ORIGINAL RESEARCH ARTICLE

The Diffusion of Corporate Social Responsibility Within an Organizational Field: an Analysis through the Complementary Lenses of Neo-Institutional Theory and Actor-Network Theory

Arnaud Gautier^1* and Elise Bonneveux²

¹CNRS-GREDEG, Université Côte d’Azur
²VALLOREM (E.A. 6296), IAE Université de Tours

 

 

Abstract

Since the Brundtland (1987) report and its definition of sustainable development, the societal responsibility of businesses leans toward institutionalization within organizational fields (Palmer et al., 1995). This can be explained by the growing visibility of the repercussions of businesses’ activities and by the fact that the economic liberalism and state nonintervention have led to a shift in responsibility for regulation from governments to organizations. Thus, the corporate social responsibility (CSR) represents a paradigm shift, which translates into the evolution of the dominant paradigm, a liberalism destabilized by ‘abnormalities’, toward an emergent paradigm of sustainable development bringing new answers (Kilbourne, 2004). This institutionalization of CSR, which has been supported for around 20 years by various actors, governments, public and private organizations, networks, and professional associations (Lenox & Nash, 2003), is similar to the rationale of autoregulation (Norman, 2011). It is built on various measures such as codes, auto-declarations, certification programs (Prado, 2013), development of guidelines, standards, and CSR labels, including the international standards ISO 26000, SA 8000, and OHSAS 18001, or industrial programs such as Worldwide Responsible Accredited Production, Social Accountability International, the Fair Labor Association, the Ethical Trading Initiative, and the Workers’ Right Consortium (O’Rourke, 2006).

The creation and outreach of these systems are not neutral initiatives for organizations and, more widely, for organizational fields (Scott, 1991). According to Reay and Hinings (2005), the organizational field can be understood as

[…] the normative contextual pressures that maintain stability, as well as dynamics that precipitate change (Greenwood and Hinings, 1996). […] We argue that since organizational fields are composed of actors who make up communities and are characterized by the interactions between these actors (Scott, 1994), structure, logics and political factors are all important to field level change. Actors within communities hold different institutional logics, and all fields can be characterized by competing institutional logics to some degree. At the field level, when a dominant institutional logic exists, it is because other logics are subordinate. (Reay & Hinings, 2005, p. 351)

In this context, the microfoundations approach explains the causal and recursive link between a macroenvironmental phenomenon and interactions at a microenvironmental level. By establishing a link between two levels of analysis (macro and micro), this approach enables us to establish that a macroenvironmental social phenomenon is the consequence of the behavior of one or several individuals (Coleman, 1987, 1990). By studying these two levels and delving into the intricacy of the links that emerge at different levels of analysis, it is possible to give a more precise explanation of the evolution of institutional logics. However, like Harmon, Haack, and Roulet (2018), we believe that analysis of an intermediate level (meso), which is often missing from research on microfoundations, leads to a better understanding of the social positioning of individuals within an organization and the habitus that shapes their way of thinking and acting. We follow an actor-network theory (ANT) approach to study this intermediate level as it enables us to go beyond the macrostructure – the microstructure dichotomy, which is specific to microfoundations – and to break from the paradox ‘between strategic choice for managers and the definiteness of the structures in place (resources, capabilities and routines)’ (Steen et al., 2006, p. 307). In line with Steen and Liesch (2007), we use ANT to show how the relationships between actors operate and which alliance strategies develop within a network.

By decrypting in a concomitant and complementary manner the collective and individual actions within an organizational field (in this case, France IT), this research takes an innovative look at the diffusion of a socially responsible practice within a network.

This work, based on a combination of neo-institutional theory (NIT) and ANT, enables us to:

[…] put an emphasis on the role of actors, whilst reintroducing the importance of context in the study of organizational behavior, and building new and more mesoscopic models. It therefore builds a bridge between levels of analysis and links individual actions and macro-societal influences together (Huault, 2017, p. 179)

This leads us to ask the following question: How does the implication of actors act as a condition of success in the diffusion of CSR within an organizational field? To answer this question, we study the diffusion of the Entreprise Numérique Responsable (ENR) label within the cluster network France IT using two levels of analysis. The first level of analysis is the macroenvironmental level, which we study following NIT. Literature relative to the institutionalization of the CSR identifies this process as:

[…] a group of regulations and behaviors going through the process of institutionalization […]. This process refers to the emergence and the dissemination of social norms. However, institutionalization does not fit the strict definition of behavior or the strict enforcement of a rule. On the contrary, agents benefit from appreciation and flexibility; the frontier of deliberation remains blurred and involves political and ideological values. (Dupuis & Le Bas, 2009, p. 85)

The second level of analysis is the meso-environmental level, which we study following the ANT. Under this method, the CSR is considered to be an innovation and is diffused within a network not because of its intrinsic characteristics but because of the actors mobilized around it. If the CSR is regarded as a new institution, its diffusion is translated by the development of apparatus, such as standards, labels, etc. (Slager et al., 2012). It is therefore relevant to consider the diffusion of CSR as the creation of an innovation through the introduction of a management subject that ‘carries’ this new institution. These two analysis frameworks are complementary and are able to fill each other’s theoretical gaps as they both give legitimacy to actors inside and outside of organizations.

The first half of the article presents the two theoretical frameworks, along with their contributions and limitations. First, we present the field of investigation and the methodology followed. We then detail our research, which takes the form of an exploratory qualitative study carried out in cooperation with the governing bodies of the France IT cluster network and member businesses. Finally, we present our findings and our six research proposals, which we then discuss.

Presentation of the theoretical framework

We follow two theoretical approaches to understand the diffusion of the ENR¹ label within the digital organizational field and the France IT organization. These are NIT, which enables a macroscopic level of analysis, and ANT, which looks at the formulation of facts at a meso-environmental level. These two levels of analysis are complementary and enable us to bridge the gap between ‘individual actions and macro-social influences’ (Huault, 2017, p. 179), while allowing us to simultaneously study the behavior of individuals and of organizations. Several academic works have used NIT for the analysis framework of the CSR (DiMaggio & Powell, 1983; Meyer & Rowan, 1991; Oliver, 1990; Suchman, 1995; Wood, 1991). Organizations’ strategies do not change only as a result of pressure from stakeholders. Changes to legislation and regulation, economic developments within a given sector, and technological innovations can also make an organization adopts a strategy which is more in line with CSR. Using NIT, therefore, seems an appropriate approach for studying organizations within distinct, yet complementary, environments, and for studies focused on the search for legitimacy of organizations.

The ANT model looks at organized forms of action and mechanisms for the elaboration of social facts. This theoretical approach has been used to study several management fields, such as innovation (Callon, 1986; Latour, 1989), change (Akrich et al., 2006), and line management (Amblard et al., 1996). It enables the analysis of situations; the ‘actants’, and the interactions within which socio-technical innovations appear, are stabilized and become established facts that are no longer questioned.

A neo-institutionalist reading of CSR

The institutional approach is a theoretical path which enables consideration of the constraints that are external to the company, the study of intraorganizational interactions, and the search for the legitimacy of companies. Suchman (1995, p. 574) describes this NIT anchor point, saying: ‘legitimacy is a generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate, within some socially constructed system of norms, values, beliefs, and definitions’.

A macroenvironmental process of institutionalization of CSR

Institutionalization is a process through which organizations integrate existing practices so that they become the rule and the dominant institution. Institutionalization can be achieved in two different ways: (1) when actors respond to organizational pressures and choose to conform, or (2) when institutionalization is not necessarily a constraining phenomenon for organizations but is, rather, the result of the proactive behavior of actors who are carriers of institutional change (Ben Slimane & Leca, 2010).

The question of institutional change has been at the heart of scientific debate for some decades (Dacin, Goodstein, & Scott, 2002), particularly regarding the behavior of actors who are able to mobilize resources for the purpose of institutional change (Battilana et al., 2009; Lawrence et al., 2011). Indeed, as DiMaggio (1988, p. 14) explains, ‘organized actors with sufficient resources (institutional entrepreneurs) see them as an opportunity to realize an interest that they value highly’. These actors of institutional change hold prior legitimacy within their organizational field (Maguire et al., 2004; Phillips et al., 2000), a high level of social skills (Battilana et al., 2009), and formal authority over other environmental actors (Phillips et al., 2000). As shown by Maguire et al. (2004, p. 657), these qualities are fundamental: institutional work is translated by ‘activities of actors who have an interest in particular institutional arrangements and who leverage resources to create new institutions or to transform existing ones’.

Institutional work enables consideration of the different actors and different types of action (Lawrence & Suddaby, 2006). According to Ben Slimane and Leca (2010, p. 64), ‘in the case of institutional work, the aim to trigger change is a constituent part of the observed activity. For institutional work to exist, actors who are leading it must exercise enough reflexivity in relation to the institution, in order to realize that it is a social construct’. This involves actors of institutional work being aware of the need to consider different institutions and the effects this might entail (Dorado, 2005).

According to Gondo and Amis (2013), the discourse of actors plays a key role in the introduction of a new practice because it makes the practice comprehensible within organizations. According to Callon (1998, 2006), discourse is performative as soon as it contributes to the construction of the reality it describes. For practices to be accepted and developed at an organizational level, institutional work must, therefore, rest on the discourse of the actors (Lawrence & Suddaby, 2006; Maguire & Hardy, 2009), who produce, disseminate, and consume texts, ‘allowing objects and concepts to be brought to reality and creating, or transforming, cognitive schemes through which actors interpret and give meaning to reality’ (Ben Slimane, 2012, p. 149). This observation corroborates the work of institutional theory on microfoundations by Powell and Colyvas (2008), according to whom, language plays the role of a mediator within the development of organizational routines.

Institutional work: A mesoscopic analysis of the organizational field

According to Greenwood, Suddaby, and Hinings (2002), the institutionalization process should be carried out in several stages. The starting point of the process is the social event, which focuses on regulation or technology that comes to question the dominant institution. Actors who have the ability to analyze their organizational field offer innovative solutions drawn from other institutions in closely related organizational fields (Levy & Scully, 2007). In this ‘pre-institutionalization’ stage, actors seek to surround themselves with professional actors and experts with whom they can develop alliances (Battilana et al., 2009) and to mobilize material and immaterial resources (Hardy & Maguire, 2008).

Next comes the ‘theorization’ stage, during which new solutions are diffused through practice. In this second stage, two phases, called ‘specification’ and ‘justification’, are developed conjointly (Tolbert & Zucker, 1983). During the specification phase, the goal is to share new-found solutions and to prove their appropriateness to different actors. The challenge of this institutional discursive work is to build upon arguments that will resonate with the interests of the organizational field’s actors (Hardy & Maguire, 2008; Maguire et al., 2004). During the justification phase, the challenge is to demonstrate to actors the moral or pragmatic legitimacy of their solutions. The construction of this legitimacy then goes through a process of translation of everyone’s interests, from which emerges a common social meaning in favor of the proposed change (Hardy & Maguire, 2008; Zilber, 2009). By using cognitive, normative, and/or political means (Lawrence & Suddaby, 2006), actors must exercise skill to diffuse the proposed innovative solutions, while taking account of the existing institutional environment.

The work of Daudigeos, Boutinot, and Jaumier (2015) or of Aggeri (2017) corroborates these propositions by demonstrating that the analysis of the power struggle between actors, also called ‘institutional war’ (coercive pillar) (Hoffman, 1999), is essential for understanding the diffusion of a practice. However, it becomes apparent that the inclusion, at a societal level, of collective representations (cognitive pillar) is all the more important as these representations not only play a triggering role but also influence every stage of institutional change at a collective level. If these practices are integrated by enough actors, the new institution is then diffused to the entire organizational field by way of isomorphism. This justification is materialized by the creation of an apparatus which serves as a vehicle for the discourse. This is how apparatus, such as guidelines (Yin & Zhang, 2012) and standards (Postel & Rousseau, 2008), is developed in order to diffuse and justify new CSR practices. Through their study on the institutionalization of the ISO 26000 norm, Helms, Oliver, and Webb (2012) underline the importance of the cognitive dimension in this context. This cognitive dimension refers to discourse formulation, which plays a key role in the introduction of a new practice (Gondo & Amis, 2013). In this second stage, social relationships within the organizational field are developed to support the diffusion of a new practice (Zietsma & Lawrence, 2010). The discourse, therefore, plays a key role in rallying the different network actors to the project for change. Ultimately, the reinstitutionalization stage translates into the repetition of these new practices by the actors without the initiating actors having to play a part.

If the approach to institutional work used by the NIT literature is an appropriate theoretical framework for the analysis of mimetic behaviors of companies in relation to CSR, it nonetheless has limitations when it comes to answering our research problem. For example, it cannot explain the diffusion and the appropriation of CSR principles by companies. According to Desreumaux (2004, p. 45):

The question is why certain ideas or managerial techniques acquire remarkable visibility, whilst others are not as successful; or, why certain administrative patterns are disseminated without modification from one organization to another, whilst others are reinterpreted each time and transformed without finding answers in the current version of neo-institutional theory.

Another important limitation of the institutionalization project, as defined by Greenwood et al. (2002), is that it only takes account of actors in the organizational field. If we look at studies that focus on the institutionalization of CSR, they tend to focus on a type of actor that is decisive to the process without comprehending all the aspects of diffusion. For this reason, Marais (2014) shows the importance of the leadership and style of company governance in the acceptance of CSR practices. Yet, there are ‘external’ actors whose ‘opinion’ or ‘interest’ should also be considered. Similarly, the institutionalization project rests on objects and apparatus from other institutional fields (Levy & Scully, 2007). Their conciliation within the project raises questions but has not, to our knowledge, been the subject of research in management science. Lastly, the institutionalization project suffers from a limitation that seems important to us. As presented by Greenwood et al. (2002), the institutionalization process takes place in a linear manner, from a proposal for a new practice to its integration and its institutionalization. However, the fact that the diffusion of CSR is mainly carried forward within organizational fields by innovative management systems leads us to suggest that we should not exclude the idea of a less linear approach to these steps, some of which may be concomitant if not iterative. The ANT offers an appropriate framework through the whirlwind model, which ‘allows the multiple socio-technical negotiations which give shape to the innovation to be followed’ (Akrich & Callon, 2003, p. 212).

The diffusion of CSR at the meso-environmental level

Having presented the CSR institutionalization process in order to give legitimacy to the diffusion of an innovation for its close environment, we next consider how ANT completes the analysis of the diffusion of an innovation within a socio-technical network where it was initiated. These processes together shape structures and individuals.

The translation of innovation within a socio-technical network

Initially used in scientific output, ANT ‘supports the central idea, which is that a fact is not capable of imposing itself, but is built around a ‘controversy,’ which is resolved after a series of operations started by the actors (Latour & Wooglar, 1979, pp. 230–235). The controversy, i.e., an objection to a question, an opinion, or a doctrinal point, is identified, according to Latour (1989), by characteristics specific to it. The complexity of the controversy translates into elements which are difficult to comprehend – interdependencies and ambiguities. This complexity can be examined by opening the ‘black box’, which represents the system. This allows us to uncover the alliances and the power struggles that lead to the emergence of the innovation. The innovation is, therefore, understood as a sequence of experiments and adjustments where various actors interact within the network. Indeed, the ANT theoretical framework seems appropriate for analyzing the diffusion of CSR as a translation practice by actors in the same network:

By translation we mean the set of negotiations, intrigues, acts of persuasion, calculations, acts of violence by which an actor or a force accords, or allows itself to be accorded, the authority to speak or to act in the name of another actor or force: ‘your interests are our interests,’ ‘do what I want,’ ‘you cannot succeed without me.’ As soon as an actor says ‘we,’ he or she translates other actors into a single aspiration of which she or he becomes the master or spokesperson. They start acting for the many, not for themselves. They gain strength. They grow. (Callon & Latour, 2006, pp. 12–13)

The translation is initiated by what Callon (1986) calls the primum movens. This is the actor who promotes the change and passes it onto the network’s actors. The primum movens is similar to the institutional entrepreneur as they are both responsible for the innovation. The translator is the main actor in the translation process, i.e., the one who links all the actors of the network together.

The turbulent design of innovation

According to this theoretical approach, strategies implemented by users are not the only things that shape the evolution of an innovation (social determinism model). Conceptors and users also contribute to the social construction of the innovation in a kind of constant interaction. Here, the technical object is developed in a gradual manner and can be considered as a collective co-construction. Thus, users, depending on their respective use and on the context in which the use occurs, bring modifications to the object, through conceptors. Through the different conception stages of the innovation, mediation between future users and conceptors shapes the tool in an iterative process of co-construction. The innovation can, therefore, be understood as a loop process, crystallized in the collaboration of social actors through, and within, the development of the object itself: ‘The innovation isn’t a linear process or a series of obligatory stages going, for instance, from fundamental research to development’ (Akrich et al., 1988a, p. 5).

In this approach, the socio-economic network is at the heart of innovation diffusion mechanisms. Adoption of the innovation is not explained just by the intrinsic qualities of the new product but, rather, by the way it is supported by actors who have an interest in it (product manufacturers, suppliers, distributors, prescribers, potential buyers, experts, scientific communities, funding agencies, etc.). It is through the formation of alliances that the innovation becomes established in the market because allies promote it. Research that lies within this current theoretical framework focuses on technical innovation processes by emphasizing decision-making and technical, social, economic, and political choices. For the supporters of this theoretical framework, the actors’ network constitutes a ‘socio-technical system’ (Akrich, 1993a, 1993b) or a ‘socio-technical framework’ (Flichy, 1995). Although they admit that technology has a certain autonomy within practices, they consider technical systems to be social constructs and that ‘[not] purely technical necessities, nor the imposition of certain forms of socio-politics can explain the shape that the innovations take’ (Akrich, 1993b, p. 36).

According to this model, the innovation is diffused through confrontation, compromise, negotiation, and conflict between various actors with an interest in the innovation. This is how an innovation is initially presented to a group of allies (those who have an interest in its diffusion) and to its enemies (those for whom the diffusion of an innovation constitutes a threat). By using ANT, we are able to describe the technical apparatus and its social interdependency aspect represented by the multiple actors involved in the technical apparatus. The technical object is conceived of as ‘a series of compromises between different social actors leading a project which is inscribed in their technical proposals’ (Chambat, 1994, p. 257). Innovation, therefore, is regarded as a dynamic and iterative process, crystallized in the collaboration of the social actors involved in the process and in the development of the object itself: ‘The innovation is not a linear process or a series of obligatory stages going, for instance, from fundamental research to development’ (Akrich et al., 1988a, p. 5). On the contrary, in this ‘turbulent design’, ‘innovation is a temporary interpretation of the state of nature, the technical possibilities, competing market strategies and interests; a state which will be rapidly modified by what others do with the proposed innovation’ (Callon & Latour, 1986, p. 17).

Thus, innovation is ever evolving according to the ‘tests we put it through, each new equilibrium materializing as a prototype concretely testing the feasibility of the imagined compromise’ (Pascal, 2006, p. 202). The first prototype is, therefore, rarely sufficiently conclusive, and several stages are generally necessary (Akrich et al., 1988a), leading to an iterative loop process of co-construction.

As it focuses on the networking of actors and as, unlike in social network theory, actors are not defined a priori and considered stable, ANT is an appropriate means for analyzing the diffusion of an innovative CSR practice. According to ANT, actors, resources, routines, and organizational abilities constantly evolve and, therefore, are not stable over time (Steen et al., 2006). It is, however, the struggle to develop and maintain a position within the network that leads to defining and stabilizing the role of each party.

Discussion of theoretical frameworks around the institutionalization of CSR

As raised above, NIT considers organizations to be conditioned by their environment. The level of analysis in the study of institutional change is the meso-environmental level, focused on the organizational field. While the scientific literature on institutional change highlights the importance of involving actors in the organizational field in the process, the collective dimension of actors who carry out institutional change is not adequately considered (Ben Slimane & Leca, 2010). This observation is in line with the work of Gondo and Amis (2013), who highlight the lack of this collective dimension within institutional work focusing on the diffusion of new practices. When initiated by actors in an organizational field, sectoral labels are a form of collective sectoral action. It, therefore, seems appropriate to regard the creation and diffusion of CSR labels as innovative collective actions within organizational fields. We, therefore, agree with Lawrence and Suddaby (2006) for whom institutional work is the result of several actors coordinating with each other.

We consider CSR to be a major organizational innovation as it brings a new and more modern way of thinking and represents a reforming paradigm (Damanpour et al., 2009; Eberhard-Harribey, 2006). It encourages the company to rethink relational models and managerial practices by adopting a socially responsible strategy founded on the Triple Bottom Line principles (Elkington, 1997). The CSR language and the managerial practices that result from it can generate new ideas. These ideas lead to changes to the organization (new ways of proceeding and managing and, therefore, new knowledge and/or skills) or to its environment, resulting in a unique economic model (new ways of proceeding for the environment). Lastly, the change can be new for both the company and the environment. By considering CSR as an innovation, it is possible to mobilize ANT, in which the diffusion of an innovation is linked more to social actors than technical factors. Because its diffusion in the market is the result of multiple interactions between different actors involved, it is considered successful when a high number of actors show an interest in it and when it has shown resistance to the criticism of its opponents. However, it will fail if its opponents’ network is more convincing than that of its advocates.

The role of collective action is to voice the expectations of each member of the network (Walsh & Renaud, 2010) by selecting and shaping different perspectives and by choosing certain logical reasoning, certain statistical presentations, etc.:

Translation is therefore a simplification of reality through the articulation of problems that arise in pre-existing categories of meaning. It is a role that is primarily cognitive and language-related, which is at the heart of organizations as it allows for a reduction of contradictions. (Desmarais & Abord de Chatillon, 2010, p. 77)

The work of translation focuses on the definition of what Akrich et al. (1988a) call an ‘obligatory passage point. This obligatory passage point is a statement relating to the innovation, which represents the point of agreement between all parties. The implication of individuals in the development of an innovation is one of the possible consequences of incentive and enrollment mechanisms that some individuals manage to establish and sustain. The model of interessement allows us to highlight the existence of the relationships that bring together actors of organizational innovation and emphasizes how individual and collective interests align in the context of the development and diffusion of innovation (Chan & Reich, 2011).

In this study, the obligatory passage point occurs in the pre-institutionalization stage. This is when institutional work begins to make sense, as the main actor mobilizes a combination of justifications, resources, and social relationships (Hardy & Maguire, 2008) in order to shape the institution in his favor and, thus, ensures the diffusion of the innovation. The diffusion of an innovative CSR practice is studied in the light of the model of interessement, which enables the analysis of the conception of technical solutions and an understanding of ‘how the innovation is adopted, how it moves and progressively spreads to become successful’ (Akrich et al., 1988b, p. 15). According to these authors, ‘innovation is the art of sparking an interest in a growing number of allies in order to become stronger and stronger’ (Akrich et al., 1988b, p. 17). Thus, ‘adopting an innovation is adapting it’ and this adaptation generally results ‘from collective work, which is the fruit of an increasingly broad interest’ (Akrich et al., 1988b, p. 15). The elaboration and diffusion of the innovation takes place as soon as users are enrolled, i.e., actively participating in the collective action.

Table 1 shows the complementarity between the two theoretical approaches used to analyze the diffusion of a CSR practice within a cluster network.

Table 1. Complementarity between NIT and ANT for the analysis of the diffusion of a CSR practice

	NIT	ANT
Levels of interaction	Macro-environmental	Meso-environmental
Conception of the CSR	Emerging paradigm	Innovation
Initiating actor for change	Institutional entrepreneur	Translator
Stakeholders	Actors in the organizational field	Actors in the network
Innovation diffusion process	Linear design	Turbulent design
Innovation diffusion conditions	Moral and pragmatic legitimacy Institutional isomorphism	Obligatory passage point Enrollment and mobilization of allies
ANT, Actor-Network Theory; CSR, Corporate Social Responsibility; NIT, Neo-Institutional Theory.

It is in this context that we use both the NIT and ANT theoretical perspectives, which we consider to be complementary in understanding the diffusion process of a CSR practice.

NIT allows us to analyze an organization’s behavior at a macro-environmental level. The institutionalization of the CSR is not an innocuous change in practice as it refers specifically to social norms and existing regulation systems and is, therefore, embedded in its context (Marano & Kostova, 2016). Our results corroborate the work of authors who use this theoretical framework and show that the CSR institutionalization process inevitably accompanies a change of paradigm (Angelini et al., 2013). Our study is also in line with Waddock (2008) and Brammer, Jackson and Matten (2012), according to whom, a real CSR development depends on the ability of organization to create an institutional setting modeling the behavior of companies. This institutionalization process leads to the creation of a theoretical framework for inter-enterprise social responsibility embedded in an institutional setting (Frigant, 2015).

ANT enables analysis at a meso-environmental level through the decoding of network formation around an innovation practice. We agree with Lawrence and Suddaby (2006, p. 242), according to whom, ‘ANT is extremely promising when it comes to expanding the understanding of institutional work’ in relation to CSR. If the CSR is to be accepted as a new institution, its diffusion translates into the deployment of responsible systems (Slager et al., 2012). In light of this, the diffusion of CSR can be considered as being similar to the creation of an innovation through the introduction of a managerial object to support this new institution. By focusing on interactions at a meso-environmental level, ANT, in a dynamic way, takes account for the work achieved by the actors in charge of the development and the diffusion of CSR practices at an organizational level. ANT enables a more thorough analysis of the observed phenomenon than the microfoundations approach. Indeed, it is possible to understand how macro-environmental factors are influenced by actions and interaction at a microenvironmental level by linking the behavior of individuals and the emergence of collective and organizational results at a meso-environmental level.

Field and methodology

In this second part, we present the field of study, the objectives of the ENR label, and our methodology within the framework.

Presentation of France IT

In seeking to answer our research question, we studied the creation and diffusion of a socially responsible label: the ENR label. This label, created in 2011, is the result of collective action initiated by France IT, the French association of digital clusters.

There are several benefits of focusing on the digital sector. First, it is an economically dynamic sector. According to the Observatoire du Numérique (2015), the digital sector has over 97,000 companies with more than 10 employees. In 2010, Eurostats statistics showed that since the 1990s, the digital sector has been responsible for between 4 and 5% of economic growth in most European countries and for between 7 and 9% of U.S. GDP (Coe-Rexecode, 2011). This growth in the number of companies can be explained by the democratization of information communication and technology in the private and public sectors. Digital technology is also at the heart of the European strategy for employment and growth through the ‘Europe 2020’ plan and, particularly in France, as reflected in the granting in 2010 of €4.5 billion of the Grand Emprunt for future digital technology investments. The issue of CSR within the digital sector and its companies is, therefore, doubly interesting because, although digital practices can be the answer to many problems, they also have their share of questionable methods (planned obsolescence, power consumption of data centers, etc.).

The French association of digital clusters, France IT, was created in 2009 and is composed of 12 clusters. The association’s activity is structured around two main missions: to support the performance of clusters and to support the performance of the clusters’ member companies. Performance in the context of France IT should be understood broadly. France IT’s role with regard to clusters is to facilitate rapprochement, cooperation, and the development of collective projects. Seven of the 12 clusters are involved in this socially responsible collective labeling. Created in 2002 in the Loire region, the C1 cluster has 10 labeled companies out of 130 members. The C2 cluster grew mainly around the digital hub of Poitiers-Futuroscope, within which six companies have the ENR label. As a member of France IT since 2013, cluster C3 is based in the Limousin area and has three labeled companies. C4 is an Alsatian digital cluster, three of whose member companies have the label. Cluster C5 was founded in 2006 in the Rhone Valley, where four of the companies are labeled. It should be noted that the C5 and C1 clusters merged at the end of 2015. Founded in 2011 in the Midi-Pyrénées region, cluster C6 has three labeled companies. Finally, cluster C7 is located in the Lyon-Grenoble area, where one company has the ENR label.

Goals and targets of the ENR label

The ENR label was created to structure the digital sector around responsible practices and to provide oversight of provider/client relationships. Labeled companies want to use the label to legitimize their CSR practices in the eyes of their stakeholders, particularly those of their clients. They also want the label to be a tool for sharing socially responsible practices within the digital sector.

Since 2011, the ENR label has been available to cluster member companies. The label is structured around an evaluation of the applicant’s CSR practices and its stakeholder ethics. Every year, France IT invites all companies in the clusters to engage in a ‘labelling wave’. Each wave lasts a year and follows three main steps. First, a new wave is launched each February and runs until October, when companies must submit their applications to the Centre International de Ressources et d’Innovation pour le Développement Durable (CIRIDD). Second, companies complete a customer survey between October and December via a CIRIDD online platform, and the applications are then examined. Third, in January, applications are presented to the Comité d’Attribution du Label (CATLAB), which approves the granting of the label and arbitrates contentious cases. The CATLAB is composed of the Fondation Internet Nouvelle Génération (FING), Fédération des Syndicats des Métiers de la Prestation Intellectuelle, Conseil de l’Ingénierie et du Numérique (CINOV), Association Nationale des DSI (ANDSI), Association Française des Éditeurs de Logiciels et Solutions Internet, and Institut Mines Télécoms. Companies that are granted the label are allowed to use its logo in their communications for 2 years. The label is only valid for that period of time, after which companies must apply for renewal through a new labeling wave. Such companies must meet higher standards in the new wave, and the CATLAB pays particular attention to improvements identified during the first labeling period.

Presentation of the methodology

As highlighted by Perez and Rodriquez del Bosque (2012), case studies are specifically employed in the study of relative phenomena of CSR. The purpose of this study is to advance theoretical knowledge that articulates the diffusion of CSR, NIT, and ANT with new thinking that emerges from our empirical investigation.

We used three complementary research methods to triangulate our data to ensure its validity (Miles & Huberman, 2003). Between 2013 and 2015, we conducted semi-directed interviews (Bernard & Ryan, 2010) with several actors involved in the process at the stage when applications were considered and then following the results of the labeling waves (cf. Table 2). These interviews were conducted following interview guidelines, which evolved through three different versions until they were settled during the first stage of data coding. First, we held interviews with actors who had designed the label, such as the current President of Cluster C1, the former President of Cluster C1, and the Director of France IT. These actors were interviewed several times over the course of our investigation. The interviews were complemented by further interviews with leaders of digital Small and medium enterprise (SMEs), who had contributed to the drafting of the label. We also conducted interviews with leaders of labeled companies during the first four labeling waves, and with the CIRIDD auditor, members of the CATLAB, cluster project managers, and SME leaders in charge of their cluster’s governance body. We conducted a total of 46 interviews with 40 actors in the sector. The interviews, which lasted for an average of 90 min, were recorded and transcribed in their entirety. Data from these interviews were strengthened by observation (Denzin & Lincoln, 2000) during the examination of application files by the CIRIDD auditor, and by the collection of secondary data within these individual organizations (Yin, 2014). The data collected were processed in three stages following a coding method. We first chose our unit of analysis, in this case, a sentence or a group of sentences relating to a common theme (Bernard & Ryan, 2010). Then, we carried out a first layer of descriptive coding of the themes exposed. We then carried out a second layer of coding in order to construct a thematic tree ‘by regrouping summaries in a smaller amount of themes or in more synthetic conceptual elements’ (Miles & Huberman, 2003, p. 133) (Table 3).

Table 2. Actors interviewed

Clusters	Number of companies	Actors interviewed
Cluster C1	130	Labeled leaders (C1-A to C1-J), skeptical leaders (C1-N1 à C1-N3), project manager (chargé de mission) (CM-1)
Cluster C2	98	Labeled leaders (C2-A to C2-F), skeptical leader (C2-N1)
Cluster C3	64	Labeled leaders (C3-A to C3-C), project manager (CM-3)
Cluster C4	73	Labeled leaders (C4-1 to C4-D)
Cluster C5	47	Labeled leaders (C5-A to C5-D), cluster director (C5-P).
Cluster C6	170	Labeled leaders (C6-1), project manager (CM-6)
Cluster C7	130	Labeled leaders (C7-1), project manager (CM-7)
France IT		Director of France IT (AA1)
CIRIDD		Auditor (AA2)
CATLAB		Trade union representative (AA3)
CATLAB, Comité d’Attribution du Label; CIRIDD, Centre International de Ressources et d’Innovation pour le Développement Durable.

Table 3. Coding stages and data processing

Coding stage
Verbatim	Descriptive coding	Thematic tree construction coding
‘Given the size of our SME, we don’t have the capacity to integrate an ISO type label, or a label which would be completely oversized compared to our structure. So I thought it was interesting to have a label which is nonetheless of national scale’. (C2-A)	Benefit of the CSR label for SMEs	Sectoral certifications/positive image	Image Positive (IMA-POS)
‘I am going to ask a very simple question: what do we mean by sustainability for people developing software? I wonder how a company creating software can modify its behavior in order to be greener, apart from having it on paper’ (CN-A1)	Skepticism about sustainability in the context of the digital sector	CSR perception/digital sector/skepticism	RSE sector (RSE SEC)
Analysis stage
Vertical analysis per code	Presence/absence/temporality	IMA-POS RSE-SCE
Horizontal analysis per code	Confrontation	IMA-POS RSE-SCE

To analyze the development of the ENR label, we broke the process down into five iterative stages based on the first five stages of the socially responsible collective action: diagnosis of the sector, establishment of a task force to draft the label, launch of the program, expansion during the second labeling wave, and, finally, a trend toward stagnation during the last two labeling waves. In addition to labeling new companies, all labeling waves are characterized by the integration of new clusters into the process and by the emergence of new profiles in labeled companies. Throughout these five key stages, we analyzed the coded data on two levels using NVivo 10 software-QSR International. The use of thematic trees allowed us to improve our level of inference by identifying the relationships between each theme. We carried out vertical analysis to identify the main categories for each interview in order to spot the presence or absence of themes and their order of appearance. We then carried out horizontal analysis of our data. A theoretical dictionary played the role of interface between our theoretical framework and the phenomenon studied, allowing us to proceed to an iterative and progressive interpretation of the phenomenon observed (Bernard & Ryan, 2010).

Presentation and discussion of the findings

We employed a form of double analysis by combining stages of the institutional entrepreneurship process, originating from NIT, with stages of the innovation diffusion process within the network, as understood in ANT. As mentioned previously, the process of institutional entrepreneurship is composed of three stages: the pre-institutionalization stage, the theorizing stage, and the reinstitutionalization stage. The diffusion of innovation, as understood in ANT, is composed of four stages: the designation of the actor initiating the change, the identification of allies for the deployment of the innovation, the analysis of the diffusion process within the network, and the analysis of the conditions of diffusion. The following table synthesizes the findings of the analysis.

The results of our research led us to make six research proposals, which we discuss later (see below table 4).

Table 4. Complementary analysis of the ENR label according to NIT and ANT

	Pre-institutionalization	Theorization	Reinstitutionalization
Initiating actor and allies	Primum movens: The three presidents of the clusters ----------------------------------------------------- Member leaders of the clusters The CIRIDD (Proposal 1)		The translator: France IT --------------------------------- The evaluation committee of the label composed of CINOV IT, SYNTEC, AFEDL, ANDSI, the Institut Mines Telecom, the FING (Proposal 5)
Diffusion process	According to NIT: Linear design through the three stages According to ANT: Turbulent design through the three stages (Proposal 3)
Diffusion conditions	According to NIT: Legitimation of labeling process According to ANT: emergence of the controversy that starts the diffusion of the innovation (Proposal 2)	According to NIT: Institutional According to ANT: Obligatory passage point for the creation of the label (Proposal 4)	According to both NIT and ANT: Label credibility ensured by the effective enrollment of stakeholders involved in the diffusion of the innovation (Proposal 6)
ANT, actor-network theory; CIRIDD, Centre International de Ressources et d’Innovation pour le Développement Durable; CSR, corporate social responsibility; ENR, Entreprise Numérique Responsible; NIT, neo-institutional theory.

The pre-institutionalization stage

Building on NIT and, in particular, on the concept of institutional entrepreneurship, we note that the stages of the institutionalization process, as described by Greenwood et al. (2002) diagram I (namely pre-institutionalization, theorization, and reinstitutionalization), are present in the diffusion of the ENR label within France IT. At the pre-institutionalization stage, the actors who initiate institutional change are the cluster presidents and some company leaders. The comprehensive knowledge of the sector of the actors who initiate change is key to their understanding of the interests of their organizational field, which enables them to make an economic and strategic diagnosis of the digital sector and to communicate with the actors within it. Thus, these results confirm the work of Greenwood et al. (2002) in that these actors have several qualities, particularly legitimacy, within the organizational field. According to ANT, the actor who initiates the change is referred to as the primum movens. This actor is the person who initiates the change within the network, and whose primary mission is to diagnose the situation. He first makes a diagnosis related to the actor’s perception, because of an informational asymmetry between digital SMEs and their stakeholders:

Diagram 1. The institutionalization process. Adapted from Greenwood et al. (2002)

We’re dealing with completely abstract jobs so the buyer of the digital service doesn’t really understand what he is actually buying and what we are delivering. So we very often have buyers who are unhappy or concerned. (C1-D)

The analysis of our results also reveals several environmental and societal issues linked to the digital sector:

In IT, we deal with high power consumption, which increases our motivation, both environmentally and economically. The most sensitive point in this domain is the server rooms. (C1-J)

The second level of diagnosis focuses on existing CSR systems and their limitations:

We had a charter of good practice here. So of course we had 70 or 80 companies that signed the charter, but it’s all local, it’s not very gratifying […] Before the problem was that there were many charters ‘scattered’ over all the clusters, everyone had written their own individual charter. (AA1)

One of the challenges for the institutional entrepreneur is, therefore, how to use elements of the diagnosis to support their argument.

Furthermore, through a reading of NIT, this study shows that the actors who initiate change take account of the importance of mobilizing the stakeholders in their organizational field to secure the development of the innovation both internally and externally. Many actors will be mobilized and integrated into the process over the course of this CSR diffusion work. Actors in the digital sector use a two-level logic in the CSR institutionalization process: they mobilize actors in the organizational field by offering them this new institution, and they align the ‘project’ with the ‘agenda’ of the sector’s actors and with the existing ‘rationales’ within the organizational field.

We observed that during this pre-institutional stage, the initiating actors build relationships with institutional actors, particularly with the certifying third-party body, the Centre International de Ressources et d’Innovation pour le Développement Durable (CIRIDD), which then becomes a stakeholder in the labeling process. ANT considers stakeholders as ‘actants’, namely human actors participating in the process of co-construction of the ENR label. Actants are also nonhuman actors, for example, the France IT network or the ENR labeling system:

During the process, five SMEs work groups have been created. Then, rationalization of their ideas by consultant and diffusion to all the companies among the five groups in order to have their approval. (AA1)

At the pre-institutionalization stage, the actors at the center of the diffusion process are not yet recognized as actors of institutional work by their stakeholders. At this stage, few actors are aware of the imbalance within their organizational field, but many do not consider changing the situation:

With the dot-com bubble there were all sorts of people who took advantage of the scarcity of supply to offer products and services that were extremely overpriced with various degrees of quality. People who had little feedback, little clarity on what should be good service delivery rushed towards the nicest, most pleasant or charming person. It often went wrong. (C1-D)

Thus, in digital companies, future actors of institutional work start to seek a solution through their diagnosis in order to institutionalize socially responsible practices by developing a label.

With regard to dissemination conditions, NIT highlights the importance of legitimizing the labeling procedure in the eyes of the organizational field’s actors. ANT brings complementary elements to the conditions for the diffusion of innovation, which must be initiated through controversy. In our case, the controversy is analyzed through the diagnosis undertaken within the digital sector and raises the following question: Is it relevant to integrate CSR practices in this line of business? It is through controversy that the diffusion of innovation begins.

These findings lead us to present research proposals 1 and 2. (Table 4) Actors who initiate the change come from the organizational field and have a good understanding of it (Proposition 1). Our case highlights the importance of actors who initiate institutional change, and more precisely the ‘provenance’ of these actors. Indeed, while ANT underlines the expertise of the primum movens, NIT shows that the diagnosis and the first solutions offered to address a deteriorating institutional environment must come from the organizational field in question (Maguire et al., 2004). Expertise alone is not sufficient. Knowledge of actors and institutions in the organizational field is a critical requirement for the formulation of a diagnosis and the development of a common agenda to facilitate change. Thus, in proposing a relevant object (label, chart, audit, etc.), the CSR diffusion process must not be limited to technical issues. Rather, and more importantly, it requires an understanding of the issues raised by the actors and the institutions, so that the solutions appear legitimate while respecting the actors/institution equilibrium.

According to NIT and ANT, the institutionalization process and diffusion of the innovation have the same starting point, i.e., a jolt, an event that questions the established framework through an externally imposed change or through internal diagnosis (Proposition 2). The two theoretical frameworks overlap when considering the starting point of the pre-institutionalization process. For NIT, an event (jolt) first disrupts the equilibrium of the institutional framework; the event may be social, regulatory, or technological (Greenwood et al., 2002). Following this event, actors of institutional work offer solutions to restore the balance of the institutional framework. If we apply this NIT approach to the development of CSR, it means that the organizational field is destabilized by an event and CSR comes along as a solution to reconfigure the institutional framework through the development of an innovative object, such as a label or a certification. In ANT, however, the event that modifies the institutional framework is the emergence of an innovative system created within a developing network. Therefore, ANT consists of enhancing the social dimension of technical innovation and in identifying the different network actors and their respective roles in development of the innovation.

The theorization stage

In the theorization stage, the label begins to be diffused within the organizational field. The ‘specification’ translates into the converging of the collective and individual interests of those participating in the labeling process. In our case, this work is carried out by the three presidents of the clusters: i.e., initiators of change. The ‘justification’ is made possible by the work of legitimization undertaken by these same actors. We note that, from this step, the development of the label and the legitimization work no longer drive institutionalization of the CRS. They are an instrumentation process of CSR, through organizational innovation development. It is no longer simply the CSR that needs to be diffused, but also a socio-technical apparatus. The labeling apparatus, when seen as a socio-technical object, implies a broadening of the diffusion. For that reason, it is useful to use ANT to understand the internal diffusion of this socio-technical system by the network’s actors.

The stakeholders are the third certifying, and SMEs leaders participate in working groups which define and draft the ENR label standards:

The collective action focused on achieving sustainable development, in partnership with the CIRIDD: four companies in the cluster thus took part in a sustainable marketing themed operation. (AA1)

At this stage of its development, the innovation diffusion process is sequential, according to NIT, because two distinct phases are identified: ‘specification’ and ‘legitimization’. According to ANT, this is an iterative loop process of constant innovation, structured in such a way that interests and stakeholders are regularly reexamined in phases of compromise. The success of this process depends on the emergence of new actors, who can help in the translation of new requirements specific to CSR programs (Slager et al., 2012). These actors and their discourse become the link between the institutional equilibrium at a macro level and the approval of solutions and socio-technical systems at a meso/microenvironmental level.

Following NIT, the creation of a label as a vehicle for CSR institutionalization can be explained by mechanisms of institutional isomorphism (DiMaggio & Powell, 1983). As the actors’ goal is to influence the practices of digital businesses, the label offers them a way to create a standard within the sector, providing a mutual frame of reference for the actors (normative isomorphism). The introduction of the label by pioneer companies can influence other companies to follow suit, to become labeled themselves, and, by extension, to modify their practices (mimetic isomorphism). Over time, it is possible that the label and the practices relating to it will become a condition of access to certain markets by ordering parties and clients (coercive isomorphism). According to ANT, the creation of a label then becomes an obligatory passage point to establish this a national collective of actors.

Two research proposals emerge from our findings of the analysis of the theorization stage. Contrary to what is suggested by the classical approach to institutionalization, which is founded on NIT, parts of the process are turbulent and nonsequential (Proposition 3). The development of the label and the legitimization work are integrated into the process of co-construction of the network around the innovative CSR approach. According to the classical NIT approach, the institutionalization process is composed of linear stages (Greenwood et al., 2002; Lawrence et al., 2001). Our analysis, which combines NIT with ANT, enables us to highlight the turbulent nature of some of the institutionalization process of a CSR label. In the pre-institutionalization stage, the diagnosis made by some actors in the organizational field, the development of solutions recognized as acceptable, and the mobilization of strategic resources are in line with what Greenwood et al. (2002) describe as a sequential process. However, the theorization stage appears to be more in the form of a turbulent design. Indeed, the theorization work does not follow a simple linear progression but appears to take the form of an increasing broadening of the network and its social relationships, founded on the investment of internal and external stakeholders (Table 4).

At this stage in the institutionalization process, the search for organizational isomorphism and the creation of an obligatory passage point come together (Proposition 4). This proposal is at the heart of the CSR diffusion process, and it is at this stage that both theoretical frameworks become complementary and strengthen the analysis of the diffusion of the labeling apparatus. Indeed, actors who are central to the diffusion process ensure the presence of an organizational isomorphism (DiMaggio & Powell, 1983). The work in this context consists of defining translatable points with the interests of the organizational field’s actors (Hardy & Maguire, 2008; Maguire et al., 2004; Zilber, 2009). If the network is in development, it must be strengthened and made irreversible. It is then appropriate to proceed to enroll actors, assigning precise roles to all of those who join the network. Enrollment requires determination and the stabilization of tasks attributed to each party. It consists of explaining to the different actors in the network what they must do, in line with their respective interests, to obtain their membership. The incentives are successful once the actors accept, after a series of negotiations, the role that was assigned to them. Henceforth, ‘the relevance of a change within an organization depends on the level of cohesion it generates’ (Amblard et al., 1996, p. 139). The network progressively becomes stronger. Power games – involving spokespeople, problem solving, and middlemen – fade away, and relationships are strengthened. Actions and choices accepted by the collective become ‘black boxes’ that are no longer debated. As incentive and enrollment are renewed, the network’s influence grows:

At the center, a group of subjects and objects that have been gathered around an obligatory passage point thanks to a translation operation […]. New entities must be added to the core to make it stronger, whilst strengthening the project which both carries and is carried by the network. (Amblard et al., 1996, p. 163)

The network is expanded by triggering an interest in the innovation in new people and by finding new ways to unify them around the project for change. It means simultaneously multiplying entities (structures, deals, and means), composing the network, and creating an interest in actors to have a strategic role (opinion leaders, financial administrator, etc.). The more developed the network becomes, the more the innovation becomes irreversible. The growing number of actors involved in and of resources allocated to the innovation leads to an increase in the coordination efforts and administrative action within the network. Step by step, the innovation takes shape. Throughout this process, what is perceived as new or aiming for change becomes an operating mode. The action of the translator, essential at the start of the process, becomes less and less necessary (Table 4).

The reinstitutionalization stage

During the reinstitutionalization stage, we observe that as leaders take on the label, the initiating actors withdraw (President and former President of cluster C1). At this stage, the labeling process has been launched, and it is appropriate for France IT and other actors to lead the rest of the collective work. France IT then assumes the role of translator within the network:

We did the preparatory work for this label, but afterwards for ethical reasons, the work was passed onto France IT. (C1-D)

The diffusion of the ENR label, considered to be an organizational innovation, cannot be achieved without the investment of the network’s stakeholders. The challenge for actors, according to Akrich et al. (1988b), is to attract a growing number of actors who become legitimate and who are recognized as spokespersons. France IT’s assessment committee is composed of several institutions, which has helped the network to expand and makes the innovation irreversible:

It remains an external body because it is composed of CINOV IT, the other digital professional union. There is SYNTEX, the AFEDL (Association Française des Éditeurs de Logiciels), the ANDSI (Association Nationale des DSI), the Institut Mines Telecom, the FING (Fondation Internet Nouvelle Génération). (AA1)

During the third and fourth labeling wave, the label’s diffusion process stops expanding. This observation is shared by the director of France IT, who notices a stagnation in the number of labeled groups:

We are in a phase of growth, which is too slow, I am a sort of impatient guy, and to achieve this national communication, we must have a threshold. (AA1)

This leads to actors who are skeptical about the label feeling able to openly speak up:

What do we mean by sustainable development for people who develop software? I wonder how a company developing software can modify its behavior in order to become greener, apart from having it on paper. (C1-N1)

Contrary to the classical approach to the institutionalization phenomenon, in this final stage, the process is turbulent and nonsequential. We observe that the process is characterized by overlapping phases, and even by a turbulent dynamic which aims to create a network inside the sector.

This disengagement can be explained by the nature of the apparatus itself. By relying on the label, conceptors have to launch collective work on an apparatus, which requires a high level of credibility. The label is granted this credibility through the working groups who drafted the terms of reference, the examination of applications carried out by a third party, and the approval of the CATLAB. Yet, to ensure this credibility, conceptors must also be independent from these organizations. Thus, they MUST withdraw in order to ensure this credibility in the eyes of companies in the sector and of their stakeholders. Withdrawing conceptors in this way ‘freezes’ the labeling apparatus. According to ANT, this withdrawal shows that the initiating actors and primum movens were successful in enrolling stakeholders and getting them to invest personally in the development of the apparatus and contribute to the extension of the network, thus making the innovative apparatus irreversible.

These findings lead to the formulation of two research proposals (Table 4). The process of withdrawing initiating actors can be explained by the process of reinstitutionalization and by the socio-technical system (Proposition 5). Greenwood et al. (2002) explain that during the reinstitutionalization stage, actors gradually withdraw as soon as the actors in the organizational field digest and repeat practices relative to the new institution. While this approach to the institutionalization process is relevant, it does not enable analysis of how institutionalization takes place through apparatus, which implies the disengagement of the initiators. In the case of a socially responsible labeling apparatus, those who initiated the apparatus may be expected to withdraw from the first stages of diffusion, so that the developers of the label are not perceived to be both judge and executioner. This approach, centered on the socio-technical system, is explained by ANT, which helps in understanding how the innovation is developed inside the network. The actors’ attempts to introduce, at a collective level, the ENR label within France IT are examined according to every possible outcome: success, hijack, or completely dropping the responsible labeling apparatus. According to ANT, the object that becomes a vector of CSR is not neutral in the diffusion stages. It influences the relationship of actors at the center of the diffusion process. Thus, while in Proposition 1, we underline that the qualities of the socio-technical system appear to be secondary in the pre-institutionalization stage and that they should not influence the actors/institution equilibrium, we observe the opposite relationship in the reinstitutionalization stage. Once this equilibrium is achieved and the institutionalization is established, the socio-technical system starts the process of the more-or-less gradual implication and withdrawal of initiating actors.

Actors who initiate the CSR diffusion process only mobilize stakeholders in the organizational field, whereas the labeling apparatus, understood as a socio-technical object, involves a broadening of the diffusion to a wider spectrum of actors (Proposition 6). The only analysis grid that NIT can offer for understanding the CSR institutionalization process, therefore, appears incomplete. Our case study shows that actors who initiate the institutional work are focused on making their solution legitimate in the eyes of their organizational field. This requires the development of a common agenda with the actors in the organizational field within which institutional actors attempt to disseminate the labeling apparatus. ANT enables us to complete the analysis of the diffusion of the CSR labeling apparatus, along with the action of actors which goes beyond the frame of the organizational field. Indeed, the CSR institutionalization process goes beyond the organizational field for two reasons. First, the social responsibility of organizations is not limited to the problems and interests of a single line of business. Second, it goes beyond the borders of the sector by involving stakeholders in a broader way and by questioning organizations about their interactions with society.

The diffusion of the CSR must also be demonstrated to stakeholders. Thus, as shown by our analytical results, the diffusion depends on management objects such as the label. This object requires a certain visibility and credibility for a growing number of stakeholders. Therefore, while the quest for moral and pragmatic credibility (Suchman, 1995) of a CSR practice necessarily involves discursive institutional work (Hardy & Maguire, 2008) inside the organizational field and is, therefore, a precondition for the success of the diffusion of a practice, the enrollment of the stakeholders is the logical and indispensable continuation of this. A sequential reading of the diffusion process can, therefore, lead actors to focus on a frame that constitutes the organizational field. NIT has shown through our research proposals the need to consider the existing institutional frame. ANT completes this theory by enabling the analysis of the co-construction of a management system led by actants. It also enables the choice of object used to diffuse CSR principles to be taken into account. Diagram 2 details the research proposals that result from the analysis of our findings.

Diagram 2. Model resulting from findings

Discussion and conclusion

We began with the observation that the diffusion of a CSR practice within an organizational field is mostly performed following NIT or ANT and is, therefore, limited to analysis at a macro- or microenvironmental level without sufficient bridging of these two levels of analysis. This study, carried out inside a network of organizations, allows this link to be made by focusing on how interactions between individuals lead to the emergence of collective and organizational results, and how relationships between macro-environmental factors are influenced by the actions and interactions that happen at a microenvironmental level (Abell et al., 2008).

The findings from our analysis of the diffusion of the ENR label within France IT lead us, in this section, to propose and discuss two contributions. The first is a proposal for an analysis grid which provides a dynamic decryption of the diffusion process of a CSR practice from a macro- to a meso-environmental level. The second contribution consists of proving that discursive institutional work enables both levels of analysis of the diffusion of a CSR practice inside an organizational field to be bridged.

First, the analysis of our findings shows that it is appropriate, in the specific context of CSR, to use both NIT and the ANT to understand the complexity of the diffusion process. The reason for using this theoretical framework is that CSR practice cannot be seen only as a socio-technical object. If we approach that practice from the standpoint of a new institution within the organizational field, then the steps of the diffusion process can be explained according to the sequential approach described by Greenwood et al. (2002). CSR principles are developed in three stages: pre-institutionalization, theorization, and reinstitutionalization. However, the institutionalization process, as envisaged by the actors who initiate the project, also relies on technical apparatus. Thus, ANT is necessary for analyzing the diffusion of the socio-technical apparatus within the network. The diffusion of the CRS practice is undertaken in four stages: the designation of the actor initiating change, identification of allies of deployment of the innovation, analysis of the diffusion process inside the network, and analysis of the diffusion conditions.

The second contribution relates to the discourse on institutional work (Hardy & Maguire, 2008). Our findings show that the institutional work inside France IT is based on the combination of the three levers proposed by Hardy and Maguire (2008): justification, strategic resources, and social relationships. The importance and gradual integration, through discourse, of social relationships is made explicit via the creation of allied groups (Battilana et al., 2009) and by effective persuasion of actors who feel threatened by the change (Chambat, 1994). This forming of alliances, defined by Hampel, Lawrence, and Tracey (2017) as the relational part of institutional work, requires the development of a network of actors in order to amplify motion and intensity within institutional work (Bertels et al., 2014), so that the solutions proposed appear to be the only acceptable option inside the organizational field. This also requires the mobilization of material and immaterial resources, as shown by ANT theorists via the development of networks around primum movens (Callon & Latour, 2006) and by NIT theorists through the material dimension of institutional work (Hampel et al., 2017). It is through building networks around discourse that the two approaches are complementary. For NIT theorists, institutional actors build and propose a persuasive discourse around signs and symbols (Ruebottom, 2013), similar to what is observed in the development of a CSR label. Following the microfoundations approach, this macroenvironmental phenomenon can also be explained by the interactions that happen at a microenvironmental level. Thus, we complete this reflection through ANT and by integrating the idea that, as long as obligatory passage points are respected, discourse can be distorted and adapted by actors of the network.

Thus, the discursive approach to institutional work becomes the link between the NIT and the ANT theoretical framework, which enables us to propose an analysis grid that combines institutional and socio-technical interests. The phenomenon of CSR institutionalization is the basis for questioning inside the organizational field while raising the question of the diffusion of a socio-technical system, such as a certification system. We put forward the idea that from an academic and a managerial point of view, it is relevant to consider this type of institutionalization as a complex phenomenon. This phenomenon simultaneously requires taking account of interests related to institutional pressures and to the legitimacy of actors as well as of the social relationships within an organizational field.

From a managerial point of view, the main conclusions of this work provide organizations with an operating framework, which respects CSR principles and encourages actors to situate themselves with regard to their ‘outlook’ about the world and the purpose of their organization inside a line of business, as well as on a more global level. This raises questions about the scope and the diffusion of a labeling apparatus. The institutionalization process also involves the creation of a technical object, which becomes central to the formation and the development of the network surrounding the innovation. Furthermore, our findings show that the stagnation of the label diffusion process inside France IT is the consequence of several elements: the relevance of the apparatus and the interest of external actors in the labeling apparatus. Thus, failure lies first and foremost in the collective approach, which does not manage to bring together the wider project stakeholders and has to deal with the skepticism of some actors.

Moreover, despite incentives offered by the network (access to training programs and involvement in the creation of the label) and the positive consequences, not all the actors endorse the project because of how they perceive the nature of the label. For company leaders who mainly perceive the label as an organizational contribution, in terms of formalization and the development of new practices, the labeling campaign is a success. However, failure is more common for the actors who perceive the label only as a commercial tool, particularly those whose companies already have the label but who seek prominence by not renewing it. The apparatus we studied, thus, faces several limits in its diffusion process. Here again, the combination of both theoretical frameworks allows us to consider the conditions of diffusion as a whole. Thus, the diffusion of a certification apparatus collides with preexisting norms within the digital organizational field. The diffusion of the institutionalization project via a socio-technical object requires existing institutions to be considered (moral and pragmatic legitimacy) along with the actors in the network. Furthermore, initiators must consider the environment so that the moral and pragmatic legitimacy of the CSR is recognized by the sector’s stakeholders. At the same time, it is necessary to meet everyone’s interests by creating an obligatory passage point, after debating the controversy with the network’s members.

With regard to the limitations of our research, the longitudinal case study started in 2012, when the collective labeling action had already begun. Thus, the first phase of label development was reconstituted a posteriori through the interpretation of actors who were directly involved. It is difficult to observe such a research field from its primary phase when actors are diagnosing their organizational field. Moreover, and as noted in our discussion of the findings, initiators of the label find themselves somewhat inside a process prior to its success. Lastly, this study focuses more on the elaboration and the diffusion of the labeling apparatus and less on the appropriation of CSR principles by company leaders.

However, these limitations constitute avenues for future research. First, consideration could be given to carrying out these interviews with leaders again in order to study the evolution of the labeling apparatus inside France IT, as well as the factors of success and failure in the establishment of socially responsible practices within their companies. In keeping with the work of Delacour and Liarte (2013), it would also be interesting to study the role of macro-actors who are central to the development and stability of the network, and to analyze whether they marginalized the actants, who they regarded as inconvenient to the stabilization of the network. Indeed, power dynamics between macro-actors and actants who are discarded to the periphery of the network can lead to resistance mechanisms, which partly explains the failure of the diffusion of an innovative practice inside a network. Finally, the work by Elbasha and Wright (2017) offers new perspectives, which could be developed in the context of this study. They recommend using Giddens’ structuration theory, along with Stones’ (2012) strong structuration theory. Like the NIT/ANT duo, these theories enable both macroenvironmental and meso-environmental analysis, to explain the changes that take place inside and between organizations. Employment of these two theoretical frameworks could then lead to a more precise analysis of the relationship between macrostructures and individual forces and would go beyond the study of strategic micro-behaviors. These different exploratory paths, which would complement the findings of this work, demonstrate the value of carrying out investigations in this fruitful field of research.

Acknowledgments

The authors would like to thank the anonymous reviewers, Vincent Mangematin and Thomas Roulet, for their helpful comments and valuable suggestions. The content of this paper, including any errors and omissions, remains the responsibility of the authors. This document is benefited from Ellen Editorial’s copy editing.

Biography

Arnaud Gautier is an associate professor at Université Côte d’Azur. His work focuses on interorganizational relationships, in particular in the context of SMEs. He is also working on the concept of shared value and the generation of shared value.

Elise Bonneveux is an associate professor in strategic management at Tours University, France, and is a member of the Vallorem Research Center. Her current work focuses on CSR in SMEs and the role of local networks. She also works on inclusive workplace practices.

References

Abell, P., Felin, T. & Foss, N. (2008). Building microfoundations for the routines, capabilities, and performance links. Managerial and Decision Economics, 29, 489–502. doi: 10.2139/ssrn.982823

Aggeri, F. (2017). How can performativity contribute to management and organization research? Theoretical perspectives and analytical framework. M@n@gement, 20(1), 28–69. doi: 10.3917/mana.201.0028

Akrich, M. (1993a). Les formes de la médiation technique. Réseaux, 60, 87–98. doi: 10.3406/reso.1993.2368

Akrich, M. (1993b). Les objets techniques et leurs utilisateurs. De la conception à l’action. In B. Conein, N. Dodier & L. Thévenot (Eds.), Les objets dans l’action (vol. 4, pp. 35–57). Éditions de l’EHESS, Collection Raisons Pratiques.

Akrich, M. & Callon, M. (2003). The key to success in innovation Part II: The art of choosing good spokespersons. International Journal of Innovation Management, 6(2), 207–225.

Akrich, M., Callon, M. & Latour, B. (1988a). A quoi tient le succès des innovations, premier épisode: L’art de l’intéressement. Annales des Mines, série “Gérer et comprendre”, 11, 4–17.

Akrich, M., Callon, M. & Latour, B. (1988b). A quoi tient le succès des innovations, deuxième épisode: L’art de choisir les bons porte-paroles. Annales des Mines, série “Gérer et comprendre”, 12, 14–29.

Akrich, M., Callon, M. & Latour, B. (2006). Sociologie de la traduction: Textes fondateurs. Ecole des mines de Paris.

Amblard, H., Bernoux, P., Herreros, G. & Livian, Y.-F. (1996). Les nouvelles approches sociologiques des organisations (3rd ed.). Edition du Seuil.

Angelini, C., Lombardo, E. & Pignatel, I. (2013). La diffusion du concept de RSE dans les firmes industrielles: l’émergence de nouveaux paradigmes managériaux. Recherches en Sciences de Gestion, 96(3), 131–151. doi: 10.3917/resg.096.0131

Battilana, J., Leca, B. & Boxenbaum, E. (2009). How actors change institutions: Toward a theory of institutional entrepreneurship. Academy of Management Annals, 3(1), 65–107. doi: 10.1080/19416520903053598

Ben Slimane, K. (2012). Retourner sa veste, toujours du bon côté: Travail institutionnel discursif dans le déploiement de la télévision numérique terrestre en France. M@n@gement, 15(2), 145–179. doi: 10.3917/mana.152.0146

Ben Slimane, K. & Leca, B. (2010). Le travail institutionnel: Origines théoriques, défis et perspectives. Management & Avenir, 7(37), 53–69. doi: 10.3917/mav.037.0053

Bernard, R. H. & Ryan, G. W. (2010). Analyzing qualitative data. Systematic approaches. Sage.

Bertels, S., Hoffman, A. J. & DeJordy, R. (2014). The varied work of challenger movements: Identifying challenger roles in the US environmental movement. Organization Studies, 35(8), 1171–1210. doi: 10.1177/0170840613517601

Brammer, S., Jackson, G. & Matten, D. (2012). Corporate social responsibility and institutional theory: New perspectives on private governance. Socio-economic Review, 10, 3–28. doi: 10.1093/ser/mwr030

Brundtland, G. H. (1987). Report of the world commission on environment and development: Our common future. United Nations General Assembly Document A/42/427.

Callon, M. (1986). Eléments pour une sociologie de la traduction, la domestication des coquilles Saint-Jacques et des marins pêcheurs de la baie de Saint-Brieuc. L’Année Sociologique, 36, 169–208.

Callon, M. (1998). Introduction: The embeddedness of economic markets in economics. The Sociological Review, 46(1), 1–57. doi: 10.1111/j.1467-954X.1998.tb03468.x

Callon, M. (2006). Sociologie de l’acteur réseau. In M. Akrich, M. Callon & B. Latour (Eds.), Sociologie de la traduction: Textes fondateurs (267–276). Presses des Mines.

Callon, M. & Latour, B. (1986). Les paradoxes de la modernité. Comment concevoir les innovations? Prospective et Santé, 36, 13–25.

Callon, M. & Latour, B. (2006). Le grand Léviathan s’apprivoise-t-il ? In M. Akrich, M. Callon & B. Latour (Eds.), Sociologie de la traduction. Textes fondateurs (pp. 11–32). Les Presses de l’École des Mines.

Chambat, P. (1994). NTIC et représentation des usagers. In A. Vitalis (Ed.), Médias et nouvelles technologies. Pour une socio-politique des usages (pp. 45–59). Apogée.

Chan, Y. & Reich, B. (2011). Rethinking business-IT alignment. In R. Galliers & W. Currie (Eds.), Management information systems, critical perspectives and new directions (pp. 81–98). Oxford University Press. doi: 10.1093/oxfordhb/9780199580583.001.0001

Coe-Rexecode. (2011). L’économie numérique et la croissance : Résumé et principales conclusions. Document de travail n°24. Coe-Rexecode.

Coleman, J. S. (1987). Microfoundations and macrosocial behavior. In J. C. Alexander, B. Giesen, R. Münch & N. J. Smelser (Eds.), The micro-macro link (pp. 153–173). University of California Press.

Coleman, J. S. (1990). Foundations of social theory. Belknap Press of Harvard University Press. doi: 10.2307/1962898

Dacin, T. M., Goodstein, J. & Scott, W. R. (2002). Institutional theory and institutional change: Introduction to the special research forum. Academy of Management Journal, 45(1), 45–56. doi: 10.2307/3069284

Damanpour, F., Walker, R. M. & Avellaneda, C. N. (2009). Combinative effects of innovation types and organizational performance: A longitudinal study of service organizations. Journal of Management Studies, 46(4), 650–675. doi: 10.1111/j.1467-6486.2008.00814.x

Daudigeos, T., Boutinot, A. & Jaumier, S. (2015). The historical study of institutional change over long periods: Pitfalls and perspectives. A commentary on the article by Hélène Peton and Stéphan Pezé. M@n@gement, 18(3), 254–260. doi: 10.3917/mana.183.0254

Delacour, H. & Liarte, S. (2013). Je t’aime… Moi non plus! De la marginalisation à l’intégration de l’industrie du cinéma pornographique dans le réseau Blu-Ray. M@n@gement, 16(3), 238–263.

Denzin, N. K. & Lincoln, Y. S. (2000). Handbook of qualitative research. Sage. doi: 10.1111/j.1365-2648.2001.0472a.x

Desmarais, C. & Abord de Chatillon, E. (2010). Le rôle de traduction du manager: Entre allégeance et résistance. Revue Française de Gestion, 205, 71–88. doi: 10.3166/rfg.205.71-88

Desreumaux, A. (2004). Théorie néo-institutionnelle, management stratégique et dynamique des organisations. In I. Huault (coord.), Institutions et gestion (pp. 29–47). Vuibert.

DiMaggio, P. & Powell, W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48, 147–160. doi: 10.2307/2095101

DiMaggio, P. J. (1988). Interest and agency in institutional theory. In L. Zucker (Ed.), Institutional patterns and organizations (pp. 3–22). Ballinger.

Dorado, S. (2005). Institutional entrepreneurship, partaking, and convening. Organization Studies, 26(3), 385–414. doi: 10.1177/0170840605050873

Dupuis, J.-C. & Le Bas, C. (2009). Analyser le processus d’institutionnalisation de la responsabilité sociale des entreprises: L’apport de trois approches institutionnalistes. Revue Française de Socio-Économie, 2(4), 83–98. doi: 10.3917/rfse.004.0083

Eberhard-Harribey, L. (2006). Corporate social responsibility as a new paradigm in the European policy: How CSR comes to legitimate the European regulation process. Corporate Governance: The International Journal of Business in Society, 6(4), 358–368. doi: 10.1108/14720700610689487

Elbasha, T. & Wright, A. (2017). Reconciling structure and agency in strategy-as-practice research: Towards a strong-structuration theory approach. M@n@gement, 20(2), 107–128. doi: 10.3917/mana.202.0107

Elkington, J. (1997). Cannibals with forks. The triple bottom line of 21st century business. Capstone. doi: 10.1023/A:1006129603978

Flichy, P. (1995). L’action dans un cadre sociotechnique. Comment articuler technique et usage dans une même analyse? In J.-G. Lacroix & G. Tremblay (Eds.), Les autoroutes de l’information, un produit de la convergence (pp. 405–433). Presses de l’Université du Québec.

Frigant, V. (2015). Beyond the business case and sustainable chain management: Why do we need to build a theory of interfirm social responsibility? M@n@gement, 18(3), 234–253. doi: 10.3917/mana.183.0234

Gondo, M. B. & Amis, J. (2013). Variations in practice adoption: The roles of conscious reflection and discourse. Academy of Management Review, 38(2), 229–247. doi: 10.5465/amr.2010.0312

Greenwood, R., Suddaby, R. & Hinings, C. R. (2002). Theorizing change: The role of professional associations in the transformation of institutionalized fields. Academy of Management Journal, 45(1), 58–80. doi: 10.2307/3069285

Hampel, C. E., Lawrence, T. B. & Tracey, P. (2017). Institutional work: Taking stock and making it matter. In R. Greenwood, C. Oliver, T. B. Lawrence & R. E. Meyer (Eds.), The Sage handbook of organizational institutionalism (pp. 558–590). Sage. doi: 10.4135/9781526415066

Hardy, C. & Maguire, S. (2008). Institutional entrepreneurship. In R. Greenwood (Ed.), The Sage handbook of organizational institutionalism (pp. 198–217). Sage. doi: 10.4135/9781446280669.n11

Harmon, D., Haack, P. & Roulet, T. (2018). Microfoundations of institutions: A matter of structure vs. agency or level of analysis? Academy of Management Review, 43, 132–155. doi: 10.5465/amr.2018.0080

Helms, W. S., Oliver, C. & Webb, K. (2012). Antecedents of settlement on a new institutional practice: Negotiation of the ISO 26000 standard on social responsibility. Academy of Management Journal, 55(5), 1120–1145. doi: 10.5465/amj.2010.1045

Hoffman, A. J. (1999). Institutional evolution and change: Environmentalism and the U.S. chemical industry. Academy of Management Journal, 42(4), 351–371. doi: 10.5465/257008

Huault, I. (2017). Paul DiMaggio et Walter W. Powell – Des organisations en quête de légitimité. In S. Charreire-Petit (Ed.), Les Grands Auteurs en Management (pp. 166–181). EMS Editions. doi: 10.3917/ems.charr.2017.01

Kilbourne, W. E. (2004). Sustainable communication and the dominant social paradigm: Can they be integrated? Marketing Theory, 4(3), 187–208. doi: 10.1177/1470593104045536

Latour, B. (1989). La science en action. Gallimard.

Latour, B. & Wooglar, S. (1979). Laboratory life: The social construction of scientific facts. Sage.

Lawrence, T. B. & Suddaby, R. (2006). Institutions and institutional work. In S. Clegg, C. Hardy, W. R. Nord & T. B. Lawrence (Eds.), Handbook of organizations studies (pp. 215–254). Sage.

Lawrence, T. B., Suddaby, R. & Leca, B. (2011). Institutional work: Refocusing institutional studies of organization. Journal of Management Inquiry, 20(1), 52–58. doi: 10.1177/1056492610387222

Lawrence, T. B., Winn, M. I. & Devereaux Jennings, P. (2001). The temporal dynamics of institutionalization. Academy of Management Review, 26(4), 624–644. doi: 10.5465/amr.2001.5393901

Lenox, M. & Nash, J. (2003). Industry self-regulation and adverse selection: A comparison across four trade association programs. Business, Strategy and Environment, 12(6), 343–356. doi: 10.1002/bse.380

Levy, D. & Scully, M. (2007). The institutional entrepreneur as Modern Prince: The strategic face of power in contested fields. Organization Studies, 28(7), 971–991. doi: 10.1177/0170840607078109

Maguire, S. & Hardy, C. (2009). Discourse and deinstitutionnalization: The decline of DDT. Academy of Management Journal, 52(1), 148–178. doi: 10.5465/AMJ.2009.36461993

Maguire, S., Hardy, C. & Lawrence, T. B. (2004). Institutional entrepreneurship in emerging fields: HIV/AIDS treatment advocacy in Canada. Academy of Management Journal, 47(5), 657–679. doi: 10.5465/20159610

Marais, M. (2014). Quelle marge de manœuvre pour l’engagement socialement responsable des dirigeants? Enjeux et perspectives du gouvernement d’entreprise. M@n@gement, 17(4), 237–265. doi: 10.3917/mana.174.0237

Marano, V. & Kostova, T. (2016). Unpacking the institutional complexity in adoption of CSR practices in multinational enterprises. Journal of Management Studies, 53(1), 28–54. doi: 10.1111/joms.12124

Meyer, J. W. & Rowan, B. (1991). Institutionalized organizations: Formal structure as myth and ceremony. In W. W. Powell & P. J. DiMaggio (Eds.), The new institutionalism in organizational analysis (pp. 41–62). The University of Chicago Press.

Miles, M. & Huberman, A. M. (2003). Analyse des données qualitatives. De Boeck Université.

Norman, W. (2011). Business ethics as self-regulation: Why principles that ground regulations should be used to ground beyond-compliance norms as well. Journal of Business Ethics, 102, 43–57. doi: 10.1007/s10551-011-1193-2

Observatoire du numérique. (2015). Observatoire du numérique, Chiffres clés 2014-Online document. Retrieved from http://www.observatoire-du-numerique.fr/chiffres-cles-2014

Oliver, C. (1990). Determinants of interorganizational relationships: Integration and future directions. Academy of Management Review, 15(2), 241–265. doi: 10.5465/amr.1990.4308156

O’Rourke, D. (2006). Multi-stakeholder regulation: Privatizing or socializing global labor standards? World Development, 34(5), 899–918. doi: 10.1016/j.worlddev.2005.04

Palmer, K., Oates, W. & Portney, P. (1995). Tightening environmental standards: The benefit-cost or the no-cost paradigm? Journal of Economic Perspectives, 9, 119–133. doi: 10.1257/jep.9.4.119

Pascal, A. (2006). Conception d’une solution TIC pour favoriser l’émergence de projets innovants: Une approche par les usages – L’expérience KMP. Thèse de Doctorat en Sciences de Gestion, Université de Nice Sophia Antipolis.

Perez, A. & Rodriquez del Bosque, I. (2012). The role of CSR in the corporate identity of banking service providers. Journal of Business Ethics, 108(2), 145–166. doi: 10.1007/s10551-011-1067-7

Phillips, N., Lawrence, T. B. & Hardy, C. (2000). Inter-organizational collaboration and the dynamics of institutional fields. Journal of Management Studies, 7(2), 23–43. doi: 10.1111/1467-6486.00171

Postel, N. & Rousseau, S. (2008). RSE et éthique d’entreprise: la nécessité des institutions. M@n@gement, 11(2), 137–160. doi: 10.3917/mana.112.0137

Powell, W. W. & Colyvas, J. A. (2008). Microfoundations of institutional theory. In R. Greenwood, C. Oliver, R. Suddaby & K. Sahlin (Eds.), The Sage handbook of organizational institutionalism (pp. 276–298). Sage. doi: 10.4135/9781849200387.n11

Prado, A. M. (2013). Competition among self-regulatory institutions: Sustainability certifications in the cut-flower industry. Business & Society, 52(4), 686–707. doi: 10.1177/0007650313493990

Reay, T. & Hinings, C. R. (2005). The recomposition of an organizational field: Health care in Alberta. Organization Studies, 26(3), 351–384. doi: 10.1177/0170840605050872

Ruebottom, T. (2013). The microstructures of rhetorical strategy in social entrepreneurship: Building legitimacy through heroes and villains. Journal of Business Venturing, 28(1), 98–116. doi: 10.1016/j.jbusvent.2011.05.001

Scott, W. R. (1991). Unpacking institutional arguments. In P. J. DiMaggio & W. W. Powell (Eds.), The new institutionalism in organizational analysis (pp. 164–182). The University of Chicago Press.

Slager, R., Gond, J.-P. & Moon, J. (2012). Standardization as institutional work: The regulatory power of a responsible investment standard. Organization Studies, 33(5–6), 763–790. doi: 10.1177/0170840612443628

Steen, J., Coopmans, C. & Whyte, J. (2006). Structure and agency? Actor-network theory and strategic organization. Strategic Organization, 4(3), 303–312. doi: 10.1177/1476127006067033

Steen, J. T. & Liesch, W. L. (2007). A note on Penrosian growth, resource bundles and the Uppsala model of internationalization. Management International Review, 47(2), 193–206. doi: 10.1007/s11575-007-0011-7

Stones, R. (2012). Causality, contextual frames and international migration: Combining strong structuration theory, critical realism and textual analysis. Working Papers, International Migration Institute. University of Oxford.

Suchman, M. C. (1995). Managing legitimacy: Strategic and institutional approaches. Academy of Management Review, 20(3), 571–610. doi: 10.2307/258788

Tolbert, P. S. & Zucker, L. G. (1983). Institutional sources of change in the formal structure of organizations: The diffusion of civil service reform. Administrative Science Quarterly, 28(1), 22–39. doi: 10.2307/2392383

Waddock, S. (2008). Building a new institutional infrastructure for corporate responsibility. Academy of Management Perspectives, 22(3), 87–108. doi: 10.5465/amp.2008.34587997

Walsh, I. & Renaud, A. (2010). La théorie de la traduction revisitée ou la conduite du changement traduit. Application à un cas de fusion-acquisition nécessitant un changement de système d’information. Revue Management et Avenir, 9(39), 283–302. doi: 10.3917/mav.039.0283

Wood, D. J. (1991). Corporate social performance revisited. Academy of Management Review, 16(4), 691–718. doi: 10.2307/258977

Yin, J. & Zhang, Y. (2012). Institutional dynamics and corporate social responsibility in an emerging country context: Evidence from China. Journal of Business Ethics, 111(2), 301–316. doi: 10.1007/s10551-012-1243-4

Yin, R. K. (2014). Case study research. Design and methods (5th ed.). Sage. doi: 10.3138/cjpe.30.1.108

Zietsma, C. & Lawrence, T. B. (2010). Institutional work in the transformation of an organizational field: The interplay of boundary work and practice work. Administrative Science Quarterly, 55(2), 189–221. doi: 10.2189/asqu.2010.55.2.189

Zilber, T. B. (2009). Institutional maintenance as narrative acts. In T. B. Lawrence, R. Suddaby & B. Leca (Eds.), Institutional work: Actors and agency in institutional studies of organizations (pp. 205–235). Cambridge University Press. doi: 10.17/CBO9780511596605

Footnote

¹. Entreprise Numérique Responsable, translated as Responsible Digital Corporation.