Abstract
International Framework Agreements negotiated between Global Union Federations and multinational companies are a major innovation which bring collective bargaining to an international level. We focus on the role of the union negotiators and analyse two cases, drawing on studies of innovation processes in business organizations. We identify types of learning processes and the exchange and creation of knowledge, and examine relations inside international trade union networks. We find that two conditions are necessary to develop an innovatory role: international unions must coordinate exchanges between different communities, holders and creators of specific union knowledge, and must also facilitate the dissemination of knowledge by maintaining the regularity and quality of exchanges between network members.
Keywords
Introduction
For two decades, a new global negotiation space has been under construction with the concluding of International Framework Agreements (IFAs), signed by multinational companies (MNCs) and Global Union Federations (GUFs). The aim is for worldwide implementation inside the company borders, possibly including suppliers and subcontractors. IFAs generally deal with fundamental labour rights and refer to ILO Conventions (Hammer, 2005; Sydow et al., 2014). This negotiation space is significantly expanding, and an innovatory process is developing. Bringing collective bargaining to an international level is itself a major innovation, as the negotiators find new interests in regulating labour standards and invent new ways to negotiate.
The role of GUFs in negotiating and implementing IFAs has been widely researched (Ford and Gillan, 2015; Papadakis, 2008), but few authors have analysed these as a type of innovation and through a knowledge perspective (Bourque, 2008). Indeed, GUFs have to overcome more impediments than MNC senior management when co-constructing this new negotiation space. First, unlike corporate management, they are required to demonstrate their legitimacy as international union contacts with their bargaining counterparts as well as their affiliates, national union federations and employees (Helfen and Sydow, 2013). This legitimacy in representing employees is not easily achieved, because of the geographical and cultural distance of GUFs from the grassroots level. Second, GUFs innovate by creating a kind of union network based on numerous links created between union levels and through knowledge exchange and creation. This network coordinates the different levels involved in negotiation and implementation processes (GUFs, national federations, unions at a subnational level). Our aim is to study how GUFs invent working methods, practices and processes so as to negotiate and implement IFAs, outside any international legal framework.
For this purpose, we give a brief overview of previous research examining the role of GUFs in IFA negotiation and implementation and we present the conceptual framework introduced by Amin and Cohendet (2009) to analyse learning processes in organizations, explaining how this framework can be applied to the union actors and to the negotiation and implementation of IFAs. We then present two cases with quite different characteristics: the agreements between the International Union of Food and Allied Workers (IUF) and the Danone Group (10 IFAs concluded from 1988 to 2016), and between UNI Global Union and the France-Télécom-Orange Group (2 IFAs in 2006 and 2014). We explain the research context and our methodology. After this, we present our results on knowledge exchange and creation during negotiation and implementation processes. Finally, we discuss these results to highlight the role of GUFs and international trade union networks (ITUNs) in this innovation process, focusing on the nature of exchanged and created knowledge and the functioning and communication devices of ITUNs.
Negotiating and implementing IFAs: literature and theoretical framework
A new GUF mission in the 2000s
Many research studies have focused on the negotiation, content and, more recently, implementation of IFAs (Niforou, 2012; Papadakis, 2008; Sydow et al., 2014) and on the role of GUFs in these processes. Why do GUFs negotiate IFAs (Fichter et al., 2011)? First, to gain recognition of unionization and bargaining rights (Helfen and Fichter, 2013). Second, to create social norms to improve workers’ employment conditions (Ford and Gillan, 2015) and promote uniform protection for workers within a MNC (Daugareilh, 2006; Helfen and Fichter, 2013). For this, GUFs must be recognized as international union interlocutors of MNCs. But this recognition is not easily earned. The main difficulties they face are selecting the union negotiation team (Alès and Dufresne, 2012; Dehnen and Pries, 2014) and obtaining ‘strong’ IFAs (Sydow et al., 2014), with ‘substantive’ provisions, strict monitoring and compliance procedures (Barreau and Ngaha, 2012). GUFs help to mobilize the union resources needed to lead the negotiation, which are financial and also human (Fichter et al., 2011) and organizational (Dehnen and Pries, 2014).
IFA implementation also reveals the coordination role of GUFs between the different actors (Fichter, 2013). First, they lead the dissemination of the agreement, using various communication channels and networks (Fichter, 2013). Second, once local actors have understood its content (Helfen and Fichter, 2013), they support local negotiations (Niforou, 2012).
The final challenge concerns enforcement. Local unionists play a significant role in measuring effectiveness and identifying partial or deferred compliance. Then GUFs mobilize and coordinate union resources on a global scale (Ford and Gillan, 2015) to denounce non-implementation. They mostly put pressure on top management, provided that they are recognized interlocutors (Barreau and Ngaha, 2013; Hennebert, 2017; Niforou, 2012).
These different stages of negotiation and implementation processes reveal the importance of links between actors, especially between various union levels by creating networks (Barreau and Ngaha, 2013; Helfen and Fichter, 2013). Within these trade union networks, GUFs facilitate the coordination of actors and resources. This coordination represents a real innovation for GUFs, which must create new practices to negotiate and enforce IFAs and to interact with multi-level (global, national, local) union actors. The adaptation capacity of unions has already been revealed by research (Ford and Gillan, 2015). However, knowledge exchange and creation within ITUNs have not yet been analysed through case studies.
‘Architectures of knowledge’ applied to international bargaining innovation
Negotiating and enforcing IFAs represents a genuine innovation for GUFs and associated unions. In doing so, they have to gain new skills and create new knowledge. We consider the cognitive approach developed by organization theoreticians suitable to explain how the complex and entangled links created within ITUNs enable knowledge exchange and creation. Amin and Cohendet stress the centrality of knowledge in generating innovation. For them, and authors following their theoretical approach, ‘knowledge in action’ and ‘communities’ are key determinants in bolstering creativity and innovation in organizations. ‘Knowledge in action’ or ‘knowing’ is viewed as an experience-based process and practice, which is built through ‘everyday interaction in which thinking and acting are combined in an inseparable unity’ (Amin and Cohendet, 2009: 11). ‘Communities’ are groups of people linked by common interests and practices who interact and create knowledge in action or experience-based knowledge. The interactions between heterogeneous communities are the source of knowledge creation and a key factor ‘in the organization and management of corporate innovation and creativity’. These interactions lead to knowledge creation if two conditions are met: repeated interactions and good communication between different communities (Amin and Cohendet, 2009: 112, 122–123). This essential assignment of knowledge creation does not only rely on proximity links but also (and maybe essentially) on remotely created social links or, more broadly, on a ‘relational proximity’ (Amin and Cohendet, 2009: 93).
GUFs and networked unions as communities and creators of knowledge
Hyman (2007) notes that recent literature on organizational learning is relevant to trade union analysis. It seems particularly interesting to apply to GUFs the theory of Amin and Cohendet (2009) outlined above. GUFs are organizations (Cotton and Gumbrell-McCormick, 2012), entities whose members aim at common objectives and, in doing so, implement division and coordination of work. Their members are autonomous national sectoral union federations. When negotiating IFAs and demanding recognition of fundamental labour rights, GUFs and affiliated unions defend employees’ common interests.
The most challenging issue for GUFs is coordination, necessary for the different components of organizations to be kept together and organizational objectives to be achieved (Mintzberg, 1982). Coordination is particularly important for GUFs, which face cultural, organizational and political differences among their affiliates (Cotton and Gumbrell-McCormick, 2012). They operate coordination inside ITUNs, which enable communication and organization of the exchange of information and experience (Platzer and Müller, 2011). However, knowledge exchange and creation play a more important role than information exchange. Indeed, ‘an organization is not a mere information-processing machine, but an entity that creates knowledge through action and interaction’ (Nonaka et al., 2001: 492). The exchange of knowledge and know-how between GUF-affiliated unions and ITUN members produces a learning process within the union network.
GUFs are composed of trade unions which can be considered heterogeneous ‘communities’, holders of knowledge ‘embedded’ in national industrial relation systems. Within GUFs and European Trade Union Federations (ETUFs), encountering knowledge from heterogeneous communities leads to the creation of new knowledge that is essential to negotiating and acting at a transnational level. This process can be applied at all union levels, from local to international. Unions must constantly create socio-cultural, experience-based knowledge as they face new cases of workers’ rights being breached. Union knowledge is partly acquired on the job (Murray et al., 2014; Sawchuk, 2001); union activism implements knowledge which stems from experience (‘invested knowledge’), and more formalized knowledge which is produced, acknowledged and passed on by ad hoc institutions (‘instituted knowledge’); it is created through a collective and relational process of confrontation and mutual enhancement (Di Ruzza, 2009: 30). At local level, unionists know living and working conditions in real terms, thanks to daily contacts and immersion in local work organizations. They know how to negotiate with local management, and their legitimacy to negotiate is mostly based on this knowledge linked to day-to-day and close contacts with employees. At global level, international trade unionists are very far from the grassroots. When, at the beginning of the 21st century, GUFs moved towards a new strategy, IFA negotiation (Croucher and Cotton, 2009: 61), they had to acquire new core competences. Using this theoretical framework, our objective is to analyse what kind of knowledge GUFs create, how and with whom during IFA negotiation and implementation processes.
Research context and methodology
To address the issue of ‘using and creating union knowledge’ during IFA negotiation and implementation phases, we adopted the point of view of ground observers. We rely on two longitudinal case studies (Yin, 2014) to illustrate the interweaving of knowledge and the structuring of a new social dialogue space at international level. Longitudinal studies are relevant because negotiation is a process which must be considered in a context (Strauss, 1978) and can take several years. This process is influenced by the economic and social history of the firms. It is also modelled by union evolution and notably by the relations between different union levels which have been built over time. Our two cases, the IUF and BSN-Danone and UNI and France-Télécom-Orange are emblematic because the two GUFs and French MNCs have differentiated characteristics.
Two GUFs and two MNCs with contrasting characteristics
The IUF was created in 1889 and reconstituted in 1920; it is composed of 430 affiliated unions in 130 countries representing over 10 million workers. It engaged against MNCs such as Nestlé and Coca Cola in the 1970s and 1980s. UNI was created in 2000 from the merger of four international federations and is composed of 900 affiliated unions spanning over 150 countries, representing 20 million employees. Regional organizations of the two GUFs exist on the five continents, pursuing independent activities in close collaboration with the GUFs and their governing bodies. The IUF prefers high-quality IFAs, even if there are fewer in number. Thus, in 2006, it broke off negotiations with ACCOR, which wanted to conclude an IFA on sustainable development but did not agree undertakings on unionization rights stronger than those contained in the 1995 IFA (Barreau and Ngaha, 2012). UNI by contrast wants to conclude as many IFAs as possible, even if the content is limited (Barreau and Ngaha, 2012; Platzer and Müller, 2011: 286). At the end of 2017, the IUF had signed IFAs with only 8 MNCs, and 10 out of 19 with Danone; while UNI had signed more than 50 IFAs covering more than 10 million workers around the world.
BSN was created in 1966 and merged with Gervais-Danone in 1973. The group became Danone in 1994. It is a big private food group which internationalized very early. Several of its leaders embraced social objectives and supported proactive social dialogue. The French public telecommunication operator was a public administration until 1990 and became France-Télécom (FT) in 1991. In 1997, it became a limited liability company, progressively internationalizing. It became Orange in 2012. During the 2000s, top management was only moderately open to social issues and drastically cut down its workforce. A new top management established in 2010 after a period of social crisis has instigated a new social contract with trade unions. These two companies represent two very different social, economic and geographical configurations. Danone employees are present in more than 60 countries (24 percent in Europe, 9 percent in Africa, 23 percent in Latin America and 20 percent in Asia Pacific and the Middle East). FT-Orange operates in 8 European countries and 20 African and Middle Eastern ones which are mainly French-speaking. In all, 61 percent of employees are in France.
The IUF established contact with BSN-Danone top management in the mid-1980s. In 1988, they concluded the first IFA, the BSN-IUF Joint Position on providing quality information, promoting equality of men and women in the workplace, developing skills training and respecting trade union rights. They have negotiated nine subsequent IFAs, including a 1989 Convention on economic and social data in Groupe Danone companies, a 1989 Convention for the promotion of equality of men and women in the workplace, a 1992 Convention on skills training and a 1994 Convention on trade union rights. In addition, they negotiated a Convention in the event of changes in business activities affecting employment or working conditions in 1997, a Convention on the setting up of social indicators at group level in 2005, a Convention on diversity in 2007, a Convention on health, safety, working conditions and stress in 2011 and an agreement on sustainable employment and access to rights in 2016.
Since 2004, transnational negotiation in FT-Orange has been led by the steering committee of the FT-Orange-UNI-Alliance, which encompasses all trade unions present in the company and affiliated to UNI. In 2006, UNI concluded a ‘global agreement on fundamental social rights’. In 2010, UNI and FT-Orange negotiated the creation of a Global Works Council (GWC). Then, in 2014, the parties signed an IFA on health and safety at work.
Data collection and method of analysis
We collected and analysed several types of data. Since 2007, we carried out 27 interviews with international, European, national and local unionists. Our sample is mainly composed of national representatives engaged in transnational negotiation with Danone and FT-Orange. Trade unionists in charge of production sites act at the local or subnational level but sometimes act at several levels. To obtain contextual elements, we interviewed four top executives in human resources (two from Danone and two from FT-Orange) and two consultants who have supported Danone unions.
The face-to-face or phone interviews lasted from 45 to 150 minutes. Interviews focused on how relationships between executives of the MNC and the GUF (and other unions) have been built; identification of the social actor who suggested the opening of a negotiation process; negotiation issues for the two protagonists, negotiation context, and negotiation, dissemination and implementation processes. This interview template enables us to identify new knowledge, so that GUFs could carry out each process, and identify knowledge brought by different union levels.
We attended, as observers, the Danone CIC (European Works Council) preparatory meeting (in 2009). We also worked with documentary sources (academic and expert studies), reference reports, corporate social responsibility reports, press releases and documents provided by our interviewees.
All interviews were recorded, transcribed and analysed using content analysis techniques. Finally, we proceeded to a triangulation of documentary information and interview contents to strengthen the internal validity of the research (Eisenhardt, 1989). The external validity of this study can be appreciated in the light of other research studies (Miles and Huberman, 1994); however, the results produced for the two case studies cannot be generalized. We present these results in the next section.
GUFs as key actors in knowledge exchange and creation when negotiating and implementing IFAs
GUFs create different types of new knowledge through exchanges within multi-level union networks. Both the IUF and UNI have learnt to become interlocutors of MNC headquarters, to introduce topics and establish an international negotiation mandate, to form a negotiating team, and to implement IFAs. Implementation involves four stages: dissemination, establishing ‘local ownership’ of the IFA (Helfen and Fichter, 2013), monitoring and enforcement. We focus on the enforcement stage, as dissemination is not a new GUF mission, ‘local ownership’ is the task of national federations and monitoring is the duty of employee representatives. Our narratives of the IUF-Danone case (especially the 2011 IFA) and the 2006 and 2014 UNI-FT-Orange negotiation and implementation processes highlight communities in action and the knowledge they bring to GUFs.
Knowing how to create international contacts with MNC headquarters: the role of home-country unions
The first knowledge that a GUF or a union alliance needs to create before initiating and completing a negotiation process is how to become international contact of MNC headquarters. In both cases, the credit for convincing senior management and organizing preparatory meetings belongs to the home-country unions. They also provide behavioural knowledge to foreign unions and the GUF so as to build long-term relationships with headquarters. ‘It is difficult to establish an international social dialogue without any home bridge. Headquarters unionists’ support and contribution are absolutely essential. They have ‘privileged’ contacts with the management’ (IUF official, September 2009).
The IUF was recognized very early as the international union contact of BSN top management. In the mid-1980s, the food workers’ federation of Force ouvrière (FO) organized a meeting between the CEO of BSN and the IUF general secretary. They decided to set up regular meetings between senior management and representatives of IUF affiliates present in the company. In 1988, the first IFA was concluded. In 1996, an information and consultation committee, Danone-CIC, was created, becoming a GWC in 2009.
By contrast, the FT-Orange-UNI-Alliance had to struggle for a year to become recognized as the international union contact of senior management. The idea of integrating representatives of all UNI affiliates in the FT-Orange group in an ‘Alliance’ came from African unionists needing to get their views across to top management. Until then, their claims were relayed by the French unions. The Alliance was created during the 2003 UNI congress and recognized by top management in 2004. This is a tribute to French trade unionists who ‘lobbied’ top management and approached them about an IFA. ‘Other UNI affiliates rely a lot on French ones. We represent three-quarters of FT employees and are close to the senior management. This enhances our legitimacy’ (CFDT official, July 2008). In 2010, significant managerial changes, the retirement of several union negotiators and the creation of the GWC brought into question the status of the Alliance as the international union contact of top management. In March 2014, the ‘Abidjan FT-Orange group and FT-Orange-UNI-Alliance joint declaration’ restored this status, after some conflicts arose inside African FT-Orange subsidiaries. ‘We reminded HR managers of Alliance prerogatives and its international union role’ (CFDT official, March 2017).
Knowing how to introduce IFA topics and establish an international negotiation mandate: the role of affiliates, the EWC/GWC and GUF expertise
To negotiate legitimately on behalf of employees and propose suitable objectives for an IFA, GUFs need to know national situations and challenges in detail, and must then rely on the knowledge of national unions and worker representatives. In the Danone case, there is permanent consultation with affiliates, which enables the IUF to identify their key demands. ‘Sometimes, in a workshop, our affiliated union says: “we have already negotiated about these issues that have been dealt with in the IFA” or “these issues are not at all the ones that we have to deal with just now”’ (IUF official, September 2009). Conversely, in FT-Orange, feedback on African members’ demands is much more difficult because the unions are new and social dialogue is little developed, despite an attempt to create an African inter-union organization.
In the both cases, topics submitted to senior management were drawn up through union networks, the Danone-CIC, FT-Orange-GWC and FT-Orange-UNI-Alliance. Indeed, strong links exist between members of these bodies, who sometimes hold several union functions and representative mandates concurrently. They discuss problems they encounter in MNC subsidiaries, especially during preparatory union meetings before formal EWC/ GWC meetings. The Danone-CIC steering committee is now the IFA negotiator, while all members of the FT-Orange negotiation team are GWC members. These employee representatives highlight specific problems during preparatory, plenary and follow-up meetings and through informal relations. For example, the 2011 Danone IFA on working conditions followed a request formulated by CIC members and IUF affiliates. ‘During the 2009 CIC meeting, lots of critical comments were expressed concerning stress at work, working conditions (as usual in Mexico and Indonesia). An international claim resulted’ (Belgian negotiator, June 2012).
Once the IUF has collected demands from affiliates and workers’ representatives, it must perform essential tasks based on its expertise before proposing a draft IFA to top management. It must align the claims to any previous agreements signed with French unions, and with European directives. Then, with the help of home-country unions, it translates them into demands acceptable to top management and defines a clear mandate for the negotiation team. Thus, the 1997 IFA is based on previous restructuring experiences (creation of redeployment units for example) and on ‘basic requirements of unions which always defend the opportunity to express and propose alternatives to managerial restructuring projects’ (IUF official, December 2009). The negotiator of the 2011 IFA on health, safety, working conditions and stress explained that the union proposal factors in the European directive on stress and the 2010 Danone agreement negotiated with French unions.
In UNI-FT-Orange, the process is the same except that the GUF role is played by Alliance, which highlights the topics. ‘The 2006 IFA draft came from UNI-FT-Orange-Alliance…. In 2014, we were concerned about HIV/AIDS, Ebola…. As to Orange top managers, they needed to harmonize corporate health/security provisions’ (CFDT official, March 2017). The 2014 IFA topics were selected during a meeting in Abidjan with the new Orange top management, organized at the initiative of the Alliance. This health and safety topic seemed to fit the African affiliates’ situation and the managerial need to harmonize health and safety mechanisms within the group. This meeting also helped to recognize the role of the Alliance as an international union contact of top management and to strengthen international social dialogue.
At this stage of the process, the GUFs also rely on experience acquired in several transnational negotiations. They give examples of wording and content, they warn against pitfalls and suggest a general negotiation framework as defined by their mandate. In the course of negotiation, the mandate could be revised. The IUF may find it necessary to come back to affiliates to obtain their support or their consent to break off the negotiation process, leading to the reconsideration of union demands. ‘At the end of the negotiation process, we added requirements we did not have at its beginning. For in the meantime, we had a CIC meeting and consulted our affiliates on this occasion’ (IUF official, January 2009). However, in the FT-Orange case, affiliates were not consulted during the 2006 and 2014 negotiation processes. Union negotiators think that the mandate granted to them, and the presence of a UNI official and an expert on the communication sector, exempt them from providing feedback to their constituents, except in case of ‘sticking points during the negotiation process’ (Alliance ex-president, February 2017).
Knowing how to form an international negotiation team: choosing experienced officials
To form an international negotiating team, the two GUFs have used different solutions, though both have relied on the bargaining experience of national officials. In addition, the French trade unionists brought knowledge about top management: how to contact them, identify decisive factors, conduct debates, use correct wording, avoid pitfalls and so on. Their role appears to be crucial. The French unionists serve on several bodies (works councils, Group Committees …) and build relationships with top management in the context of French negotiations.
In Danone, members benefit from experience gained in four different national (French, Belgian, Spanish and Dutch) negotiations. Their long tenure in the international negotiation team is a valuable asset. Thus, in 2011, the main union negotiator had eight years’ membership of the international bargaining team and many more in national (Belgian) negotiations: IUF officials told me: ‘you’re the one who will coordinate and be a bridge between IUF and Danone top-management’. I have spent eight years in the Danone EWC steering committee and I have gained contact and friendship-based experience. I possess twelve-year experience of local and national bargaining. Without experience, agreements stay theoretical and their enforcement is impossible. (Belgian negotiator, June 2012)
This knowledge of international and national collective bargaining facilitates negotiation of ‘substantive’ IFAs. It is cross-referenced to the knowing of other national federation representatives. The latter express their views on content of the IFA, on arguments and ways to obtain a satisfactory outcome. Interpersonal networking links created over time make informal inter-unionist exchanges easier.
The French trade unionists’ role is manifest in the 2006 FT-Orange IFA. Negotiations were led by the Alliance steering committee, composed of the Senegalese chair and representatives of the three French unions. The latter brought their knowledge of FT-Orange industrial relations and the company’s economic situation. They also brought experience from IFA negotiations carried out in other French public firms, such as EDF. They were, however, confronted with a new negotiation experiment that they hoped to use in future transnational bargaining. In the 2014 IFA, French and African unionists were new members of the international negotiation team, except for the CFDT representative who participated actively in creating the Alliance and retained experience from the 2006 IFA. ‘I was aware of the whole Alliance story. Yet other negotiators were new: new union teams and HRM’ (CFDT official, March 2017). So knowledge was lost and new forms of learning were under way.
Knowing how to enforce IFAs: the crucial role of local unions
Monitoring by EWC/GWC members is necessary but not sufficient. GUFs must also intervene in cases of non-compliance with IFA provisions. This requires that local unions inform them about these cases. When a non-compliance case occurs, knowledge is mobilized, exchanged and created. The GUF and other union levels have to learn how to get feedback and force top managers to intervene with local managers so that they comply with IFA provisions. To flag up non-compliance, local union knowledge is essential. Based on their knowledge of local situations and the interpretation that they give to the provisions of the agreement, they can draw up demands to remedy the discrepancies. For this to happen, the GUF (or the FT-Orange-UNI-Alliance) and the national federations should identify ‘interested’ local unions in the agreement and ‘translate’ some provisions the content of which may not be well suited to a local situation. To achieve this, international trade unionists can refer to implementation examples of similar provisions contained in previous IFAs. The reporting of non-compliance is handed over to the GUF, which must relay this to top management, possibly creating a favourable balance of power by threatening an industrial dispute.
This process corresponds to implementation in Danone. Thanks to the trusting relationship between senior management and the IUF, the company accepted the IUF request for recourse to an expert chosen by the union (as provided in the 1997 IFA), with the mission to examine alternative solutions to the closure of a Hungarian biscuit plant in 2001. Management eventually decided to implement the expert’s solution. Similarly, top management accepted a union request and asked North American managers to accept employee unionization in 2009. ‘We encountered a concern about a new IUF-affiliated union recognition in a US plant. This union contacted IUF because local managers did not want the union to get in the plant’ (Belgian negotiator, June 2012). At the request of IUF, top management has interceded to convince a local management team to implement IFA measures. Since 2009, ‘all North-American Danone plants have been unionized’ (IUF official, September 2009). This ‘harmony’ in relationships owes much to senior management who wish to develop social dialogue and avoid social conflict. The relationship between IUF and other MNCs could be different: it might organize employee mobilization if IFAs are not enforced and adapt ‘traditional’ know-how to contemporary situations thanks to the knowledge of the national and local unions involved.
The FT-Orange Alliance has often engaged senior management over non-compliance with the 2006 IFA, after a dispute broke out in an African subsidiary. ‘Local unions warn Alliance of unconventional practices. Then we alert Orange top management. This is usual and I must say that top management is mindful of these issues’ (CFDT official, July 2008). Thus, Alliance intercession occurred in crisis situations up to 2014. For instance, in 2008, in the Cameroonian subsidiary, the UNI affiliate used the IFA to obtain a revision of the paid leave policy. It did not succeed in finding a compromise with local management and approached the Alliance steering committee, which then informed top management. So the top HR manager was appointed to intercede locally, and every employee obtained a five weeks’ paid leave. ‘We won about paid leave’ (Cameroonian unionist, May 2008). So both GUFs brought employee claims to the attention of senior management. However, until 2014, FT-Orange top management has reacted only after a dispute has erupted and a balance of power at a plant or country level has been created.
The innovation role of the GUFs is demonstrated because the contents of IFAs are improving as well as their effectiveness in regard to their main objective. In effect, unionization grows in the two MNCs (Danone plants in North-America and FT-Orange subsidiaries in Africa). Hennebert (2017) obtains the same finding when studying two other cases.
Discussion: GUFs at the heart of an innovative and social process
Our empirical results highlight the role of GUFs and ITUNs in this innovative process of negotiating and enforcing IFAs. We now discuss the nature of the knowledge created and exchanged between heterogeneous communities and the functioning and communication devices of GUF-led ITUNs.
The nature of knowledge created and exchanged between heterogeneous communities
GUFs have learned to negotiate IFAs, of which the contents are increasingly binding and the procedures increasingly detailed. Knowledge exchanges generate a virtuous learning circle: successful learning, improving processes, imitation and cross-fertilization, knowledge-sharing effects. This learning process can be identified thanks to the repeated experience of IFA negotiations (2 for FT-Orange and 10 for Danone) and reveals a genuine capacity to learn (Lévesque and Murray, 2013). It is possible if some stability of negotiators exists. In our two case studies, GUFs have learnt to put pressure on top management to enforce IFAs locally. This learning is the product of knowledge exchanged among local, international and national unions. As underlined by Amin and Cohendet (2009: 122), this new knowledge emerges from exchanges between heterogeneous communities.
The GUF-led ITUN is formed with different union communities. In both case studies, GUFs are leaders in negotiation and implementation processes, as shown in previous research (Croucher and Cotton, 2009; Dehnen and Pries, 2014; Fichter et al., 2011). But GUFs could not manage these processes without the help of other international, European, national and local unionists who represent ‘heterogeneous communities’. This heterogeneity shapes the specific knowledge held by each of these communities. It is ‘embedded’ in national systems of industrial relations. Heterogeneity is also associated with specific sectoral characteristics. For instance, the IUF is a community that defends food workers’ interests; but its affiliates are very heterogeneous, some representing tobacco workers, others hotel and restaurant workers and so on. Also, IUF officials are usually specialized by sector and thus develop specific knowledge. Each national federation is also heterogeneous because its members are local unions which represent interests of employees working in competing companies.
National unions have to deal with contradictory demands (Hyman, 1997), whereas GUFs pursue objectives that aim to improve all employees’ working conditions and rights. Although there might be some conflicts of interest within GUFs (Cotton and Gumbrell-McCormick, 2012), in our study, we did not observe divergences with the objectives of affiliates during the international bargaining process: they defended the common interests of all employees, aiming to obtain the recognition of unionization and collective bargaining rights and to improve the living conditions of employees. Once these essential claims have been achieved, union negotiators may differ in their subsequent priorities. For example, in the 2006 FT-Orange negotiations, French unionists would have liked to mention public service obligations (FO unionist, Alliance steering committee member, December 2008), while African unionists wished to introduce the fight against pandemics and corruption. The first claim was not acceptable to top management, but the second was accepted. Senior management power prevails in IFA negotiation.
The IFA negotiation and implementation processes require knowledge creation on the basis of knowledge provided by the heterogeneous communities belonging to the GUF-led ITUN. Our study specifies the kind of knowledge exchanged and created at different union levels. Each level brings specific knowledge. Home-country federations provide necessary knowledge to establish international social dialogue within MNCs. Knowledge provided by affiliates and EWC/GWC bodies enables a union negotiating mandate to be established and guides the negotiators throughout the whole process. Preliminary bottom-up communication is necessary to transmit local demands to the negotiators and produce realistic IFAs. During IFA negotiation meetings, knowledge acquired by experience of international and/or national bargaining is used by members of the negotiating team. Exchanges between union negotiators outside the bargaining process are also useful in learning how to obtain strong commitments about unionization rights, employee protection in case of restructuring and how to word these managerial commitments. Interactions between ITUN members enable learning of new methods of international bargaining to be used in subsequent IFA negotiations. The implementation phase depends on local unions’ knowledge and its transmission to the GUFs.
Union knowledge is mostly ‘invested knowledge’ (Di Ruzza, 2009) or experience-based knowledge (Murray et al., 2014; Sawchuk, 2001). This knowledge ‘can be transferred without codification (for instance, by recreating the learning context, as in the case of driving lessons)’ (Amin and Cohendet, 2009: 21). Therefore, in the bargaining team composed of unionists of diverse nationalities, each brings his or her own nationally acquired experience, in this way contributing to the creation of new knowledge suited to the context: the positions, methods, wording of top management representatives with whom they negotiate, the international situation of the company, the contents of IFAs already concluded. Of course, unionists have to master ‘scientific knowledge’ or ‘instituted knowledge’ (Di Ruzza, 2009) as well.
During negotiation and implementation processes, GUFs and their ITUNs exchange information but also create knowledge. Knowledge is a very complex, loose and flexible structure. A piece of information received by an individual or a group is sometimes immediately rejected. Sometimes, it is processed and added to the existing stock of knowledge, leaving the structure unchanged or contributing to its complete reorganization (Amin and Cohendet, 2009:18–19). As shown in our two cases, a GUF must collect information and knowledge from its affiliates before suggesting a draft IFA to senior management. Thus, in 2009 several South American Danone CIC/EWC members reported evidence of bad working and health and safety conditions. This evidence reinforced the knowledge of IUF officials about Southern employees’ working conditions. In order to propose a draft IFA on that issue to top management, the IUF needed to collect knowledge held by other union levels (national and sub-national): for example, bargaining conditions and contents of national working condition agreements and Danone policies on health and safety at work. The information and knowledge collected contributed to broadening IUF knowledge about how to propose this IFA and improve its content compared to previous IFAs.
GUF-led ITUNs and their communication devices
Many authors point out that GUF-led ITUNs are critical in developing international bargaining. We examine the main features of ITUNs and whether the two conditions required for knowledge creation are met. Our results show that global union networks can be formal and informal and are characterized by their multiplicity and flexibility; this complements the research of Helfen and Fichter (2013).
GUFs are at the centre of formal networks which are composed of their affiliates, national federations and their regional organizations. Alliances are another form of network. The UNI-FT-Orange-Alliance is a formal trade union network gathering UNI affiliates present in the FT-Orange group under the responsibility of an elected chair. In both the IUF and UNI, affiliates are national sectoral federations which have been accepted as members. However, in some African host countries of FT-Orange, unions are not organized in federations, and UNI accepts corporate unions as full members. EWC/GWC networks generally have their own chair, but the IUF has required that the Danone-CIC/EWC should be managed by itself and national federation officials. Then, the union bargaining team is the interface between the GUF-led ITUN and the EWC/GWC network. This union negotiation team is exclusively composed of European trade unionists, although only 24 percent of Danone employees are located in Europe. By contrast, since its creation in 2003, each successive president of the UNI-FT-Orange-Alliance has been an African, who is a member of the international negotiation team and can present African unions’ claims during IFA negotiation. Each global union network is a part of the GUF-affiliated union network as it comprises only the affiliates present in one MNC. The headquarters unionists play a great role. The scope of the union network evolves with corporate boundaries.
Beside formal relations linking network members, some informal relations build up. Affinities or a common language can bring together internal or external members of these networks (e.g. subcontractor or supplier unions). Hence, these subgroups contribute to exchanging and creating the knowledge required in IFA negotiation. In these networks, local unions play a particular role in the IFA implementation process; they alert the GUF to cases of non-implementation.
Our study shows that IFA negotiation and implementation are time-consuming processes that involve ongoing relations between actors. Interactions between heterogeneous communities are beneficial if two conditions are met: ‘a high level of repetition of interactions’ and a good quality of communication between different communities (Amin and Cohendet, 2009: 122–123). If the first condition is usually fulfilled by international unionism, the second is more demanding. The GUFs keep up regular relationships with their regional organizations (such as UNI Africa), their national affiliates, alliances and EWC/GWC bodies, during regular meetings (executive committees, congresses, EWC/GWC and steering committee meetings) and also seminars, expert groups, conferences, training courses …. For some of these meetings, GUFs use facilities available in Brussels (the International Trade Union House) or Geneva (international union and ILO headquarters). These cities are nodal points for the international union movement (Hyman, 2005).
Moreover, links are created through virtual communications: union press, electronic messages and mail exchanges (Pulignano, 2009). Thus, knowledge exchanges occur ‘in the achievements both of face-to-face contact and physical proximity and of distantiated and relational proximity facilitated by various space-spanning devices’ (Amin and Cohendet, 2009: 10). However, union actors in negotiation and implementation processes encounter language and cultural difficulties (Cotton and Gumbrell-McCormick, 2012; Fichter et al., 2011). For FT-Orange, communication between the different union levels is facilitated thanks to a common language (the group is mainly extended in French-speaking Africa). But more than 20 languages are spoken inside the Danone group. However, the communication channels seem to work well, so that the IUF gets rapid feedback on non-compliance cases from affiliates, CIC-EWC members or directly. In FT-Orange, union communication networks are now being implemented. Many reasons could explain this delayed implementation: late corporate internationalization, recent relations between the Alliance and senior management and sometimes difficult communications with African unions because of technical issues and union activist overload.
Conclusion
Our objective was to explore the innovation role of GUFs: how learn to negotiate and implement IFAs through knowledge exchanges between multiple union levels. Our two longitudinal case studies highlight knowledge exchange and creation processes inside ITUNs, and the nature of the knowledge exchanged and created. They demonstrate that two conditions are necessary for GUFs to play an innovation role and satisfy core union issues, at least partially. They must coordinate exchanges between heterogeneous communities, holders and creators of specific union knowledge and also facilitate the dissemination of knowledge by maintaining regularity and quality of exchange between ITUN members. These top-down and bottom-up interactions create a virtuous circle. In the two contrasting cases, the dynamic of knowledge exchange and creation enhances GUF legitimacy towards top management and enables working conditions to be improved on a global level. But these GUF-led ITUNs defend only a minority of workers in the world.
Our article, framed by organizational theory, adds to existing literature in industrial relations. Analysing two cases, we provide further insights into union organizational learning as defined by Hyman (2007). It also complements research by Helfen and Fichter (2013) on exchanges within transnational union networks. Our research has limitations, given the uniqueness of each case and the ongoing process of IFA negotiation and implementation. However, it allows us to conduct in-depth analysis of union knowledge specificities and creation processes in different contexts. IFA implementation is particularly difficult to study because of the global scope of agreements. If the international union movement continues to focus on concluding IFAs, it will remain important to analyse further the GUF innovation role in knowledge exchange and creation, on the basis of a similar methodology. New longitudinal case studies on French MNCs which have concluded several IFAs could broaden the basis for comparison of learning processes inside GUF-led ITUNs.
Footnotes
Acknowledgements
We thank the editor and reviewers for the quality of the comments we received. We also thank John Hughes who corrected our English.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Part of this research was conducted within the programme ‘Le potential régulatoire de la RSE’ (2004–2011), financed by the Agence Nationale de la Recherche (ANR) under the scientific responsibility of Julienne Brabet.
