by Bansal et al. (2012) and Sharma and Bansal (2020). This was complemented by interviews with a panel of academic experts working in the field of CSR/ SD studies, which helped to make sense of the current challenges faced by researchers and to refine the report's research agenda.

Defining the research context and formulating questions
In line with its Centenary Declaration for the Future of Work, published in 2019, and the blueprint contained therein for the promotion of a more human-centred world of work, the ILO called for a literature review focusing on the role played by CSR/SD functions and the impact of these functions within the workplace (that is, on employees), on the workplace (that is, on organizations' business models) and through the workplace (that is, on external stakeholders). Once this context was defined, the authors of the present review engaged in discussions with ILO staff with a view to formulating a set of research questions that would address the ILO's specific areas of interest. It was determined that the review should focus on whether and how CSR/SD functions and professionals can transform workplaces (and on the relevant political dynamics and considerations) and should strive to advance organizations' knowledge of CSR/SD matters. Three topics were identified as being central to this literature review: Topic #1: Embedding CSR/SD functions and professionals; Topic #2: Managing stakeholder relations through CSR/SD boundary processes; and Topic #3: Shaping interactions with employees and trade unions through CSR/SD boundary processes.

Locating and selecting studies
the following key areas were identified: organization and management theory, CSR, governance studies, organizational behaviour and human resource management. The members of the research team discussed the temporal and disciplinary scope of the review with their ILO contacts. Timewise, it was decided to review articles published from 2000 onwards so as to better understand current thinking and findings. A list of prospective journals falling within the scope of the review was collectively drawn up and subsequently revised to include, notably, some journals from proximate disciplines in management (such as accounting) and in the social sciences (such as industrial relations, development studies and sociology), as well as some non-English-language journals (from France). Reflecting the research team's determination to combine rigour and relevance, the review covers practitioner-focused journals as well as academic journals. 3 A similar set of keywords was used as a filter to identify all the relevant articles in the chosen journals. Slightly different keyword strategies were deployed to identify articles within generalist journals (such as Academy of Management Journal) on the one hand, and specialist journals (such as Journal of Business Ethics) on the other. The title, abstract and keywords of each prospective article were examined by one of the co-authors before a decision was taken on whether to include it in the review or not. The article's relevance to one or more of the three core topics was rated on a three-point scale by the research team, with the review focusing on the most relevant articles (rated 2 or 3) for each topic.

Analysing, synthesizing and reporting
The next step was to identify the key characteristics of each article, including the type (empirical study, literature review and so on), methodological orientation (qualitative, quantitative or mixed-methods), theoretical focus (for example, reliance on institutional theory) and various other elements that could help address the three core topics. As a result of this process, the level of analysis (individual, organizational, and institutional field) was also identified -a useful characteristic to record. The distinction made by Gond and Moser (2021) between different micro levels proved valuable in distinguishing CSR/SD functional dynamics from the work of CSR/SD professionals. However, a conceptual framework was required to make sense of the organizational and institutional dynamics that underlie CSR/SD functions and their institutionalization and to organize the review more effectively. The following subsection describes how a framework concentrating on "boundary processes" was developed to support the analysis.

Framing the review: An overarching framework of boundary processes
Central to the development of CSR/SD functions within organizations and the concomitant rise of new professional actors responsible for CSR/SD matters are dynamics that connect forces operating outside organizations (such as institutional, market and legal forces) to intra-organizational functional processes and to individual practices and reactions (Acquier 2016). Accordingly, the present review focuses on the environment-organization interface and relies on the notion of "organization as open polity" (Weber and Waeger 2017, 890), which emphasizes how the external and internal political environments of organizations are connected in ways that give rise to new occupational roles and organizational units and have the potential to transform organizational as well as environmental dynamics (Waeger and Weber 2019). The concept of a "boundary process" -defined as a process "through which an organizational polity interacts with and is thus coupled to its external environment" (Weber and Waeger 2017, 889) -is used to explore how an organization interacts with its political environment in ways that give rise to CSR/SD functions and that make these activities and the staff engaged in them potentially influential and impactful.
This concept is extended here to capture a broader variety of mechanisms whereby CSR/SD functions are shaped by and shape organizations' capacity to deliver on the CSR/SD agenda. The ultimate impact of CSR/ SD functions, policies and programmes on organizational outcomes and stakeholders' welfare depends on boundary processes, where political forces from the environment interact with organizational and individual dynamics. By mapping these boundary processes, this report sought to identify the channels through which norms and standards, such as the ones developed by the ILO, influence whether and how CSR/SD functions can achieve their ultimate goals. Figure 1 provides an overview of all the processes identified through the literature review. In Chapter 2 of the report, these processes are outlined and the main findings for each one presented. The tensions and debates emerging from the juxtaposition of all these studies are also discussed. Lastly, Chapter 3 provides a critical synthesis of the current state of knowledge.

Developing a research agenda: Interviewing experts in the CSR/SD field
The research team interviewed a panel of six academic experts who have contributed to the literature on CSR/SD functions and professionals. These experts were selected on the basis of their publications, with each having published at least two articles in top-tier journals from among those identified during the review.
Semi-structured interviews were conducted with each expert, the average duration of the interviews being 54 minutes. The interview grid was designed to stimulate a conversation with purpose and broadly covered the three core topics mentioned in subsection 1.3.1. The questions were tailored to each interviewee's publications in order to ensure that their specific expertise could be fully tapped. The main ideas arising from each interview were then synthetized with reference to the three core topics. This empirical material was used, together with the results of the review, to inform analysis of the current challenges in this research field and to develop the research agenda proposed in Chapter 3.

X 2 How CSR/SD functions emerge and interact with stakeholders: Reviewing the literature from the perspective of boundary processes
This part of the report describes the full range of boundary processes that were identified through the review. First of all, the findings on the boundary processes that underlie the embedding of CSR/SD functions and professionals within organizations are presented (section 2.1). The report then "zooms in" on how CSR/ SD boundary processes shape organizations' capacity to interact with their stakeholders (section 2.2) and on how these processes influence organizations' interactions with employees and trade unions (section 2.3).

Embedding CSR/SD functions and professionals
The unprecedented adoption of CSR discourse and practice at the corporate level over the past 20 years has been accompanied by the emergence of new organizational functions devoted to the management of CSR/SD and of new professionals operating within these functions with dedicated occupational roles, such as "Chief Sustainability Officer", "Climate Change Manager" or "CSR Manager". In analysing these trends the research team was guided by three overarching questions: • How have CSR/SD functions been institutionalized and professionalized within organizations?
• Who are the new CSR/SD professionals performing these functions?
• How do professionals with CSR/SD functions operate?
The review undertaken for this report suggests that these trends have been the subject of highly disparate studies, which have relied on a variety of theoretical approaches (such as institutional theory and sense-making theory) and have dealt with specific aspects of the embedding of CSR/SD functions (for example, the structuring of CSR/SD functions, or the impact of newly created occupational roles) and of the development of CSR/SD-related professional roles (for instance, the struggles faced by CSR/SD managers, or the competencies required to support the implementation of CSR/SD policies). Adapting the concept of a boundary process (see Figure 1), this report considers CSR/SD functions and professionals as boundary-spanning entities that connect extra-and intra-organizational forces, and analyses how they may draw on or shift boundaries within organizations -for instance, between the departments dedicated to public relations, human resource management and communications -and between organizational actors. The boundary processes identified through the review are classified into three broad categories: (a) processes triggered at the macro level by an organization's environment (such as regulation), which also include processes that play out at the environment-organization interface (such as professionalization); (b) processes involved in the structuring and deployment of CSR/SD functions (such as structuration and translation); and, lastly, (c) processes occurring at the level of individual CSR/SD professionals (such as institutional work and self-work). Table 3 provides definitions and illustrations of these three categories of boundary processes, which, in the following subsections, are presented in detail from the macro to the micro level of analysis.
Government-corporate relations shaping CSR/SD policies [Gond, Kang and Moon 2011] Standards as enabling a "licence to critique" CSR/SD [Christensen, Morsing and Thyssen 2018] Commodification Creation of CSR/SD markets of consultancy products and services that make organizations susceptible to CSR/SD practices and tools and enable such practices and tools to achieve wide diffusion.
Rationalization of CSR/SD through a business case logic [Shamir 2005] CSR/SD consultants as "issue translators", "boundary negotiators" and "enactors of responsive regulations" [Brès and Gond 2014] Consultants as engineers and diffusers of CSR/SD tools [Gond and Brès 2020] Professionalization Emergence of CSR/SD professionals within and across organizations who share relatively similar aims, knowledge and practices and who form associations and/or communities of practice.

Development of CSR/SD professional associations [Weller 2017]
Responsible careers of CSR/SD professionals [Tams and Marshall 2011] Emergence of CSR/SD-focused communities of practice [Weller 2020] Constitution of a fragile CSR/SD knowledge base [Brès et al. 2019] Structuring of CSR/SD functions Structuration Set of top-down, middle-ground and bottom-up processes that lead to the creation and positioning of formal CSR/SD functions within organizations.
Creation of Chief Sustainability Officer post [Strand 2014] Issues of coordination between different functions [Egan and Tweedie 2018] Situated interactions between individuals that operate as structuring forces in organizations [Soderstrom and Weber 2020] Translation Interpretive processes whereby external demands for CSR/SD are assimilated by an organization as actors there make sense of CSR/SD functions and activities.
Typologies and dimensions of CSR/SD sense-making strategies [Cramer, Van Der Heijden and Jonker 2006] and organizational processes [Basu and Palazzo 2008] Framing, localizing and normalizing stages of translation [Wright and Nyberg 2017] Implementation and decoupling Organizational and cultural processes whereby CSR/SD discourses are turned into actual policies and outcomes within organizations in ways that are more or less connected to external demands or pressures (from an organization's institutional environment).
Models showing the stages of CSR/SD implementation [Maon, Lindgreen and Swaen 2010] Decoupling of CSR/SD policies from practices and outcomes [Graafland and Smid 2019] Loose coupling of CSR/SD with core activities that emerges as managers "muddle through" with CSR/SD considerations [Crilly, Zollo and Hansen 2012] Decoupling of CSR/SD policies and practices as "aspirational talk" [Christensen, Morsing and Thyssen 2013], "helpful hypocrisy" [Glozer and Morsing 2020] or a transitory phase [Haack, Martignoni and Schoeneborn 2020] CSR/SD Professional work Institutional work Purposive and practical actions of individuals within organizations that help to create, disrupt or maintain CSR/SD functions.
Relational work and "issue-selling" to promote CSR/SD [Wickert and de Bakker 2018] Strategifying work to embed CSR/SD within an organization's strategies [Gond, Cabantous and Krikorian 2018] Justification work to explain changes in CSR/SD strategy [Demers and Gond 2020] Political work (coercive or deliberative) through the embedding of CSR/SD functions [Acosta, Acquier and Gond 2019]

Macro-level boundary processes: Embedding of CSR/SD functions and professionals
The emergence of CSR/SD functions and professionals can be linked to "outside-in" boundary processes, whereby the institutional and market environment of an organization leads to its adoption of a formal CSR/ SD organizational structure. Three dynamics are of particular relevance in this respect: regulation, commodification and professionalization.

Regulation
Although CSR/SD has traditionally been defined by North American scholars as a set of voluntary corporate actions that go beyond legal requirements ( Concepts such as "corporate citizenship" (Matten and Crane 2005*) or "political CSR" (Scherer and Palazzo 2011*) have been coined precisely to capture the processes whereby private actors have taken over the role of governments, through their delivery of services traditionally offered by a government or through the participation of MNCs in deliberative democratic forums and emerging global governance platforms that seek to tackle social and/or environmental issues, such as human rights or global warming. Such a "retreat of the State" to the advantage of private actors suggests that the development of CSR/SD functions and professionals is related to macro-trends such as globalization (Scherer, Palazzo and Matten 2014*).
However, more recent studies of political CSR (Kourula et al. 2019*;Scherer et al. 2016*) have diagnosed a "return of the State" in the CSR realm, with governments playing the role of a strategic actor in the CSR/SD space that can intervene purposively and in various ways to mobilize and potentially capture private initiatives (Schrempf-Stirling 2018*), or even fully regulate them through a traditional command-and-control process. Schneider and Scherer (2019*) have clarified the multiple mechanisms underlying such governmental CSR interventions, which combine the design of incentives with the provision of expertise, knowledge and legitimacy to key actors in order to support the corporate adoption of CSR/SD. Particular attention has been paid to India, as its Government was one of the first -together with that of Mauritius in 2007 -to pass a law (the Companies Act 2013, Section 135) making it mandatory for companies to spend a certain amount of their profit on CSR/SD activities, in a move that radically calls into question the voluntary nature of CSR/SD (Gatti et al. 2019*). A subfield of study dedicated to "CSR and government" (Knudsen and Moon 2017) investigates the various mechanisms of governmental CSR intervention within and beyond national borders. Focusing on France, Giamporcaro, Gond and O'Sullivan (2020*) identify a series of hard, softer and indirect CSR/SD governmental interventions in the realm of responsible investment that have catalysed the growth of the asset management industry over the past 15 years. The regulation of CSR/SD information disclosure by corporations (New Economic Regulations Act), together with the creation of state-owned pension funds with a CSR/SD focus (Fonds de Réserve pour les Retraites, Établissement de Retraite Additionnelle de la Fonction Publique) and stronger regulation of institutional investors, has fostered the expansion of CSR/ SD functions at asset management companies and in publicly held corporations in France. 4 However, the CSR/SD regulatory sphere is far from being dominated solely by governments. A growing number of studies have analysed the role of private governance and non-governmental standards as key drivers of CSR/SD adoption, even though their impact may be limited by their redundancies and overlapping nature. Gilbert, Rasche and Waddock (2011) have pointed to a proliferation of "international accountability standards" that constitute a new regulatory infrastructure for CSR/SD at the global level and encompass standards issued by organizations as diverse as the Global Reporting Initiative, the ILO or the International Organization for Standardization (ISO). Researchers have investigated the complex regulative effects inherent in this multiplication of CSR/SD standards generally (Wijen 2014*) and in specific industries (Reinecke, Manning and von Hagen 2012*) and have studied the development and impact of standards such as ISO 26000 (Helms, Oliver and Webb 2012*). Although earlier discussions focused on whether these standards should be strict or loose, more recent studies have concentrated on how standards can enable CSR/SD professionals to act as voices of dissent within organizations, fostering "a culture devoted to (self)-reflection and discussion about basic assumptions, shared values, ideals and practices" (Christensen, Morsing and Thyssen 2013*) that can advance the achievement of CSR/SD objectives.
In general, there has been a continuous expansion of public and private regulations in the CSR/SD realm, which may be ascribed to trends such as globalization and pressures from civil society and investors and is driven by national governments and supranational institutions, as well as by private organizations and NGOs. This push for greater organizational accountability has fostered the development of dedicated CSR/ SD functions in organizations and promoted CSR/SD professionalization by creating a common set of norms that make up the knowledge base of CSR/SD experts. Researchers have only just begun to explore how these multiple layers of regulation interact and shape the behaviour of organizations.

Commodification
Regulatory developments do not take place in a market vacuum: several studies have noted how a new professional category of CSR/SD consultants have acted as an "outside-in" pressure, leading to the embedding of CSR/SD functions in organizations. Such consultants have actively contributed to a process of "commodification" (Brès and Gond 2014*; Gond and Brès 2020*; Shamir 2005*), whereby social and environmental issues are turned into commodities and the size of the "market for virtue" is continuously expanded (Vogel 2005). There are several aspects to this process. First, by "furthering awareness and advancing CSR standards" (Molina 2008*, 31), consultants contribute to the boundary process of regulation, whereby hard and soft regulations are translated into CSR/SD concepts and practices for businesses, reducing the ambiguity inherent in such regulations (Edelman 1990;1992). For example, Brès and Gond (2014*) show that "responsible purchasing" practices were diffused among leading corporations in Quebec through a forum designed by consultants to exchange best practices. 5 Through commodification, consultants play an active role in enacting "responsive regulations" by mobilizing, promoting and translating CSR/SD regulations.
Second, earlier studies note how consultants have reshaped the meaning of CSR/SD discourses in ways that are directly compatible with business interests (Gond and Brès 2020*; Molina 2008*; Shamir 2005*), especially by advancing a "business case" rhetoric focused on the benefits of adopting CSR/SD (Skouloudis 4 The New Economic Regulations Act (specifically article 116) requires listed companies to disclose in their annual report how they address the environmental and social impact of their activities. The regulation has been in place since 2001 and was supplemented in 2002 by a list of 19 topics to facilitate company reporting. Article 173 of the Energy Transition for Green Growth Act came into force on 1 January 2016; for the first time, institutional investors in France were required to report on their environmental, social and governance approaches on the basis of the "comply or explain" principle. Beyond France, numerous hard and soft legal developments have fostered the development of CSR/SD functions, such as the King Reports on Corporate Governance in South Africa. Knudsen and Moon (2017) provide a review of such legal frameworks. 5 Interestingly, a key figure in this process was "an ex-employee of the ILO and the UN [who] acted as an active translator of international laws and standards into practices that fit the 'practical expectations' of purchasing managers" Gond 2014*, 1359).
and Evangelinos 2014*). Shamir (2005*, 240), for instance, in his ethnographic study of how actors from "market-based NGOs" imported CSR into Israel, observed that consultants: […] typically invok[ed] a nonconfrontational rhetoric, typically abstain[ed] from highlighting potential tensions between profit-making imperatives and social commitments, and, on ground level, typically invoke[ed] a business-grounded rationale for the need to develop CSR.
However, other studies of CSR/SD consultants suggest that this focus on a business case rhetoric is only one of the many narratives that such consultants are juggling in order to make sense of their activities, as they must also balance the search for profit with more socially and environmentally meaningful practices (Ghadiri, Gond and Brès 2015 Third, a recent study has shown how CSR/SD consultants construct the material infrastructure underlying the development of corporate CSR/SD functions and practices through their active role in designing and adapting management tools, such as life-cycle assessment methods, auditing tools or Deming's inspired "Plan-Do-Check-Act" frameworks, to the specific needs of CSR/SD professionals (Gond and Brès 2020*). These tools are often a key resource for smaller CSR/SD functional units, which, without them, would not be able to devise or implement CSR/SD strategies. There is clearly an exchange of personnel and, hence, expertise between consultancies and CSR/SD units within organizations.
To sum up, the boundary process of commodification, which is largely driven by consultants, can augment, replace, anticipate and sidestep the influence of regulations. Consultants help to operationalize and rationalize new CSR/SD-focused business practices and tools that are then adopted by CSR/SD professionals; they may themselves also shape such regulations, which are usually aligned with their business interests (Brès and Gond 2014*). Through commodification, they continuously extend the reach of CSR/SD ideas, tools and practices to new organizations.

Professionalization
The emergence of CSR/SD consultants in the "market for virtue" reflects a more general boundary process in the institutional environment, namely the professionalization of CSR/SD activities and the endeavour to turn CSR/SD into a "profession" (Brès et al. 2019*). A profession can be loosely defined, following Abbott (1988), as an occupational group leveraging abstract expertise and proprietary techniques to establish exclusive control over an area of work (jurisdiction) within and across organizations. Professionalization involves translating these resources into social and economic rewards, thereby drawing (and redrawing) boundaries within organizations to create dedicated functional units, while establishing new professional organizations at the institutional field level. Institutional analysis has shown that the emergence of new occupational or professional groups within organizations is driven by regulation (Edelman 1990;1992) and by the efforts made by social movements to normalize new organizational responsibilities (Augustine 2021*).
The professionalization of CSR/SD is indicative of the massive impact that corporate transition towards sustainability is expected to have on the workforce (ILO 2019b). It manifests itself not only in the emergence of in-house CSR/SD managers, dedicated CSR/SD consultants and environmental, social and governance analysts in financial markets, but also in the networking of nascent CSR/SD professionals that occurs through professional associations, which in turn foster the establishment of "communities of practice". This networking usually involves CSR/SD professionals from various organizations within the same industry, 6 CSR/SD consultancy is provided by various actors. These include non-profit organizations (such as Forum for the Future and Business for Social Responsibility), the "Big Four" accounting and consulting firms (notably PwC and Deloitte), general strategy consulting firms (such as McKinsey and Accenture) and smaller dedicated consultancies (such as DNV GL and Corporate Citizenship). The type of CSR/ SD consultancy -non-profit organization, "market-based non-profit organization" (Shamir 2005*), "Big Four" firm, specialized consultancy -does seem to be relevant, with different types of consultants bringing different resources to corporate actors and CSR/SD functions (for example, the Big Four can offer the business legitimacy of well-established techniques, while social and environmental legitimacy is typically provided by non-profit organizations) (Brès and Gond 2014*).
but it may also occur across multiple industries and markets. Weller (2017*) looks at the emergence of CSR/SD professional associations in the United States -in particular the creation of Business for Social Responsibility in 1992 and the Corporate Responsibility Officer Association in 2005, which subsequently became the Corporate Responsibility Association -and discusses the competing role of ethics and compliance as a related professional domain (albeit one that is more concerned with the development of regulations). Associations of CSR/SD professionals have also mushroomed in Europe in recent years. 7 Efforts have been made by industry associations to codify shared knowledge and establish standards. For example, the Association for the Advancement of Sustainability in Higher Education, which was set up in 2005 and is focused on North America, has designed a voluntary sustainability standard that is currently used by sustainability professionals at over 1,000 colleges and universities to report on their work. Members of the United Nations-supported Principles for Responsible Management Education initiative, launched in 2007, sign up to six principles guiding management-related higher education. Moreover, professional associations such as the International Society of Sustainability Professionals and the Institute of Corporate Responsibility and Sustainability have developed certification programmes based on a set of qualifications that reflect a shared body of knowledge and expertise among those working in CSR/SD. Weller (2017*; 2020*) has found that CSR/SD associations facilitate the formation of CSR/SD-focused communities of practice, that is, "groups of people who share a concern or a passion for something they do and learn how to do it better as they interact regularly" (Wenger 2011, 1). This emphasis on shared passion is corroborated by Augustine (2021*), who found that sustainability managers saw themselves as part of a broader "movement for sustainability". Such managers convene regularly through online forums and in-person conferences (Wright, Nyberg and Grant 2012*). Communities of practice operate within and across organizational boundaries and are characterized by the production and practical application of shared knowledge by their members (Brown and Duguid 1991), who may be said to possess a common identity as a group (Wenger 1998). CSR/SD-focused communities of practice directly foster the development of CSR/ SD functional units in organizations by equipping their members with a common knowledge, identity and purpose. In that sense they constitute a key boundary process connecting the CSR/SD activities of different organizations. Studies of the career paths of CSR/SD professionals indicate that most of them are driven by a quest for societal and ecological purpose across organizations. Other important factors are their alertness to changes in professional CSR/SD landscapes, which help them to orient their careers (Tams and Marshall 2011*), and their perception of the opportunity that such positions offer them to promote their take on "sustainability" (Koistinen et al. 2020*).
Current studies suggest that CSR/SD professionalization remains fragmented and that CSR/SD professions are regarded as still being in a nascent state (Brès et al. 2019*). For example, in the context of South Korea, Shin et al. (2021*) delineate four distinct and competing discourses of CSR professionalism held by CSR/SD practitioners that contribute both to legitimizing, and to fragmenting and undermining, the development of this institutional field. Furthermore, a recent dialogue between scholars from the CSR/SD field and scholars from the sociology of professions highlighted a number of peculiar features of CSR/SD professions, such as the lack of a clear knowledge base, relatively vague boundaries and recurrent questioning of the social and ecological impact of such work. It is debatable whether CSR/SD can ever become fully professionalized: indeed, it has been argued that CSR/SD could be a failed professional project, even though CSR/SD roles have become part of organizations' structures (Brès et al. 2019*;Risi and Wickert 2017*). This conclusion was reached by Risi and Wickert (2017*) on the basis of 85 interviews with CSR/SD managers from German and Swiss corporations. They found that, over time, these managers were not able to advance their professionalization projects, but rather were marginalized to the organizational periphery. Accordingly, the institutionalization of CSR/SD functions and the professionalization of CSR/SD roles could follow diverging trajectories. Nevertheless, most scholars agree that CSR/SD roles have undergone, and are still undergoing, many of the typical processes involved in professionalization. 7 The European Association of Sustainability Professionals brings together national CSR/SD associations to exchange practices, organize transnational initiatives and develop joint positions on policies. Its members include the Collège des Directeurs du Développement Durable, which was set up in 2007 to bring together SD directors across France, and the Institute of Corporate Responsibility and Sustainability, which was originally founded in 1984 to support CSR/SD practitioners in the United Kingdom.
Although earlier research has highlighted the conceptual and empirical potential of focusing on CSR/SD professionalization, only a few studies have taken CSR/SD professions as the core unit of analysis. Moreover, knowledge of the many existing CSR/SD associations and communities of practice remains scarce. Future analysis could concentrate on the global-local tensions inherent in CSR/SD professionalization. Looking at settings such as "field-configuring events" (for instance, conferences or standard-setting discussions) and using tools such as network analysis could help to trace the development of various communities of CSR/ SD professionals more accurately and to shed further light on how commodification and professionalization interact in the emergence of CSR/SD functions within organizations.

Meso-level boundary processes: Structuring CSR/SD functions
The institutional field-level boundary processes supporting the embedding of CSR/SD functions significantly overlap with the boundary processes at the organizational level that underlie the deployment of these functions. They include structuration, translation and sense-making, and implementation and decoupling.

Structuration
Several loosely connected streams of research investigate the structuring of CSR/SD functions, following a top-down, middle-ground or bottom-up approach. The top-down approach is consistent with earlier analyses of corporate social responsiveness, which insisted on the need for executives to commit to the development of corporate capacities for dealing with social issues (Ackerman and Bauer 1976). Starting from the "upper echelons", the first set of studies focused exclusively on the development of formal senior managerial positions related to CSR, exploring the various labels and job descriptions used to analyse what today are most commonly referred to as chief sustainability officer (CSO) positions (Strand 2013*;. This research has led to studies of senior management teams that usually rely on upper echelons theory -an approach that explores whether and how top managers influence their organizations (Hambrick and Mason 1984) -to determine whether and how CSR/SD positions or their holders' beliefs and characteristics shape subsequent CSR/SD performance. Fu, Tang and Chen (2018*) found that the existence of a CSO position reduces subsequent "corporate social irresponsibility" more than it enhances CSR; Wiengarten, Lo and Lam (2017*) found the effect of having a CSO on corporate social performance to be significant and positive only when the position was filled by someone from inside the organization, the effect being stronger for female CSOs. The findings of Peters, Romi and Sanchez (2019*) are more nuanced, suggesting that the positive impact of a CSO on subsequent CSR-related performance is relatively long-term (three years) and conditional on earlier CSR performance and on the chief executive officer's (CEO) knowledge of sustainability. Kanashiro and Rivera (2019*) found the effect of a CSO position on subsequent corporate environmental performance to be stronger in industries with a high level of environmental regulation.
Other studies based on upper echelons theory have explored the influence of CEOs' ideological orientations, belief in CSR/SD and personality traits on their organization's CSR profile. For instance, drawing on their analysis of the ideological disposition of 249 CEOs, Chin, Hambrick and Trevino (2013*, 197) found that, in contrast to conservative CEOs, "liberal CEOs exhibit greater advances in CSR", that "the influence of CEOs' political liberalism on CSR is amplified when they have more power" and that "liberal CEOs' CSR initiatives are less contingent on recent performance than are those of conservative CEOs". Exploring CEOs' beliefs and business ideologies, Hafenbrädl andWaeger (2017, 1601) came to the conclusion that: […] even though they believe in the business case for CSR, managers who hold a fair market ideology [that is, a tendency to justify and idealize the market economy system] will not readily engage in CSR because they experience weaker emotional reactions to ethical problems than managers who do not hold a fair market ideology.
On the whole, these results suggest that the organizational design of CSR/SD positions at senior management levels and CEOs' beliefs may positively impact CSR/SD performance, at least when efforts are substantive rather than purely symbolic, such as in the recruitment of a CSO who is knowledgeable about CSR/ SD and the corporate context, or the presence of senior executives who espouse progressive ideologies. However, the analysis by Mun and Jung (2018) of how gender diversity is promoted at Japanese companies suggests that top-level efforts to promote CSR/SD issues may not trickle down and may end up benefiting only the senior ranks of the organization.
The necessary, if not sufficient, motivation provided by senior management's seal of approval when it comes to deploying and properly supporting CSR/SD functions is consistent with the findings of qualitative studies that have looked at how CSR/SD managers go about developing and expanding their prerogatives within organizations. For example, the study by Gond, Cabantous and Krikorian (2018*) of how CSR/SD was successfully embedded in a British energy company by a freshly appointed Head of CSR/SD highlights this manager's continuous efforts to secure the support of key board members. The operational deployment of CSR/ SD is usually left in the hands of middle managers who do not always have sufficient authority or discretion to implement CSR/SD strategies. Risi and Wickert (2017*, 622), for instance, found that only 10 per cent of the 85 German and Swiss CSR/SD managers whom they interviewed reported directly to the CEO, chairperson and/or the board of directors, and only a few of them had substantive or increasing budgets.
The lack of structural empowerment of CSR/SD managers has been noted recurrently, as has the struggle of such managers to create and/or protect their occupational jurisdictions (see, for example, Acquier 2016*; Augustine 2021*; Gond, Cabantous and Krikorian 2018*; Risi and Wickert 2017*). Studies of the interface between CSR/SD and other functions in an organization have identified coordination problems. In the case of the interface with accounting, the qualitative analysis by Egan andTweedie (2018*, 1749) of the implementation of CSR/SD reporting at a large Australian corporation revealed that, "while accountants adapted well to early changes aligned to cost efficiency, they struggled to engage with more creative sustainability improvements". Similar difficulties have been reported in earlier studies of how accountants can contribute to their organization's CSR/SD agenda (Çalişkan 2014*) -an endeavour that may require specific "hybrid" competencies (Caron and Fortin 2014*). Gond, Igalens, Swaen and El Akremi (2011*) identified "territory battles" over ownership of employee-focused CSR/SD practices between the human resource and CSR/SD areas of a number of MNCs headquartered in France (see also Acosta, Acquier and Gond 2019*). Similar tensions between distinct categories of professionals have been identified for the related domains of ethics and compliance (see, for instance, Ben Khaled and Gond 2020; Chandler 2014). Acquier (2016*) highlighted the contrasting trajectories of CSR/SD functions at two different organizations: one an MNC from the sports industry that successfully embedded CSR/SD in its multiple subsidiaries across different regions; the other a European energy group in which the CSR/SD area became marginalized because of its focus on external reporting and its lack of capacity to inform strategy. Studies following a middle-ground approach indicate that, even when a CSR/SD unit has been established, it may struggle to expand its jurisdiction across the functional areas necessary for CSR/SD changes (Augustine 2021*) and so prove unable to influence other parts of the organization.
There has been less empirical research on the bottom-up boundary processes underlying the emergence and development of CSR/SR functions (Girschik, Svystunova and Lysova 2020), even though most studies focusing on middle management emphasize the importance of securing the support of employees for CSR/SD (Gond, Cabantous and Krikorian 2018*; Haugh and Talwar 2010*). However, in a very robust qualitative study conducted over 18 months at a large biomedical company in the United States, Soderstrom and Weber (2020*) identified bottom-up dynamics that explained why some sustainability concerns were assimilated by organizational structures while others were not. Central to such assimilation is the quality of interactions among the members of an organization; for example, when individuals put forward suggestions, they receive a response and are able to adjust the initial proposal accordingly (Weick 1979). Such interactions can trigger repeated cycles of exchanges, eventually leading to the formal structuring of an organization's sustainability efforts. According to Soderstrom and Weber (2020*), successful interactions generated attention, motivation, knowledge, relationships and resources that could feed into emergent organizational structures.
The top-down, middle-ground and bottom-up approaches to CSR/SD organizational structuring complement one another with regard to clarifying how distinct elements of the CSR/SD function are structured over time, in ways that may enhance or diminish the impact of the function and that therefore impinge on an organization's ability to make progress on CSR/SD issues. Researchers have only just started to assess whether the simultaneous presence or absence of different CSR/SD functional components influences the overall implementation of CSR/SD (Halme et al. 2020*). Future studies could look more closely at how these boundary processes interact in ways that explain whether and how the CSR/SD function can influence the rest of the organization.

Translating and sense-making
Another set of studies on the structuring of CSR/SD functions within organizations focused not so much on the structural components of such functions as on how CSR/SD actors mobilize these structures to interpret and make sense of what CSR/SD issues mean for the organization. Central to these studies are the concepts of translation, whereby actors try to understand CSR/SD issues and practices and negotiate and (re)define their meaning in order to adapt them to their organizational situation (Callon 1986;Czarniawska and Joerges 1996), and sense-making, which reflects the notion that problems do not appear to practitioners as something "given" but rather they "must be constructed from the materials of problematic situations that are puzzling, troubling and uncertain" (Weick 1995, 9).
Basu and Palazzo (2008*) identified the key dimensions underlying the boundary process of CSR/SD managerial sense-making by looking at the discursive, cognitive and behavioural aspects of how organizations engage with CSR/SD issues. According to the study, the alignment or misalignment of these three facets explains how managers think about and discuss CSR/SD with the organization's stakeholders. Other studies of CSR/SD managers at organizations have analysed the crucial role of sense-making as one of the main interpretative mechanisms through which employees respond to their organization's implementation of CSR/SD (Aguinis and Glavas 2019*). Monitoring the implementation of CSR/SD strategies at 18 global MNCs, Cramer, Van Der Heijden and Jonker (2006*) highlighted the diversity of organizations' CSR/SD-related sense-making, with these MNCs opting variously to translate CSR/SD into well-defined goals (pragmatic sense-making), management control systems (procedural sense-making), strategy (policy sense-making), values and beliefs (value sense-making) or upholding the organization's reputation (external sense-making).
In one of the most comprehensive empirical studies of the translation of CSR/SD issues in organizations, Wright and Nyberg (2017*) looked at how five major Australian corporations, including CSR/SD teams and other functions, internally translated climate change issues over a ten-year period . Drawing on 70 interviews and multiple sources of secondary data, they developed a staged-model framework that explains how corporations redefine the meaning of climate change to gain the support of stakeholders (referred to as "framing"), make these new definitions locally relevant by identifying compromises between competing goals ("localizing") and redesign business practices and routines accordingly ("normalizing"). The authors found that the key tensions associated with action on climate change were neutralized as a result of this process, which ultimately led to only minor and incremental changes and to business-as-usual management processes and routines that were unlikely to address the major challenge of climate change in an effective manner. Augustine (2021*) similarly found that, when developing standards to regulate sustainability activities at colleges and universities, managers chose not to take into account many potential sustainability issues in these standards (and, therefore, in their day-to-day work), especially those concerning the social elements of the CSR/SD agenda, because of their wish to be seen as "neutral", "apolitical" actors. These studies therefore suggests that CSR/SD issues may become "lost in translation" as they are translated by CSR/SD departments and their resident "experts".
In sum, studies with a focus on translation highlight the various ways in which executives and managers can make sense of CSR/SD issues when integrating them into the work of their organization. They also point to the risk of losing the capacity to address such issues when a CSR/SD function is formally established. Although translation is an essential boundary process when it comes to making CSR/SD issues understandable and "manageable" within organizations, researchers working on CSR/SD have only recently begun to draw inspiration from the rich tradition of translation studies in management and social sciences (Spyridonidis et al. 2016).

Implementation and/or (de)coupling
A third organizational-level boundary process that explains the structuring of CSR/SD functions relates to analyses focusing on the actual implementation of CSR/SD programmes, policies and strategies within organizations. In this area, researchers have either characterized the stage of CSR/SD implementation following a prescriptive, managerial approach, usually inspired by organizational change, development or learning (CSR/SD implementation studies), or have adopted a neo-institutional theory to document and discuss whether and how CSR/SD discourses are actually turned into practices (CSR/SD (de)coupling studies). The first managerial type of study is exemplified by Zadek's (2004*) analysis of how Nike moved from being known as an irresponsible company because of child labour issues in its supply-chain in the early 1990s to a CSR/SD role-model in the early 2000s. To capture these CSR/SD learning dynamics, Zadek proposes a five-stage model of organizational growth centred on CSR/SD maturity, starting with defensive approaches to CSR/SD, then moving on to compliant, managerial, strategic and, finally, civic approaches. In a similar fashion, Maon et al. (2009*) developed a grounded model of CSR/SD organizational implementation, relying on Lewin's change model. In a broader consolidative review of such implementation studies, Maon et al. (2010*) identified, compared and integrated all the various stage models of CSR/SD implementation to specify an overarching three-stage model of CSR/SD-related cultural change: first "cultural reluctance", then "cultural grasp" and finally "cultural embedment". Although managerially useful because of their prescriptive nature, few empirical studies have actually tested or evaluated such models systematically and/ or across numerous corporate cases.
Another stream of research dedicated to implementation relies on institutional theory to explore whether and how CSR/SD policies are actually coupled to organizational practices (i.e. is much of CSR/SD just window-dressing or green-washing?). First, scholars leveraged Meyer and Rowan's (1977) seminal observation that organizations adopt organizational structures from their environment in order to enhance their legitimacy and comply with dominant "rationalized myths", even though there is no economic or technical rationale for doing so (and they never intend to actually change their practices). Research has since focused on more recent distinctions between various forms of decoupling, hybridization, loose-coupling or recoupling. Confirming the cursory observation that CSR/SD outcomes and practices are often at odds with organizational CSR/SD discourses and policies, multiple studies have explored the dynamics of CSR/SD discourse and implementation. Crilly, Zollo and Hansen (2012*) refined the coupling-decoupling continuum by exploring the managerial motivations of decoupling through a configurational analysis of 17 MNCs based on 359 interviews with internal and external actors; they found that managers' decoupling of CSR/SD responses depended on the presence of information asymmetry and external stakeholder consensus, as well as internal managerial consensus and the organizations' interests. Although Crilly et al. (2012*) identified that decoupling of CSR/SD policies can reflect an actual managerial motivation ("faking it"), in the presence of competing stakeholder expectations and a lack of managerial consensus about CSR/SD, corporate responses are emergent and reflect the complexity of navigating internal and external expectations ("muddling through"). Haack et al. (2012*) showed how the adoption of the Equator Principles by banks created multiple narratives within the organizations, even if the principles were decoupled from substantive practices at first. As a result, this initial decoupling could eventually lead to stronger coupling between discourse and practices (Haack et al. 2012*). In a later study, Haack et al. (2020*) generalized their analysis through modelling, suggesting that decoupling in the CSR/SD domain might be a more "transitory" phenomenon, as organizations engage in cycles of coupling and decoupling.
Another stream of studies regard the decoupling of CSR/SD discourse and practice as a form of "organizational hypocrisy" (Brunsson 2003), which can be a necessary evil and is not necessarily a bad thing. Christensen, Thyssen and Morsing (2013*, 373) regard the gap between ambitious CSR/SD discourse and practice as "aspirational talk" needed to leverage further resources for organizational change; as "differences between words and actions on the CSR arena may in fact be vital in order to move the field forward towards higher goals and superior standards", they argue that discourse can inspire practice change. Glozer and Morsing (2020*, 363) argue further that such hypocrisy could be "helpful" as it triggers irony, which can mobilize audiences to "critically reflect on complex ambiguities of CSR in non-moralizing ways". In a related study, Augustine (2021*) found that, although sustainability managers cut task areas from their formal jurisdiction that they thought would make them appear values-based or like they had "an agenda", they continued to pursue some of this work behind the scenes, which shows how having an individual in a CSR/SD role can help advance sustainability initiatives, even if the individual does so in a concealed fashion.
More recent studies have re-examined CSR/SD decoupling in light of Bromley and Powell's (2012) distinction between "policy-practice" decoupling (the gap between organizational policy and actual practice) and "means-end" decoupling (the gap between practice and the actual outcomes of such practice). Graafland and Smid's (2019*) study of 1,000 large companies in 24 countries provides little empirical evidence for complete decoupling, confirming the insight that policy-practice decoupling has become a rarity, even though means-end decoupling is frequent. The study moves further by showing that CSR/SD policies of a high quality (as measured by scope and level of detail) have a relatively stronger effect on CSR/SD implementation, while CSR/SD programmes of a high quality (as measured by scope and the use of targets and strict deadlines) have a relatively stronger impact on outcome. The analyses reported suggest that even low-quality CSR/SD policies have a positive influence on subsequent programmes and that assigning responsibility for CSR at the board level reduces decoupling (2019, 232). The authors reflect that "introducing a weak program[me] is still better than having no program[me] at all" (2019, 257), which confirms the "better than nothing" argument put forward by research dedicated to the usefulness of decoupling.
Another recent configurational study of CSR/SD decoupling digs even deeper into the complexity of CSR/ SD implementation. In an analysis of 19 large companies, the researchers showed that, depending on the environmental and/or social focus of the CSR/SD policies, distinct configurations of factors can explain the enhancement of CSR/SD performance (Halme et al. 2020*). This research moves beyond a one-size-fits-all take on CSR/SD implementation to identify two pathways for improved environmental performance: "an exogenous pathway for publicly listed MNCs adhering to systems and standards, and an endogenous pathway for cooperatives and family-owned firms with more internally customized approaches to CSR" (2020*, 1184). It also identifies two pathways for the improvement of social performance: one that focuses on the integration of CSR/SD through control systems, and another that focuses on strong values and stewardship policies. In contrast to these four configurations for leading to improved performance, the authors found multiple configurations of factors leading to one or the other aforementioned types of decoupling.
As a whole, research on CSR/SD implementation and/or decoupling has moved beyond linear modelling of CSR/SD implementation through organizational change to provide more sophisticated and path-dependent accounts of whether and how specific configurations of CSR/SD policies and functions shape the actual implementation of CSR/SD programmes, as well as their ultimate outcomes for stakeholders. Research shows that the gaps between CSR/SD discourse and practice/outcomes are not bad, as the phenomenon may be transitory, and, in the main, the presence of ambitious claims and positions in CSR/SD departments usually ultimately has a beneficial impact, although this does depend on the quality of the CSR/SD programmes and policies. Arguably, this suggests that advocating for the presence of CSR/SD within organizational structures makes sense, at least in the long run.

Micro-level boundary processes: CSR/SD professional work
Boundary processes of relevance to CSR/SD functions and professionals also operate at the individual or inter-individual level. Pertinent studies have looked at CSR/SD professionals working in the CSR/SD unit and the positions they occupy, and in particular at their work on deploying CSR/SD strategies, enhancing these strategies' impact and influence within and outside the organization, and dealing with the tensions inherent in the assimilation of CSR/SD issues by organizations. Recent studies can be clustered into three groups. The first focuses on the contribution of individuals to the institutionalization of CSR/SD within organizations (institutional work). The second examines the more psychological work on themselves that these professionals sometimes engage in so as to be able to perform their tasks (self-work). Finally, the studies in the third group focus on the competencies that individuals often develop in order to embed CSR/ SD within their organizations effectively (competency-building).

Institutional work
Studies of CSR/SD have relied, explicitly or implicitly, on the concept of institutional work 8 to examine the various activities that CSR/SD professionals engage in and the work performed by other organizational actors to promote or embed CSR/SD within their organization. These studies therefore address questions relating to bottom-up institutional change. Hunoldt, Oertel and Galander (2020*) analysed the efforts of CSR/SD managers to reduce the tension between an internal logic focused on profit maximization and an external demand for CSR/SD focused on social and environmental performance. They identified four core activities: "(a) establishing new formal structures and/or changing existing ones, (b) developing a corporate understanding of the logic and illustrating the need for adherence, (c) expressing the synergies between logics, and (d) enhancing executives' commitment to the logic" (2020*, 4).
The multiple facets of the work involved in CSR/SD institutionalization have been analysed in a wide range of empirical studies that explore how the boundaries of CSR/SD occupational jurisdictions are drawn or extended (see, for instance, Augustine 2021*). Focusing on the horizontal embedding of CSR/SD and the boundary-spanning positions of CSR/SD professionals, Wickert and de Bakker (2018*) highlight the "relational work" performed by CSR/SD professionals to support the "selling" of social and environmental issues to managers ("issue-buyers") from other areas of the organization, who may not always be sensitive to the normative nature of such issues. On the basis of 54 open-ended interviews with CSR/SD managers at German MNCs, they documented how these CSR/SD professionals tempered their own aspirations and motivated themselves by espousing the role of "internal change agent", while patiently building up internal influence by winning allies and establishing proximity with, and influencing the worldviews of, the key issue-buyers they were targeting in order to facilitate CSR/SD adoption. Furthermore, Gond, Cabantous and Krikorian (2018*, 242) stress the importance of embedding CSR/SD in corporate strategy, coining the term "strategifying work" to refer to "institutional work that aims to change the boundaries of strategy so that a new notion becomes regarded as strategic within an organization and potentially across multiple other organizational settings". This qualitative study shows how the head of CSR/SD at a UK utility company succeeded in connecting CSR/SD policies to corporate strategy at the cognitive, relational and material level through multiple micro-tactics in her daily work.
Other institutional work-based studies of CSR/SD have emphasized the normative and moral aspects of this endeavour. Demers and Gond (2020*), for instance, found that executives, managers and employees at an oil sands company needed to engage in "justification work" in order to defend on moral grounds the CSR/SD strategy to be implemented (see also Nyberg and Wright 2013*). Recent studies have also noted the power-imbued and deeply political nature of CSR/SD-related institutional work (Lefsrud and Meyer 2012*). For instance, a study by Acosta, Acquier and Gond (2019*) of how a global CSR/SD standard was implemented at a Colombian supplier to an MNC shows how coercive and deliberative power dynamics characterize the institutional work undertaken by competing groups of employees and managers to influence the company's adoption of implicit and explicit forms of CSR/SD.
To sum up, institutional work studies have documented a number of micro-level activities at organizations. Such studies provide a realistic picture of what CSR/SD professionals do, and shed light on the deeply relational, strategic, normative and political nature of their work. However, these micro-practices have not been compiled in an easily transferable way, and their contribution to the effective alignment of CSR/SD policies, practices and outcomes needs to be evaluated more systematically. 8 Institutional scholars have coined the term "institutional work" to refer to how individuals deal with, and often try to modify, elements that have been institutionalized within and across organizations. According to Lawrence and Suddaby (2006, 215), institutional work may be described as "the purposive action of individuals and organizations aimed at creating, maintaining and disrupting institutions".

Self-work
The complex and especially demanding nature of CSR/SD work has been captured in qualitative empirical studies that highlight the inner challenges faced by these new professionals -in particular as they strive to affirm their new status while constructing their professional or occupational identity, and as they reflect on the sometimes limited impact of their work. A number of studies focus on how CSR/SD professionals define their identity, beliefs and emotions through their work. Scholars have paid special attention to the intensive identity work inherent in CSR/SD managerial activities ( . This identity work seeks to reconcile the tensions faced by CSR/SD managers "between their own sense of self and the various work and non-work contexts in which they find themselves" Grant 2012*, 1451). The heterogeneity and lack of stability of CSR/SD managers' professional identities make such work both necessary and difficult (Lefsrud and Meyer 2012*): it often gives rise to a hybrid identity, reflecting a kind of "tempered radicalism" (Meyerson and Scully 1995). A compromise has to be struck between political ideals related to CSR/SD and the pragmatic nature of corporate life. This is encapsulated in the metaphor of an "activist in a suit" (Carollo and Guerci 2018*), who is balancing distinct identities (Ghadiri, Gond and Brès 2016*) in an attempt to sustain such a compromise.
Other studies have focused on the emotional and psychological demands of CSR/SD professional work. Nyberg (2012*, 1561) explore, in the Australian context, how 36 corporate sustainability specialists dealing with climate change operate as "emotionology workers" as they get to grips with the various emotions (such as anxiety, fear and anger) expressed in the media in response to their companies' policies, and draw constantly on their own passion for ecology to persuade their colleagues to act. The authors' analysis reveals the techniques used by CSR/SD workers to deal with their own emotions and cope with an energy-draining task that may at times seem impossible to fulfil. Sonenshein, DeCelles and Dutton (2014*, 7) conclude that "it's not easy being green" in two studies that explore the enabling but also potentially damaging role of self-doubt among professionals working on the selling of environmental issues.
The existing research therefore suggests that CSR/SD professional work is special in that it often involves the entire "self" of employees, mobilizing their energy, but also their beliefs, emotions and deepest levels of reflexivity (see also Tams and Marshall 2011*). The implications of such a moral and emotional load for the design of CSR/SD functions have rarely been discussed in detail.

CSR/SD competency-building
Adopting a more pragmatic approach, other studies focus on the set of competencies and skills required to perform the tasks of CSR/SD professionals and to develop tools for embedding CSR/SD more broadly within an organization. Various associations of CSR/SD practitioners, for instance, have developed CSR competency frameworks (as in the case of the CSR Academy in the United Kingdom, or the Sustainability University). Based on interviews with 28 CSR/SD professionals (mainly managers) from Dutch MNCs, Osagie et al. (2019) identified six distinct managerial roles -coordinating, stimulating, networking, strategic, monitoring and mentoring -and explored the cognitive, functional, social and overarching (meta) competencies required to fulfil each of these roles. Their analysis suggests that each role calls for a distinct prioritization and combination of various competencies, which confirms the complexity of day-to-day CSR/SD work. These roles stretch the limits of what is expected of managers because of their need to contend with external social and environmental pressures.
Another rigorous and practice-focused systematic exercise, undertaken by Bertels, Papania and Papania (2010), built a repertoire of individual-level actions and tactics for embedding CSR/SD within organizations.
The authors distinguish between formal and informal practices aimed at either fulfilling sustainability-related obligations or achieving innovation. They identify 20 practices for embedding sustainability into organizational culture, which are classified into four groups according to the main objective: "clarifying expectations", "instilling capacity for change", "building momentum for change" and "fostering commitment".
By focusing on practice, such studies have helped to promote understanding of what it takes to be a CSR/ SD professional. However, in view of the rapid and fragmented development of the CSR/SD field, more continuous systematic and integrative efforts may be necessary to specify the various CSR/SD competencies and roles in diverse regions of the world. Future studies could also explore whether and how combinations of competencies and types of role may lead to more effective CSR/SD outcomes.

Managing stakeholder relations through CSR/SD boundary processes
The various boundary processes identified thus far explain the embedding of CSR/SD functions and professionals within organizations. The current report now "zooms in" to consider how these organizational changes have reshaped the boundary process whereby organizations interact with their stakeholders and have altered the dynamics of stakeholder relations. Figure 2 illustrates how the boundary-process framework can be used to understand these aspects.
Specifically, the research team's analysis was guided by two questions:

• How are CSR/SD structures and operations affected by external actors and pressures?
•

What are the most common types of relationship between CSR/SD programmes and the various stakeholders?
From the literature review, two groups of studies have emerged that can explain the mechanisms underlying the boundary process behind stakeholder relations with reference to the embedding of CSR/SD functions and professionals: the first group focuses on how organizations respond to pressure from their stakeholders, while the second group focuses on the role of multi-stakeholder partnerships. The main findings of each group are presented in turn in the following subsections.

Confrontation: Impact of external stakeholder pressures
While most existing studies have concluded that stakeholders exercise direct influence on CSR/SD implementation (Helmig, Spraul and Ingenhoff 2016*), research is ongoing to identify when and why different types of stakeholder pressure have varying impacts on companies' adoption of certain CSR/SD structures and activities.  King and Soule (2015*) found that firms that were targeted by activists to a greater extent -in this case by shareholder activists -were more likely to establish CSR/SD committees at meetings of their board of directors or to start issuing a CSR/SD report regularly. The authors give examples of high-profile cases, such as that of Nike, to illustrate the effect that social movements and activists have historically had on companies' adoption of CSR/SD. Their conclusion is that Fortune 500 firms in general are more likely to adopt CSR/SD structures and practices as a result of overt targeting by activists.

Receptivity to stakeholder pressures
Taking a step back, however, some studies have found that firms are less likely to heed demands from activists or NGOs, which, along with the media, can be regarded as "secondary stakeholders"; instead, firms tend to listen to their "primary stakeholders", which typically include employees, investors (Gond and Piani 2013*), the government and customers. Building on earlier work, such as the stakeholder salience model proposed by Mitchell, Agle and Wood (1997) (see also Wood et al. 2021), Helmig, Spraul and Ingenhoff (2016*) investigated the pressures exerted by primary and secondary stakeholders on CSR/SD implementation by conducting a survey of 196 top managers of industrial firms in Switzerland, who self-reported on the stakeholder pressures faced by their organizations and on their organizations' responses in the form of changes to CSR/SD implementation. They found that these firms' decisions on CSR/SD implementation were influenced mainly by pressures coming from primary, rather than secondary, stakeholders.
However, some scholars have criticized the treatment of different stakeholder groups as disconnected entities, since secondary stakeholders often work extensively by informing and mobilizing primary stakeholders (for example, NGOs often mobilize customers through boycotts or lobby the State to provide greater oversight of CSR-related issues). In fact, Helmig, Spraul and Ingenhoff (2016*) found that pressure from secondary stakeholders influenced primary stakeholders. In this way, secondary stakeholders could still exercise influence over CSR/SD implementation; it was just an indirect influence.

Hypocrisy avoidance
While many studies indicate that firms make public concessions in response to activist targeting -especially when such targeting threatens their reputation (King 2008;Soule 2012) -there is an ongoing debate as to whether firms do so simply in a "cosmetic" fashion for the sake of public relations, or whether activist targeting can lead to longer-term or more substantive CSR/SD changes. In a new qualitative study of how activist targeting influences CSR/SD adoption and change, Waldron et al. (2020*) examine four activist organizations' campaigns aimed at pressuring firms to adopt new CSR/SD practices or change their existing ones. Drawing on a sample of successful and unsuccessful campaigns, 9 the authors found that successful campaigns emphasized how a given company was engaging in questionable ethical behaviour, offered a solution for the company based on new programmes and commitments, and laid out the potential consequences in the form of specific threats if the company did not change its ways. The successful campaigns worked by stigmatizing certain types of behaviour, and thereby also stigmatizing the companies associated with such behaviour. Significantly, the authors found that secondary stakeholders -namely the activistsoften claimed to represent primary stakeholders (especially consumers) or threatened to inform consumers and prompt them to change their buying patterns -a powerful reputational and economic threat. This is consistent with King's (2008) conclusion that boycotts are more likely to be successful when they are more threatening reputationally than economically.
One key factor found to help explain when a campaign was successful or not was the presence of a mechanism often referred to as "hypocrisy avoidance" (Waldron et al. 2020*); that is, if firms had made previous CSR commitments or claims, they were more likely to yield to future targeting. This is consistent with one of the key additional findings of McDonnell, King and Soule (2015*), who also tested whether the companies in their Fortune 500 sample that had adopted either a CSR committee or CSR report would be more receptive to activist targeting in the future and more likely to concede to activists' demands. They found that this was indeed the case.
Studies focusing on this aspect posit that, once organizations have committed to being responsible, they put themselves in the spotlight, as it were, and it is more difficult for them to go back on such commitments without coming across as hypocritical. These findings on hypocrisy avoidance suggest that CSR/SD structures and activities are relevant when it comes to changing an organization's practices in the long term.
One study even found that firms sometimes hide their sustainability credentials -in this case, their inclusion in the Dow Jones Sustainability Index -for fear of being labelled as hypocritical in the future (Carlos and Lewis 2018).
In general, studies on activist targeting do not provide a full examination of the various structures that companies put in place following targeting. For example, activist targeting is often the first step in the creation of more formalized channels for stakeholder dialogue (Pedersen 2006*), which may be followed by the establishment of partnerships to manage CSR/SD-related issues that require continuous coordination between multiple actors (including companies, civil society organizations and often also State actors). As described in the following subsection, there is extensive literature on stakeholder engagement studied from the angle of multi-stakeholder partnerships.

Cooperation: The role of multi-stakeholder partnerships
The literature on organizations' engagement in partnerships focusing on CSR-related issues has been comprehensively reviewed in a series of papers, including those by Austin and Seitanidi (2012), Clarke and Crane (2018*) and Laasonen, Fougère and Kourula (2012). Before considering the findings of these studies, it is important to have a baseline definition of what today are commonly referred to as "multi-stakeholder partnerships" (MSPs). MSPs are broadly defined as a collaborative form of governance involving mainly business actors and civil society organizations that come together to find a common approach to a complex problem that affects them all (Roloff 2008;Rasche 2012

What makes MSPs successful for an organization?
One key factor to consider is the role played by third parties in facilitating effective partnerships between business and civil society. Arenas, Sanchez and Murphy (2013*) explore this relationship in detail, concluding that third parties enable firms and civil society organizations to move from confrontational and conflictual relationships to collaborative relationships. The authors build their analysis on the baseline understanding that pressure from civil society may take the form of conflict, but when companies overcome conflict and collaborate with civil society organizations, they are more likely to succeed in assuming their role as responsible citizens, in implementing CSR policies and in improving the social and environmental impact of production processes (Aguilera et al. 2007;Peloza and Falkenberg 2009;Seitanidi and Crane 2009). Processes of transition from conflict to collaboration are important because they are more likely to lead to forms of "critical cooperation" (Covey and Brown 2001;Hamann and Acutt 2003), promote real changes in business operations and result in improved social and environmental impact (Arenas, Sanchez and Murphy 2013*, 723-724).
In their study of four cases designed to address firms' CSR/SD-related challenges, ranging from indigenous rights to supplier labour issues, the authors found that not only were third parties essential for moving beyond contentious or adversarial relationships, but that they also played a key role in moving relationships beyond mere transactions (for example, philanthropic relationships between firms and NGOs) (Arenas, Sanchez and Murphy 2013*).
Other studies highlight the positive effect of having collaborative or distributed decision-making processes in place within the partnership (MacDonald, Clarke and Huang 2019*) and the benefits of setting up governance structures that include conflict resolution procedures (Garriga 2009). Furthermore, it is important to control the size of a partnership: quantitative studies have shown that, as a partnership expands (in terms of the number of organizations involved in it), it is less likely to work effectively (MacDonald, Clarke and Huang 2019*).
There have also been a few studies on the drawbacks or unintended consequences of MSPs. One interesting study by Dentoni, Bitzer and Pascucci (2016*) finds that, over time, through their engagement in MSPs, companies' capability to interact with stakeholders increases, but that their ability to learn from stakeholders and change their ways decreases as they gain more experience in CSR/SD. The authors attribute this to the development of corporate strategies on sustainability, "which entails a shift from a reactive to a proactive attitude towards sustainability issues and which may decrease the need or motivation for stakeholder orientation" (2016*, 35). Interestingly, ongoing engagement in MSPs may result in firms being less responsive to future CSR-related issues if they see these as falling outside the scope of what they have determined as their primary CSR strategy (for example, education issues versus labour issues).

What makes MSPs successful for society?
There has been some debate on how best to measure the effectiveness and success of these types of partnership. When partnership success is measured in terms of organizational-level outcomes, MSPs appear to bring a wide range of benefits to partner organizations. For example, Clarke and MacDonald (2019*) conducted 47 interviews with local partners involved in MSPs related to sustainability planning in Canada and found that they reported cost savings, relationship and reputation gains, and an improved knowledge base as a result of their involvement in the partnerships.
However, scholars are increasingly arguing that the success of partnerships should be measured by positive changes in relation to societal-level issues and concerns, rather than by changes at the level of individual organizations (Waddell et al. 2015). For example, in the case of the MSPs for sustainability planning, this would imply measuring success in terms of emissions reductions, rather than reputation gains. On a similar note, Dentoni, Bitzer and Schouten (2018*), in examining the Roundtable on Sustainable Palm Oil, concluded that MSPs are more effective when they move on from the deliberation and decision-making stages to the enforcement stage, which often happens through the creation of voluntary standards as informal regulatory measures (discussed in subsection 2.1.1.1, "Regulation"). However, the authors also acknowledge that, despite the hope that MSPs can prompt "systemic change" (Clarke and Crane 2018*), the case of the Roundtable on Sustainable Palm Oil, which has received a great deal of criticism for the lack of change it has prompted at the organizational, societal and issue levels, shows that "it is very challenging to identify any causal relationship between organizational changes in MSPs and the breadth or depth [of] systemic [change]" (Waddell et al. 2015)" (Dentoni et al. 2018*, 349

Shaping interactions with employees and trade unions through boundary processes
In earlier sections, this report has looked at the boundary processes associated with the embedding and structuring of the CSR/SD function and the work of new CSR/SD professionals (section 2.1) and explained how these boundary processes can transform organizations' capacity to interact with their stakeholders (section 2.2). This section uses the boundary-process framework (see Figure 1 above) to examine how CSR/ SD functions and professionals have reshaped organizations' relationships with employees and trade unions. The analysis in this respect was guided by the following question:

• How is the institutionalization of CSR/SD functions related to the collective and individual relationships of employees to CSR/SD programmes and policies?
As with the embedding of CSR/SD functions and professionals, the literature review conducted for this report revealed that the above question has been addressed by a wide range of studies from various angles (such as stakeholder theory and organizational psychology). These studies seek to explain particular features of worker-and labour-specific CSR/SD issues, such as private governance (international framework agreements and codes of conduct) implemented in both transnational companies (i.e. MNCs) and suppliers, the role of unions in this process, and employees' engagement in CSR/SD at the micro level. In line with the approach adopted in section 2.1, the three broad categories of macro, meso and micro are used to classify the literature dedicated to employee-related organizational dynamics. First of all, a process taking place at the extra-organizational level is considered, namely the emergence of various forms of private governance of worker-and labour-specific CSR/SD issues (such as regulation of human and labour rights in supply chains, and global health and safety issues) (2.3.1). The extra-and inter-organizational levels are examined next, namely where unions interact with CSR/SD functions (2.3.2). Finally, the focus shifts to the intra-organizational and intra-individual level, where individual employees interact with CSR/SD functions (2.3.3). Table  4 provides an overview of the boundary processes identified and the illustrative studies analysed as part of the literature review. Figure 3 illustrates how the three CSR/SD boundary processes discussed in Part 2 are involved in worker-and labour-related issues.
X Table 4. Boundary processes linking CSR/SD functions to employees and trade unions: Key concepts and illustrative studies

Boundary processes Definition and nature of organizational boundaries involved Key concepts and illustrative studies
Private governance of worker-and labour-specific CSR/SD issues Mechanisms of the governance dynamics Field-level processes whereby workerand labour-specific CSR/SD issues are governed through "soft" laws Unions' CSR strategies Field and organizational interface where unions redefine and negotiate their own CSR/SD meaning and practices The nexus between CSR and industrial relations [Colombo, Guerci and Miandar 2019] Mechanisms [Dawkins 2010;Sobczak and Havard 2015] Interactions between individual employees and CSR/SD functions Mechanism of employee engagement in CSR/SD Intra-organizational boundary processes whereby individual employees engage in CSR/SD functions Active employee engagement [Girschik 2020;Skoglund and Böhm 2020] Employee voice [Godkin 2015] Organizational-level conditions affecting the reaction of employees to CSR/ SD policies

Private governance of worker-and labour-specific CSR/SD
In the twentieth century, industrial relations and labour and workers' rights issues in Europe and the United States were tackled through collective bargaining and industrial agreements between firms and trade unions (Egels-Zandén 2009a*). As a result, industrial, or labour, relations constitute one of the most regulated areas of CSR/SD worldwide (Matten and Moon 2008;. However, over the past two decades, the industrial relations landscape has undergone fundamental changes with the rise of global supply chains for MNCs. Egels-Zandén (2009a*) has identified four major trends in the governance of workers' rights: (a) MNC-led governance systems that partly fill institutional voids in developing countries; (b) a weakening role of unions and national governments, with a shift towards transnational private governance; (c) the increasing influence of NGOs in industrial relations; and (d) a shift from legal to voluntary governance -that is, "soft" regulations, such as codes of conduct and international framework agreements (IFAs).
The present review highlights similar trends in the recent literature on collective interactions between workers and CSR/SD functions. When it comes to worker-and labour-related CSR issues, the vast majority of the articles reviewed focused on the emergence of private forms of governance, initiated either by MNCs or by transnational MSPs (such as the Better Work programme, Reebok's worker empowerment programmes, the Accord on Fire and Building Safety in Bangladesh, and IFAs). Through institutionalized CSR/SD functions, these new forms of governance affect MNCs and their suppliers in developing countries.
The rise in soft regulations -namely, the private governance of workers' health, safety and rights issueshas mostly to do with institutional field-level boundary processes that connect MNCs' headquarters and their suppliers. Through these mechanisms, MNCs' private governance or multi-stakeholder soft regulations influence both workers and the framing of labour issues at supplier firms in developing countries. These macro and transnational labour-focused processes have been discussed in terms of their mode of functioning, their impact on workers and the factors determining their effectiveness. Accordingly, the following subsections look in turn at private governance mechanisms, their effectiveness and social impacts on workers, and the moderators of private governance implementation.

Private governance mechanisms
Labour and workers' rights are governed by dynamics that are of a transnational and multi-stakeholder nature (Egels-Zandén 2009a*). Numerous studies have explored how private governance in this area works in relation to the roles of other stakeholders. The multi-stakeholder nature of private governance empowers various stakeholders through networking (Gereffi and Lee 2016*;Mele and Schepers 2013*), and, as a result, NGOs have become an especially influential stakeholder group, orchestrating societal pressures through multi-stakeholder governance platforms (Egels-Zandén 2009a*; Fransen and Burgoon 2014*). Egels-Zandén and Wahlqvist (2007) -citing the Business Social Compliance Initiative, a project that was unsuccessful owing to the lack of NGO support -argue that codes of conduct can sometimes marginalize important stakeholders whose support is needed to establish robust and consensus-based definitions of key concepts.
Although the role of trade unions has been eroded by the rise of transnational private governance of labour and workers' rights, the importance of global and local union involvement in the governance process has been repeatedly noted (

Effectiveness and social impact of private governance
Noting the rise of transnational private governance of labour-and worker-specific CSR/SD issues, several studies have cast doubt on the effectiveness of such governance mechanisms and their ability to positively impact workers' rights, mostly at supplier firms in developing countries. There is no academic consensus on this matter, with earlier studies coming to different conclusions on the basis of empirical evidence. Most studies express scepticism regarding the effectiveness of soft regulations, which are mostly initiated by MNCs that require their suppliers to comply with their codes of conduct on labour and working con- Assessing the compliance of suppliers located in developing countries with MNCs' codes of conduct, several researchers have identified not only a clear lack of compliance (Egels-Zandén 2007*; Yu 2008*) but also these codes' lack of impact on workers' and unions' freedoms and rights even when suppliers do comply with them (Wang 2005*). According to these scholars, structural, process-related and institutional issues are responsible for this ineffectiveness. First, operational pressure associated with just-in-time products and intensive competition within the marketplace -where there is unresolved tension between profit maximization and commitments to workers' human rights -creates structural problems in the dynamics between MNC headquarters and their suppliers ( However, a small yet growing number of studies have challenged such scepticism by showing some successful cases where IFAs and codes of conduct have resulted in better labour and working conditions. For instance, looking at Chinese suppliers in the toy manufacturing industry between 2004 and 2009, Egels-Zandén (2014*) found instances of codes of conduct being "recoupled" with actual practices, evidenced by an enhanced level of supplier compliance with MNCs' codes of conduct. In these cases, the adoption of such codes led to improvements in occupational health and safety, working hours and other standard-related outcomes. Beyond the corporate-led codes of conduct, Bourguignon, Garaudel and Porcher (2019*) also found IFAs to be a good platform for CSR/SD managers and other central actors to cooperate with unions on improving the effectiveness of social and labour policies throughout MNCs. Other studies suggest that trade unions can play an active role in global production networks (Fichter, Helfen and Sydow 2011*) and can assist with the local-level implementation of private labour standards (Williams, Davies and Chinguno 2015*). Studies also suggest that social dialogue can be improved through transnational governance frameworks (Lévesque et al. 2018*), which complement company-produced codes of conduct. Egels-Zandén and Hyllman (2007*) even argue that IFAs could be considered a better approach than unilateral codes of conduct.

Moderators of private governance implementation
Moving beyond the debate on whether mechanisms for the private governance of labour-and worker-specific CSR/SD are effective or not as a whole, researchers have started to explore the factors conditioning the success or failure of the implementation of soft regulations at supplier firms over the past decade. Several authors highlight structural factors that influence the effective implementation of private governance in suppliers. For instance, Egels-Zandén (2014*) concludes that codes of conduct positively impact suppliers by increasing the demand for audits and transparency and by producing better coordination across the industry. Graz, Helmerich and Prébandier (2020*) identify three structural conditions that enhance a trade union's capacity to be influential in the context of transnational private governance: (a) the ability of workers' organizations to gain access to the processes of employment regulation, implementation and monitoring; (b) their ability to include employers and state agencies in such processes; and (c) the ability of workers to effectively exercise leverage in pursuit of particular goals. Well-functioning social dialogue between MNCs and workers' representatives from suppliers has been singled out as an important structural factor ensuring that transnational private governance has a positive impact on workers' rights (Lévesque et al. 2018*;Mamic 2005*; Reinecke and Donaghey 2020*). Lévesque et al. (2018*) found that this positive impact depends on whether managers and worker representatives are able to mobilize their capabilities across multiple levels. The authors also suggest that the implementation of IFAs could be facilitated by "the sharing of decision-making competencies at different levels and the dissemination of management and trade union capabilities across levels" (Lévesque et al. 2018*, 227).
Other studies have identified relational factors that affect the impact of private governance on workers' conditions and rights, such as interdependency among stakeholders within global value chains (Niforou 2015*; Tran and Jeppesen 2016*) and collaboration with NGOs (Egels-Zandén and Hyllman 2011*; Egels-Zandén and Wahlqvist 2007*) and governments (Hart 2010*;Tran and Jeppesen 2016*). For instance, Niforou (2015*) investigated how multilevel dynamics impact workers' use of power along global value chains, concluding that the sources of labour leverage are contingent on the degree of interdependency among different nodes of a supply chain. Whether pragmatic, strategic or political, these interdependencies in turn determine the outcomes of labour leverage, which may range from short-term isolated "victories" to long-term sustained improvement in working and employment conditions. On the other hand, Tran and Jeppesen (2016*) emphasize how the development of shared norms and awareness in relation to both codes of conduct and CSR/SD among managers and workers can help to improve working conditions.
Overall, there is still an ongoing debate on the extent to which regulations covering worker-and labour-specific CSR/SD influence working conditions at supplier firms. These regulations are the result of multilayered mechanisms of private governance, where various stakeholders collaborate with one another while having their own interests in mind, and so power and political issues are often involved (Dawkins 2019*; Reinecke and Donaghey 2020). Recent empirical studies have tried to settle this debate by looking at the factors underlying successful or failed implementation of soft regulations (see Egels-Zandén and Hyllman 2011*; Graz, Helmerich and Prébandier 2020*; Lévesque et al. 2018*;Niforou 2015*). However, more research is required to analyse how private governance of worker-and labour-specific CSR/SD issues can have a positive social impact.

Interactions between trade unions and CSR/SD functions
A second stream of research has taken trade unions and their interactions with other stakeholders as its main (meso-level) unit of analysis. Even though some of these studies focus on how unions influence global processes of private governance, they are discussed here under the umbrella of meso-level analysis.
In contrast with the studies discussed above that noted a loss of trade unions' influence as a result of private governance processes, some researchers emphasize the enduring importance of unions with regard to CSR/SD issues. In their studies, they investigated a wide range of positive institutional field-level and outside-in boundary processes, whereby unions and workers influence CSR/SD issues not only by facilitating suppliers' compliance with private governance but also by supporting the implementation of CSR/SD within MNCs and by deploying their own CSR/SD strategies.

The role of unions in private governance mechanisms
The rise of private governance in industrial relations is a key trend that emerges from recent studies on collective interactions. In the vast majority of studies, workers and unions are regarded as the beneficiaries or targets of private governance, even though it has been pointed out that workers' participation is essential to make the implementation of private governance mechanisms successful (see Graz, Helmerich and Prébandier 2020*; Tran and Jeppesen 2016*). Only a handful of studies have noted how unions and workers can intervene in various subtle ways in the implementation of private governance at supplier firms. For instance, in the Chinese context, Yu (2009*) analysed how Reebok's worker empowerment programmes -which may be regarded as a private form of labour standard -have operated through interactions with stakeholder groups, each of which has its own distinct interests. The study focuses on two aspects of worker participation: individual participation as workplace monitors, and collective participation through the trade union, where collective participation complements individual participation. For instance, when individual participation was limited, the union played a highly active consultative role in resolving many problems related to working conditions, such as issues of discrimination, child labour, forced labour, health and safety, and working hours. Unions played a moderately active consultative role on management-related issues, such as harassment and abuse by managers. A decade later, in the French setting, Bourguignon, Garaudel and Porcher (2019*) similarly highlighted the potentially active role of unions as "monitoring agents" in the implementation of private governance. Their empirical study shows that union networks can be a key resource for CSR/SD managers at MNCs who wish to be able to better monitor suppliers.

The role of unions vis-à-vis CSR/SD functions at multinational corporations
Research dedicated to the relationship between unions and MNCs has provided evidence of the active role of unions in the implementation of CSR/SD policies by such corporations, and has explored the mechanisms underlying unions' influence. First, several studies indicate that unions can drive MNCs' adoption of CSR/SD functions and practice through outside-in boundary processes. For instance, Egels-Zandén (2009b*) explored why an (unnamed) European MNC had adopted IFAs, and found that one of the motivations was to retain a relationship of trust with the trade union movement within the firm. Zhao, Tan and Park (2014*) observed how trade unions in China, India and the Russian Federation engaged aggressively with the CSR/ SD functions and policies of MNCs. Some sophisticated processes were involved: for example, the All-China Federation of Trade Unions requested MNCs to set up trade unions, and unions in India and the Russian Federation protested against MNCs over environmental pollution, health and safety conditions and the beneficiaries of philanthropic programmes. Harvey, Hodder and Brammer (2017*) highlight how trade unions can help to bring about deliberative democracy by influencing CSR/SD functions, and they suggest that unions can be both agents of democracy and democratic agents for MNCs' CSR/SD functions.
To explain the positive influence of unions, Harvey, Hodder and Brammer (2017*) note how these are embedded in institutionalized "voice channels" and point to an alignment of incentives. Similarly, Egels-Zandén and Hyllman (2006*) shed light on how unions and NGOs can enter into strategic partnerships in order to maximize their impact on CSR/SD policies at MNCs. In a quantitative study, Boodoo (2020*) identified a positive correlation between union density and employee-oriented CSR/SD policies, and a more complex relationship between non-employee-oriented CSR/SD and union density. At low levels of unionization, as union density rises, companies tend to "sacrifice" non-employee-oriented CSR/SD in order to meet the demands of unions, whereas at higher levels of unionization, firms are able to invest in policies and programmes that are of greater interest to external stakeholders while still meeting the demands of employees and unions.

Trade unions' CSR/SD strategies
At the institutional field level, the industrial relations landscape has shifted from a national to a transnational focus, not unlike the global governance literature on CSR/SD. As a result, many scholars have identified a growing overlap between the realms of CSR/SD and industrial relations (Colombo, Guerci and Miandar 2019*; Egels-Zandén 2009a*). Colombo, Guerci and Miandar (2019*) explore similarities and differences between CSR/SD and industrial relations by focusing on what unions and companies in the metal and chemical industries in Italy negotiate in relation to CSR/SD. They found that the specific legacies of these two industries account for the differences in what is considered by trade unions in the two sectors to constitute CSRrelated activity, despite the homogenizing influence of the national legal framework for industrial relations.
Exploring further the nexus between CSR/SD and industrial relations, some researchers examine what unions do in the CSR/SD domain, analysing, for example, unions' own CSR/SD strategies. They have investigated both the nature of such strategies and the mechanisms through which they operate. For instance, Dawkins (2010*), focusing on what unions do for society, introduces the concept of "labour union social responsibility", which comprises three key objectives: economic equity, workplace democracy and social justice. Each objective informs a different social responsibility strategy: the service, organizing, and covenantal strategies, respectively (Snape and Redman 2004). Although these strategies reflect different views of the role of trade unions in society, they are not mutually exclusive. Dawkins (2016*) tests some of his insights by examining the effect of social responsibility on key measures of trade union attachment, such as union commitment, job satisfaction, propensity to participate and propensity to withdraw. Drawing on a quantitative survey of carpentry apprentices in the United States, he found that members who perceived trade unions to be more socially responsible were also more committed to the union, more satisfied with their jobs, more likely to participate in union activities and less likely to withdraw from the union. This study notes the positive outcomes of unions' CSR/SD strategies not only for society as a whole but also for the unions themselves. Through a qualitative longitudinal study of selected unions in France between 2006 and 2013, Sobczak and Havard (2015*) were also able to shed light on the nature of unions' CSR/SD strategies and how these interact with four stakeholders: the central Government, NGOs, individual companies and other unions. The authors found that French unions' CSR/SD strategies ranged from reactive to proactive and were influenced by all four above-mentioned stakeholders. Unions may choose a different CSR/ SD strategy depending on how they perceive each stakeholder group.

Interactions between individual employees and CSR/SD functions
The institutional field-level boundary processes eventually cascade down to individual employees within organizations with embedded CSR/SD functions and policies. It is therefore appropriate to "zoom" in on the micro-level interactions that CSR/SD functions have through intra-organizational and individual mechanisms. These mechanisms are usually covered in micro-CSR studies that draw on organizational psychology, communication theory and organizational sociology (see Gond and Moser 2021;Jones and Rupp 2018). The present review focuses in particular on internal boundary processes where embedded CSR/SD functions interact with individual employees: that is, on how individual employees perceive, react to and engage with CSR/SD units within their organizations. Following , the relevant studies cover mechanisms of employee engagement in CSR/SD, organization-level conditions influencing individual employees' reactions to CSR/SD, and the outcomes of such reactions.

Mechanisms of employee engagement in CSR/SD
Micro-CSR studies have been conducted to look at the interactions between individual employees and CSR/ SD functions from a micro perspective, that is, considering intra-organizational mechanisms. Although numerous psychological micro-CSR studies have dealt with employees' perceptions of corporate CSR/SD initiatives and with the positive impact of these perceptions on their commitment to the organization, job satisfaction and engagement (see , the questions as to why, when and how employees contribute to their organization's CSR/SD initiatives have received considerably less attention (see, for example, Glavas 2016*; Rupp et al. 2018*). Most studies have essentially focused on how to engage employees in the organization's CSR/SD functions or policies. For instance, Mirvis (2012*) highlights three approaches to promoting such engagement: (a) the transactional approach, where CSR/SD programmes are undertaken to satisfy the needs and interest of employees who wish to participate in their company's external CSR/SD activities; (b) the relational approach, where a company and its employees collaborate to conduct CSR/SD programmes; and (c) the developmental approach, where an organization involves all employees in its CSR/SD efforts with a view to "developing" both the company and its employees so that they are able to produce greater value for business and society.
Girschik (2020*) and Skoglund and Böhm (2020*) provide empirical evidence of the potential for more proactive employee engagement through CSR/SD-related intra-organizational processes. In particular, Girschik studied how a team of employees at a Danish pharmaceutical firm developed a "framing" of their company's responsibilities and, through a bottom-up process, transformed the company so that it became more socially and environmentally responsible. These employees acted as "internal activists" by producing appealing and inspiring communication materials about their organization's CSR policies. Looking at a European energy firm, Skoglund and Böhm focused on how mundane environmental actions among employees could lead to organizational change towards a "greener" way. Their study shows how employees' everyday actions can serve as environmental activism that transcends organizational boundaries.
In his study, Godkin (2015*) presents a rare attempt at modelling the multiple intra-organizational mechanisms that influence employees' engagement in CSR/SD. His model presents the interactions between middle managers and employees and illustrates how "managers can encounter employee engagement in CSR and channel their enthusiasm effectively" (2015*, 15), notably by paying attention to employees' "ethical voice", namely the "signals" emanating from their earlier personal engagement in CSR/SD initiatives at their organization.

Organization-level conditions influencing employees' reaction to CSR/SD
Micro-level studies exploring the interactions between individual employees and CSR/SD functions also focus on the conditions that influence employees' reactions to CSR/SD -another area that is relatively under-researched . De Roeck, El Akremi and Swaen (2016*) examine how, and under which conditions, employees' perceptions of CSR affect their identification with the organization. The authors found that perceptions of overall justice influenced the level of identification, with two mediating factors playing an important role: the (perceived) external prestige of the organization, and pride in being a member of the organization. Considering two samples of 107 professionals and 205 employees from various industries, Lawrence and Kacmar (2017*) found that emotional exhaustion could lead employees to engage in unethical behaviour, and that factors such as job insecurity, the extent to which individuals are embedded in their organization, and their adaptability played an important mediating role. Hejjas, Miller and Scarles (2019*) have investigated the factors that positively and negatively influence employee engagement in CSR/SD policies. They found that organizational culture, the design of CSR/SD interventions, employees' perceptions of CSR/SD and the observed benefits of participation were key factors in that respect. Wang, Zhang and Jia (2020*) have studied how the influence of CSR/SD policies on "employee voice" is affected by the prevalence of an ethical climate and "humble leadership" in an organization. They found "other-focused" and "self-focused" ethical climates to have multiple mediating effects on that relationship; the former refers to individuals' propensity to prioritize the interests and well-being of others, the latter to their propensity to maximize self-interest when making decisions of an ethical nature. Humble leadership has a moderating effect: it positively influences the relationship between CSR/SD and other-focused climates and negatively influences that between CSR/SD and self-focused climates.

Outcomes of individual employees' reactions to CSR/SD
A third aspect covered by micro-CSR studies is the wide range of outcomes of employees' reactions to CSR/ SD initiatives -a topic that has been receiving a great deal of attention recently . Most of these studies rely on organizational psychology theories, such as fairness theory (Rupp 2011*;Thornton and Rupp 2016), social identity theory (De Roeck, El Akremi and Swaen 2016*;Farooq, Rupp and Farooq 2017) and social exchange theory (Archimi et al. 2018;De Roeck and Maon 2018*). Moreover, these studies tend to highlight positive attitudinal and behavioural outcomes in the workplace .
An exception is the study by Costas and Kärreman (2013*), who critically evaluate the influence of organized CSR/SD policies on employees and show how internal CSR initiatives are embedded into organizational control mechanisms: CSR/SD functions could potentially be used as a tool by managers to exercise control over employees.
However, most studies in this group emphasize the positive impact of CSR/SD policies on employees. For instance, Beaudoin et al. (2019*) explore how a company's commitment to CSR ultimately determines the ability of employees to make ethical decisions resisting the temptation of personal gain. These studies offer explanations for why employees react in a particular way to CSR/SD initiatives  and as well as cognitive outcomes. Their paper investigates the effects of (in)consistent external-internal CSR strategies on employee attitudes, intentions and behaviours. Drawing upon social and moral identification theory, they conducted two studies, one involving 3,410 employees and the other 1,902 employees. In the first, a cross-industry employee survey was conducted, which found that the use of inconsistent CSR strategies with greater external efforts than internal efforts increased employee's intentions to leave their jobs because of their perceptions of corporate hypocrisy and emotional exhaustion. From the second study, based on a multi-source secondary dataset, it emerged that the use of inconsistent CSR strategies increased actual employee turnover at the firm level.

X 3 Critical synthesis, current challenges and perspectives for future research
By reviewing the literature dedicated to CSR/SD functions, with a focus on three complementary topics, this report has sought to provide an overview of studies that analyse how organizations engage with CSR/SD at the individual, organizational and institutional field levels. The boundary framework proposed in the first part of the report for organizing these studies with reference to these levels ( Figure 1) has helped to provide a coherent picture of existing research and to identify knowledge gaps. In the first section of this final part of the report, these gaps are reviewed critically for each topic, as are the research questions that they raise. The second section presents four overarching challenges that limit the current understanding of how CSR/ SD functions operate, while the third and final section offers an agenda for advancing research in this area.

Critical synthesis: Knowledge gaps
Focusing on the boundary processes involved in embedding CSR/SD functions (Topic #1), managing stakeholder relations (Topic #2) and reshaping relationships with employees and trade unions (Topic #3), the analysis undertaken for this report has revealed several research gaps. Table 5 provides an overview of these gaps and relevant questions for further research on each topic.
X Lack of clarity as to which CSR/SD boundary processes are effective at translating issues.
Lack of knowledge of how "soft" regulations and policies influence the embedding of CSR/SD functions.
Over-representation of the business case perspective.
Insufficient analysis of whether and how CSR/SD functions and professionals transform organizations.
Lack of consolidated framework to make sense of the work and competencies of CSR/SD professionals.
What are the relative effectiveness and impact of the boundary processes operating at different levels? Are some CSR/SD professionals more effective than others in translating and promoting CSR/SD issues in their organization?
How are "mistranslations" of CSR/SD issues being addressed or corrected through various regulatory, market or professional field-level interventions? How do third-party organizations support the translation of CSR/SD issues?
Can CSR/SD professionals and consultants be mobilized beyond the business case to support CSR/SD implementation? If so, how?
What different types of institutional work and micro-practices are deployed in the context of CSR/SD?
What forms of cognitive dissonance are experienced by CSR/SD professionals, and how do they manage these?

Topic #2: Managing CSR/ SD stakeholder relations
An incomplete picture of the mechanisms explaining how CSR/SD boundary processes make organizations sensitive and reactive to external pressures.
Focus on localized changes of targeted organizations rather than systemic changes.
Lack of analysis of how multi-stakeholder partnerships can be turned into CSR/SD regulatory devices.
What are the various mechanisms whereby CSR/SD boundary processes make corporations more reactive and responsive to external CSR/SD issues? Which other mechanisms are there apart from hypocrisy avoidance?
What coalitions of stakeholders in multi-stakeholder partnerships are most effective in terms of affecting CSR/SD boundary processes so as to generate sustainable organizational capacity to deal with CSR/SD issues?
How can confrontational and partnering strategies be balanced to enhance the social and environmental impact of CSR/SD functional units?

Topic #3: Reshaping relations with employees and trade unions
Fragmented knowledge of the roles of unions in organized CSR/SD functions.
Unclear picture of the structuring forces underlying unions' CSR strategies.
Lack of understanding of how employees contribute to the structuring of CSR/SD functions.
Employees dominantly depicted as passive recipients of CSR/SD initiatives, and lack of discrimination between the needs and expectations of different types of employees.
What is the role of unions in private governance mechanisms? (Need to further take into account theories and concepts of political and industrial relations, such as social movement theory, collective bargaining, labour process theory and stakeholder theories).
What are the CSR/SD strategies of unions, and what impact do they have in different national settings with different economic systems?
Through which mechanisms do employees react to specific CSR/SD policies?
What are the outcomes generated through employees' engagement with CSR/SD initiatives, including employees' ability to voice their concerns?
How do employees, managers and executives differ in their reactions to CSR/SD policies?

Topic #1: Embedding CSR/SD functions and professionals
The analysis of boundary processes undertaken for this report shows that regulative, market and professionalization dynamics drive the embedding of CSR/SD functions within organizations. These trends suggest that CSR/SD functions and professionals are here to last. Several studies have clarified how the development of various kinds of regulations sustains the continuous need for CSR/SD experts in organizations, and have highlighted the crucial role of market forces and professionalization in the diffusion and adoption of CSR/SD practices and ideas. However, further research is required to investigate these dynamics, determine their relative effectiveness as channels of transmission of CSR/SD priorities and issues and clarify how they interact with one another productively. It would be interesting to explore whether and how the professionalization of the CSR/SD field -for instance, through dedicated certification of managerial competencies -could help to counter the dominant business case framing of CSR/SD and the commodification process described in subsection 2.1.1.2. Another worthwhile area of research would be how commodification and regulation interact in ways that may reshape organizational CSR/SD functions and practices.
Studies of the structuring of CSR/SD functions highlight a great variety of processes and configurations of CSR/SD departments, and the findings on organizational (structuration), interpretative (translation) and institutional (implementation/decoupling) dynamics need to be consolidated further in order to identify the most effective ways of designing and developing CSR/SD functions. Most studies point to a relatively positive impact of the creation and development of such CSR/SD functions, especially when these functions are connected to the top level of the organization's hierarchy. However, their capacity to transform organizations and achieve a strong impact remains insufficiently studied. In the view of some scholars, CSR/SD functions are condemned to marginalization; for others, their presence, even where they remain peripheral to the organization's core activities, is a necessary step towards the embedding of CSR/SD. Future studies could adopt a longitudinal design to address this gap.
As for CSR/SD professionals, the present review suggests that many aspects of their work, which encompasses relational, strategic, normative and political dimensions, have been relatively well documented, albeit in a fragmented manner. Moreover, the effectiveness of CSR/SD professionals' tactics has yet to be evaluated systematically. CSR/SD roles and jobs are highly demanding, and they are also constantly expanding, since CSR/SD functions cover many emerging social and environmental issues: for example, loss of biodiversity and climate change, or, more recently, socio-economic issues related to the COVID-19 pandemic. Studies with a focus on self-work call for greater scrutiny of these jobs by managers (in particular human resources managers), because they present emotional and identity challenges. Linking the institutional work and competency-building approaches could help improve understanding of how the ability of CSR/SD professionals to influence colleagues and the organization as a whole could be enhanced.

Topic #2: Managing stakeholder relations
A relevant question is whether confrontational activism has more potential than partnerships between business and civil society when it comes to shaping the implementation of CSR/SD and prompting companies to seek to improve their social and environmental impact, or whether it is the other way round. From the present review it may be concluded that this is almost certainly a false dichotomy. Both NGO activism and partnerships between companies and NGOs have proven to be effective drivers of the adoption of CSR/ SD programmes (and of changes to these), with the two mechanisms playing different but often mutually reinforcing roles at different stages of the CSR/SD implementation process.
Activist confrontation and the associated campaigns to stigmatize behaviour and threaten corporate reputation are often necessary in order to draw attention to certain practices and to secure initial commitments from companies (King 2008;McDonnell, King and Soule 2015*;Waldron et al. 2020*). CSR/SD commitments, structures and practices that are put in place may at first seem like window dressing on the part of companies, but these can eventually "walk the talk" and become more responsive to future targeting. This is partly due to hypocrisy avoidance, that is, where company leaders are concerned lest they appear hypocritical or lest they face reputational damage that could cost them the support of primary stakeholders, such as investors and customers (McDonnell, King and Soule 2015*;Waldron et al. 2020*). The above process also shows how small initial commitments can lead to substantive changes: once firms have taken a small step of that kind, they are now more "in the spotlight". However, this is an area that is worth exploring further, especially the role of CSR/SD managers in facilitating internal changes in an organization and promoting responsiveness to external targeting.
Partnerships show great potential with regard to bringing about systemic change to address societal-level "wicked problems" that cannot be tackled by single firms, especially when the partnerships involve some element of enforcement, rather than just dialogue (Clarke and Crane 2018*; Dentoni, Bitzer and Schouten 2018*). These partnerships can often best be facilitated by third-party organizations from civil society, which can help companies to move beyond confrontational or transactional relationships to collaborative ones (Arenas, Sanchez and Murphy 2013*). However, partnerships can fall short of their initial promise: they can stall, key partners may leave, and they may struggle to achieve measurable systemic change, perhaps because they tend to reproduce rather than alter underlying power dynamics (Dentoni, Bitzer and Schouten 2018*). Furthermore, firms can reject certain demands from stakeholders as they start to engage in CSR/ SD more strategically, becoming known for openness and innovation on certain issues while withdrawing from others (Dentoni, Bitzer and Pascucci 2016*). As Dentoni, Bitzer and Schouten (2018*, 333-334) point out, "despite their attempts of complementing public institutions in domains where governments are not able or willing to regulate (Scherer and Palazzo 2007*), the problems that MSPs seek to address remain far from being tamed or, in some cases, have become even more acute". Scholars have only just begun to understand when partnerships are most effective at bringing about organizational change as opposed to deflecting attention from companies' behaviour by simply appeasing stakeholders.
Activist pressure is valuable in MSP engagement as it can further drive CSR/SD implementation by holding organizations accountable to commitments they have made in the past and by drawing their attention to new areas of CSR/SD that they may have been ignoring. Furthermore, activist pressure has been shown to trigger the "radical flank effect" (Haines 1984), that is, where the presence of activists demanding far-reaching transformations makes those asking for less radical change seem much more reasonable (van Huijstee and Glasbergen 2010). Partnerships may be able to break out of an impasse or to advance their goals if there is renewed activist pressure on firms to go further in their CSR/SD commitments, as MSPs and the NGOs that are willing to engage in them will then appear more reasonable.
An important general finding is that there is a set of boundary processes linking firms to their external environment, and, in particular, to stakeholders who are working to influence CSR/SD commitments, structures and practices. Confrontational targeting by activists can usher in new CSR/SD commitments and structures. After firms make initial commitments to change, hypocrisy avoidance explains why they seem to be more open to future demands. Inducing companies to make initial or small changes, regardless of the full extent of their intentions with regard to CSR/SD, can lead to more openness to future change. From this review it is clear that partnerships, notably MSPs, are an increasingly prevalent approach for CSR/SD engagement with stakeholders, especially when dealing with "wicked problems", such as water management, greenhouse gas emissions, deforestation or child labour, which cannot be tackled by a single organization or sector. MSPs can often best be facilitated by third-party organizations, and they are more successful in bringing about changes in the underlying systems if they have distributed decision-making processes (that embrace and adequately manage conflicts). However, there is a lot of criticism of "weak" enforcement mechanisms and standards devised by third parties and MSPs, which confirms the need for confrontational activism that can serve as an ongoing critical voice and help to enhance voluntary enforcement mechanisms and standards.

Topic #3: Reshaping relations with employees and trade unions
Analysis of the CSR/SD boundary processes governing the relationship between CSR/SD functions with employees and trade unions reveals three dynamics: the rise of private governance of worker and labour issues, union-CSR/SD interactions, and interactions between individual employees and CSR/SD functions.
Studies have captured a growing number of private forms of governance of worker and labour issues that have become institutionalized in MNCs' supply chain networks. Initiated by MNCs and multi-stakeholder groups, these various forms of private governance provide guidelines that are meant to protect workers' rights and ensure decent working conditions. Researchers have explored key mechanisms in this area by focusing on collaborative and political relationships with other stakeholders, in particular NGOs.
Most studies tend to focus on the private and self-regulating nature of corporate governance and on NGOs' active engagement with private governance platforms. In contrast, the role of governments has been largely neglected. This is despite the fact that governments continue to be active players in CSR/SD (soft) regulation platforms. As suggested by some of the studies reviewed, governments are necessary to improve workerand labour-specific standards (Laudal 2010*) and can fill the void created by companies' reliance on the business case for CSR/SD self-regulation (Hart 2010*). It would be helpful to have more studies exploring various mechanisms where governments -not only those of the countries in which suppliers are based but also those of the countries hosting the headquarters of the MNCs -play a role in the private governance of worker-and labour-specific issues. Moreover, although there are recurrent calls for more worker involvement as a key condition in the successful implementation of private governance (Graz, Helmerich and Prébandier 2020*; Hart 2010*; Tran and Jeppesen 2016*), it is still not well understood how workers become involved in bottom-up governance dynamics, rather than just top-down governance initiated by MNCs and by external stakeholders such as NGOs. Finally, studies dealing with this aspect rely heavily on stakeholder theory and political CSR. It would be worth using new theoretical angles to explore the mechanisms, outcomes and conditions of private governance of worker-and labour-specific CSR/SD.
Studies of union-CSR/SD interactions highlight three main aspects: (a) the role of unions in the private governance mechanism at the institutional field level of analysis; (b) the role of unions in MNCs' CSR/SD at an organizational level; (c) unions' CSR strategies as the interface between the environment and the organization. However, there is still only a limited number of studies looking at the role of unions and their contribution to the development of CSR/SD functions. In particular, the various roles played by unions in influencing MNCs' CSR/SD functions and suppliers' codes of conduct have not been properly explored.
The literature dealing with Topic #3 reviewed for this report suggests that interactions between individual employees and CSR/SD functions can produce diverse attitudinal and behavioural outcomes among employees. These outcomes are relatively well documented in studies based on organizational psychology. However, research has tended to focus on CSR/SD initiatives that organizations create for, or together with, mostly external stakeholder groups. Only a few studies look at how CSR/SD functions and policies interact with employees. Moreover, studies of employee reactions to CSR/SD have been somewhat biased towards positive emotions and rarely explore the dark side of employee engagement in CSR/SD. Despite the important role of such employee engagement, the intra-organizational and individual mechanisms underlying how employees actively or even proactively engage in CSR/SD are still not well understood, and the same applies to the key factors contributing to these mechanisms. Having more studies that focus on employee-centred individual and organizational mechanisms of engagement in organized CSR/SD functions and policies would help to shed light on how employees are actively involved in organizing and developing CSR/SD functions.

Tackling four key research challenges
Beyond the gaps discussed in the above section, four overarching challenges have emerged from the literature review and from the interviews with academic experts conducted by the research team: isolation, disempowerment, glocalization and marginalization. Figure 4 presents these research challenges and a description of each one; they are discussed in depth in the following subsections.
3.2.1. Challenge #1: Isolation -The CSR/SD boundary processes at the individual, organizational and institutional field level have been studied in isolation; the fragmentation of knowledge about them undermines the potential collective impact of CSR/SD functions and professionals The first challenge is the fragmentation of knowledge about CSR/SD boundary processes at the individual, organizational and institutional field level -a situation that reflects the organization of CSR/SD scholarship and the "essentially contested" nature of CSR and SD Mitnick, Windsor and Wood 2020). This prevents researchers from being able to explain whether and how CSR/SD functions and professionals make a difference within their own organizations, as well as collectively across industries and at the societal level. The academic experts interviewed for this report also flagged this as an especially problematic issue. They emphasized how connections between the different levels of analysis are important, noting, for example, how the degree of formalization and structuring of CSR/SD functions can affect CSR/ SD professionals' work, and how dynamics at the institutional field level may lead to the superficial adoption of CSR/SD functions unconnected to the rest of the organization.
Vertical isolation -that is, the lack of studies cutting across the individual, organizational and environmental levels of analysis -makes it difficult to fully understand what makes CSR/SD functions impactful within organizations and whether the professionals who populate them can effectively influence organizational practices. This challenge has been noted in the CSR/SD field, with many studies calling for the development of multi-level research (Aguinis and Glavas, 2012;Gond and Moser 2021;Jones et al. 2019).
Horizontal isolation -that is, the lack of knowledge about how boundary processes relate to one another and interact across multiple organizational settings and cultural contexts -makes it difficult to understand how CSR/SD practices and actors cross organizational boundaries and, therefore, to evaluate the extent to which CSR/SD can sustain or promote forms of collective action for the common good.
3.2.2. Challenge #2: Disempowerment -Although CSR/SD functions enhance the reactivity of organizations, they remain insufficiently resourced and fragile; CSR/SD professionals are continuously at risk of losing their capacity to influence organizations A second challenge refers to the ability of CSR/SD functions and professionals to attract executives' attention and to support the deployment of effective CSR/SD initiatives at organizations. Studies dealing with Topic #1 indicate that appointing dedicated CSR/SD executives can have a significant impact, while studies covering Topic #2 suggest that the mere presence of CSR/SD functions and professionals is likely to make organizations more receptive to stakeholder pressures and, therefore, more likely to address CSR/SD issues.
On the whole, however, existing studies at the individual and organizational level provide a much more mixed perspective on such influence, suggesting potential disempowerment. The literature describes how individuals and organizations seeking to embed CSR/SD have continually struggled with various types of tensions, which can ultimately lead to the disempowerment of CSR/SD professionals (Berti and Simpson 2021). This point was also raised during the interviews with academic experts. Several of them were sceptical about the internal power of CSR/SD functions and professionals; they stressed that practitioners were aware of the issue and noted how a changing context might help them: These are the right people trying to go to do good things as they can, and they know what they're doing is not really changing much. They are well aware of the limitations they have in the functions. (Interview, Expert #2) Internal sustainability people have a mandate, but it doesn't extend into the capital allocation or budgeting process. So, if they want to change this, they have to take what little capital they have and go outside. (Interview, Expert #1) In the specific case of interactions with procurement departments, which play a key role in labour issues through MNCs' supply chains, CSR/SD professionals usually follow rather than lead: I think it's clear that when it comes to the power relationships it's the sourcing people who call the shots. (Interview, Expert #4).
However, all six experts agreed that "things would be worse" without these professionals. It is therefore legitimate to conclude that CSR/SD functions and professionals are still in a fragile position, partly because they need to defend their influence continuously, both within and outside the organization, and partly because they have no well-developed knowledge basis to rely on.
3.2.3. Challenge #3: Glocalization -CSR/SD boundary processes are subject to tensions between global and local forces that can undermine the implementation of CSR/SD initiatives across organizations, supply chains and countries The third challenge that cuts across the three topics and was recognized by the academic experts interviewed for this report has to do with the tension between globalization -the global forces standardizing CSR/SD organizational practices (such as global norms, and national and international regulations) -and localization, which refers to the situational contingencies inherent in the adoption and implementation of such practices across various organizational settings, ranging from subsidiaries of MNCs to distant suppliers. To use the neologism popularized by Robertson (1992) to describe this tension, "glocalization" makes it difficult to evaluate the impact of the mere presence of CSR/SD functions or professionals within a given organization, as their reach within the organization and along supply chains may be limited.
Nowhere was this tension more apparent than in studies dealing with Topic #3, investigating organizations' ability to implement forms of private governance that protect labour and workers' rights effectively. These studies highlight the need for organizations' active cooperation with local as well as national governments. As Expert #4 put it in her interview, CSR/SD managers "often have to look to HQ before acting", which slows down potentially impactful CSR/SD initiatives in the supply chain.
3.2.4. Challenge #4: Marginalization -CSR/SD boundary processes follow a top-down logic, which may fail to take into account marginalized actors (e.g. workers, women and the global South) The fourth overarching challenge manifested itself less visibly in the present literature review, precisely because it has to do with marginalized voices and stakeholders, and with the "blind spots" created and perpetuated by research that tends to focus on corporations. With regard to the embedding of CSR/SD within organizations (Topic #1), studies are biased towards larger MNCs headquartered in Western countries. The review of studies dealing with Topic #2, moreover, suggests that pressure from radical and alternative actors (such as more critical NGOs) -actors that are potentially marginalized in the institutional field or seen as "secondary stakeholders" -could be the most effective way of promoting change at the institutional field level.
Even though the various boundary processes identified during the review (see Figure 1) all support the consideration of social and environmental issues within organizations, these boundaries may accommodate a narrow and corporate-focused definition of responsibility or sustainability. Consistent with a focus on issues that have a direct financial and material impact on the organization in the short term, CSR/SD priorities defined in a top-down manner are likely to leave aside issues perceived as more complex or that are advocated by powerless stakeholder groups. As explained by one of the experts interviewed by the research team: "Boundary spanners are biased [toward corporate interests] about the boundary that they are spanning" (Interview, Expert #1). The review of studies falling under Topic #3 suggests that the under-representation of powerless actors has to do with the difficulty of assessing the perspectives of employees, who are the most directly impacted by corporate supply chains. As explained by Expert #4, in a country such as Bangladesh, the presence of company unions (or "yellow" unions) and challenging working conditions makes it extremely difficult to interview the most vulnerable actors in the supply chain. The challenge of marginalization has been aggravated by the COVID-19 pandemic.

Consolidating CSR/SD boundary processes: A research agenda
3.3.1. Research direction #1: Bridging CSR/SD boundary processes vertically and horizontally to consolidate knowledge Connecting the dots, as it were, of prior research on CSR/SD functions and their role within organizations is important to overcome the fragmentation that has been noted in this area. The multilevel CSR/SD boundary-process framework proposed in Part 1 of this report (see Figure 1) is a first step in this direction. However, that framework has revealed a lack of cross-cutting studies covering the many CSR/SD boundary processes at the individual, organizational and institutional field level. More consolidative reviews of the literature, together with theory-building efforts, are needed to bring together the findings from each level. Future research could use mixed-methods designs to obtain insights at multiple levels of analysis. Methods such as fuzzy-set qualitative comparative analysis would be well suited for this. This analysis could support studies of how specific configurations of characteristics of different boundary processes at different levels explain organizational-level outcomes in terms of CSR/SD integration or impact on stakeholders. Such methods could help to conceptualize how boundary processes operate in relation to one another, in complementary or substitutive ways. Insights from such studies could be complemented with the use of more traditional techniques, such as hierarchical linear modelling or regressions, to trace the impact of specific factors related to different boundary levels on CSR/SD-related organizational outcomes.
Clarifying how boundary processes between communities of CSR/SD practice, CSR/SD professional associations and informal networks of CSR/SD practitioners interact would require further conceptual efforts and methodological creativity. Concepts such as communities of practice or informal networks capture distinct facets of the collective dynamics underlying the practice of CSR/SD professionals. The fact that communities of practice, professional associations and informal networks all cut across multiple organizational settings provides an opportunity to conceptualize whether and how, collectively, distinct groups of CSR/SD professionals interact with one another across organizational borders. Various theoretical frameworks, such as institutional theory, social movement theory or collective action theory, could all help in different ways to explore whether and how CSR/SD functions, through institutional field-level dynamics of communities of practice or professionalization, can reinforce the impact of one another. Analytical tools of network analysis could support the mapping and investigation of formal and informal networks of CSR/SD professionals. Establishing lateral connections between grassroots CSR/SD movements within and outside organizations can also help in understanding the deeply cosmopolitan and transnational nature of CSR/SD professionalization. Future studies could look at whether such horizontal connections support the translation of CSR/ SD initiatives across multiple countries and cultural contexts with a view to identifying the cultural barriers and enablers of CSR/SD diffusion from the perspective of practitioners. Comparative studies of national and transnational communities of CSR/SD professionals would also be valuable. Understanding better how boundary processes are connected horizontally and vertically is a precondition for evaluating their transformative potential.

Research direction #2: Reconsidering temporal dynamics in the analysis of how CSR/SD boundary processes operate
The challenges of isolation and marginalization highlight the need for (re)consideration of temporal dynamics. Existing studies of CSR/SD boundary processes have looked at the embedding and structuring of these boundaries, in addition to their influence on organizational responsiveness to stakeholder pressures or on the translation of labour rights issues.
However, these studies are limited by their specific approach to temporality. Little is known about how CSR/ SD functions, issues and careers have evolved over time, not only at the institutional field level but also at the organizational and individual levels. Restoring the temporal dimension is key at the organizational level of analysis, since a growing number of studies suggest that issues such as CSR/SD disconnection from the rest of the organization (functional decoupling) may be a transitory phenomenon (Haack, Martignoni and Schoeneborn 2020*) and that decoupling can enhance contradictory attributions and interpretations of the value of CSR/SD initiatives by employees, depending on their temporal dynamics (Christensen, Morsing and Thyssen 2020*; Glozer and Morsing 2020*). Researchers could analyse how CSR/SD functions and professionals deploy their activities over time within and across organizations. Within organizations, future studies could focus on the life cycle of CSR/SD functions in specific contexts in order to support the optimization of their design. Arguably, the capacities required to address CSR/SD issues in response to stakeholder pressures may differ according to the maturity of the organization, and different types of expertise may be required depending on the degree of institutionalization of CSR/SD functions. Future studies could document the life cycle of CSR/SD maturity and CSR/SD structuring to clarify which conditions facilitate the effective embedding of CSR/SD in organizations. Another promising avenue of research with a temporal focus would be tracing the deployment of CSR/SD initiatives across multiple units or subsidiaries of the same organization over time in order to identify any CSR/SD discourse-practice gaps and analyse how such discrepancies are managed by CSR/SD professionals, as well as how they are perceived and dealt with by employees.
Enhanced awareness of temporal dynamics at the individual level can help in understanding the career trajectory of CSR/SD professionals and the differences between distinct generations of CSR/SD experts in different industries. First, depending on CSR/SD maturity at the institutional field or organizational level, distinct profiles of CSR/SD professionals could be relevant, and future studies could focus on what is required, and when, with regard to CSR/SD expertise. Second, studies dedicated to CSR/SD managers or consultants have identified multiple tensions faced by these professionals but have said little about how they experience such struggles over time and learn to deal with these tensions throughout their careers. Paying more attention to temporality could help improve understanding of whether and how CSR/SD professionals become cynical or maintain their faith in their efforts. Third, a growing body of "grey literature" suggests that generational differences are relevant when it comes to sensitivity to CSR/SD issues. Identifying differences in the attitudes and behaviours of distinct generations of employees or CSR/SD managers could shed light on individual engagement with CSR/SD issues within and across organizations.
In view of the urgency of CSR/SD issues, such as climate change, that call for immediate action by businesses and governments, developing a sound knowledge of the temporal dynamics underlying the embedding of CSR/SD functions is crucial. It would also facilitate understanding of how CSR/SD professionals could help to address the health challenges arising from the COVID-19 pandemic (Aguinis, Villamor and Gabriel 2020).

Research direction #3: Harnessing the regulatory potential of CSR/SD boundary processes
Both the literature review and the interviews with experts undertaken for this report suggest a limited ability of CSR/SD functions and practitioners to advance CSR/SD issues within and across MNCs; at the same time, however, it is clear that their absence would be even more harmful. Rather than throwing the baby out with the bathwater, as it were -something that critical scholars focusing on CSR have argued in favour of (Fleming and Jones 2013) -it is important to think about how to make CSR/SD functions and professionals more effective at the various levels.
In this respect, it would be worth reviving Bowen's (1953) early insight, according to which CSR-related processes should be seen as a regulatory tool in their own right -in addition to legislation (hard law) and market mechanisms -that can be of help when considering and managing CSR/SD issues. Accordingly, the focus of attention should be not on whether CSR/SD functions and professionals have achieved their ambitious objectives, but on how the infrastructure they form across multiple MNCs and organizational settings could be best used by other organizations -governmental organizations, NGOs, standardization bodies and professional associations -to deal more effectively with CSR/SD issues. This raises several new questions: How can CSR/SD functions and professionals be leveraged to enhance their impact? How can the regulatory effectiveness of CSR/SD functions and professionals be enhanced? Which organizations or social dynamics can help maintain the focus of CSR/SD functions and professionals on the search for the common good?
These questions can be explored across the individual, organizational and institutional field levels of analysis. At the individual level, researchers could build on recent insights about "paradoxical tensions" and their effective management through strategies to help CSR/SD managers cope with the emotional burden of CSR/SD work. At the organizational level, future studies could investigate how to prevent CSR/SD teams and managers from being sidelined, and could consider how they can be empowered through intra-organizational competencies and capacities to combat the still all too prevalent "loneliness of the CSR/SD manager syndrome". At the institutional field level, more work is needed to clarify what needs to be done to strengthen the professional, network and inter-organizational dynamics that can sustain the deployment of better equipped and more impactful CSR/SD professionals across organizations. Future studies could look at how "hard" regulators (that is, governments) and "soft" regulators (that is, standardization bodies) can engage effectively with CSR/SD think tanks, professional associations and communities of practice to enhance the capacities of CSR/SD professionals through processes such as certification, training and networking. Re-empowering CSR/SD professionals can help to strengthen the effectiveness of CSR/SD boundary-process at MNCs, providing a new transnational regulatory lever. Such an approach involves thinking more "laterally" about how new spaces of interaction between different types of CSR/SD professionals could be designed and how dedicated regulations can be developed to harness and enhance these professionals' capacity to influence organizations.

Research direction #4: Redesigning CSR/SD boundary processes to take better account of marginalized voices
Including alternative voices is necessary to tackle a long-standing challenge affecting the CSR/SD area, namely the disempowerment of certain actors and the under-representation of marginalized groups, such as women, ethnic minorities and workers in supply chains, particularly in the global South. This would be in line with the critical discussion of individual-level CSR research by Girschik, Svystunova and Lysova (2020), who call for the scope of research to be broadened "to cultivate the potential of alternative ideas, voices, and activities found in organizational life" and propose a research agenda that "embraces employee activism, listens to alternative voices, and unfolds confrontational, subversive, and covert activities" (Girschik, Svystunova and Lysova 2020, 1). Accordingly, the various CSR/SD boundary processes identified in Part 1 of this report (see Figure 1) should not be regarded as closed or fixed boundaries serving the narrow interests of a new class of CSR/SD managers and professionals, but as a socio-material infrastructure supporting the further development of CSR/SD processes (Gond and Nyberg 2017).
Capitalizing on the dual nature of boundaries as separating but also connecting distinct domains or entities (Lamont and Molnár 2012;Weager and Weber 2017), this report proposes the "opening up" of existing CSR/SD boundaries so that human rights, labour rights and societal issues can be better dealt with through CSR/SD boundary processes and professionals. Such research should involve analysis at the individual level (for example, of CSR/SD professionals, whether they are activists or not), but also consider how marginal voices can be represented, translated and amplified across levels of analysis by taking into account organizational and institutional field dynamics. Girschik, Svystunova and Lysova (2020) call for a greater focus on activist employees who support, develop and sometimes even launch social movements that may affect their corporation's policies in the environmental domain. As players in internal corporate processes and external carriers of social issues, employee activists and CSR/SD professionals can extend the boundary processes discussed in this report. They can amplify or represent marginalized actors who, even though they may be more exposed to the social or environmental side effects of corporate activities, tend to fall outside the scope of "official" CSR/SD initiatives. Engaging with marginalized actors poses unique challenges for researchers with regard to access and sampling, and methodological creativity is required to devise new research techniques that can reach these actors.
Striving for inclusivity and considering the voices of activist employees and marginalized stakeholders exposed to corporate social or environmental side effects also involves the study and design of relevant CSR/ SD boundary processes at the organizational level. Like whistle-blowers, activist employees or employees raising concerns about the negative social and environmental aspects of their organization's policy may be subjected to retaliation. Using the CSR/SD boundary processes to mobilize such employees, therefore, requires the design of robust bottom-up mechanisms that allow their voices to be heard by those in the upper echelons of the hierarchy while ensuring that they are protected. This is consistent with the findings of studies on the factors contributing to the enhancement of private forms of labour governance, which point out how the design of appropriate "complaint mechanisms" to protect workers who voice concerns is essential (Marx and Wouters 2016). Studies of the factors hindering effective unionization, such as that by Louche, Staelens and D'Haese (2020) on the flower-cutting industry in Indonesia, offer a useful template for identifying organizational factors that can make CSR/SD initiatives accessible to those individuals along the supply chain whom such initiatives are supposed to serve. Future studies could consider a broader set of issues in relation to unionization and identify ways to involve workers in the design of CSR/SD initiatives in an inclusive manner.
As for the role of institutional-field and societal-level dynamics, studies of the limitations of corporate global regulation implemented through private labour governance are also instructive (Amengual and Kuruvilla 2020; Short, Toffel and Hugill 2020). These studies suggest that the extent to which labour standards can protect workers and their rights -for instance through a reduction or suppression of local decoupling (Bartley and Egels-Zandén 2015) -depends not only on organizational factors but also the institutional context within which factories operate (Short, Toffel and Hugill 2020). In particular, the existence of public governance processes can play a central role in the effective regulation of local labour rights (Amengual and Kuruvilla, 2020;Stroehle 2017). Such insights suggest that future studies should consider not only the role of private standards and governance but also that of governments, in particular local public authorities, in making CSR/SD boundary processes more inclusive.

X Conclusion
Following these four research directions involves a number of methodological challenges and calls for innovative methods and research designs. More radically, it requires researchers to reconsider current assumptions about the nature of CSR/SD, which, under the influence mainly of Western countries, are focused on a "business case" approach. The CSR/SD boundary processes need to be reconfigured and probably extended across multiple levels of analysis so that the voices of alternative and marginalized -yet essential -actors can be heard. It is important to bear in mind that marginalized actors are not only the ultimate beneficiaries of CSR/SD initiatives designed by corporations headquartered in Western countries, but also co-designers informing the selection and content of such policies through direct involvement.