Résumés
Abstract
Could “centralized” bodies of collective representation constitute a power resource for employees in contemporary fragmented work organizations? Referring to the particular features of the institutional framework in France, this article links the theory of power resources to the concept of institutional toying to show that the power resources of managers allow them to make use of institutional shortcomings that limit the scope of these bodies as power resources for employees. The article draws on the REPONSE 2017 (DARES) survey, which sought to quantify “fragmented” organizations and their collective representation bodies. Six case studies of fragmented companies are also used to illustrate the diversity of institutional toying strategies employed by managers.
Summary
The traditional representation of a company has been undermined by processes of productive disintegration (outsourcing and geographical dispersion) and economic and financial concentration (subsidiaries). This movement has been accompanied by new collective rights in the form of bodies at the “centralized” level, which are intended to take into account the entire workforce. Referring to the particular features of the institutional framework in France, this article links the theory of power resources to the concept of institutional toying to show that the power resources of managers enable them to make use of institutional shortcomings that limit the scope of these bodies as power resources for employees. It draws on the REPONSE 2017 (DARES) survey to quantify “fragmented” companies and the collective representation bodies that operate within them. Six case studies of fragmented companies are used to illustrate the diversity of institutional toying strategies employed by managers in setting up centralized bodies. We begin by highlighting the prevalence of three forms of fragmentation (outsourcing, group, multi-site) and the limited role played within them by centralized bodies. The institutional limits of centralized bodies are then analyzed based on the power resources available to management and the diversity of strategies implemented to configure these bodies. Next, the fragmentation of companies is analyzed as an instrument for weakening employees’ power resources, as it reduces their interest and ability to participate in social dialogue, and divides and creates competition within a fragmented workforce. Finally, associational resources to overcome the divisions created by fragmentation and institutional resources to rebalance the ability to set the framework for employee consultation are discussed as solutions to restrict management leeway.
Key words:
- Outsourcing,
- subcontracting,
- subsidiaries,
- collective representation,
- centralized bodies,
- power resources,
- management strategy
Corps de l’article
1. Introduction
The traditional concept of a company is based on a dual unit: the unit of location, with all employees working in a same workplace, and the management unit, with a single employer holding legal and economic power. This dual unit makes it possible to build a framework for collective representation in which clearly identified players are grouped together in bodies that provide them with both the information and the maneuvering room to reach compromises. This was the prevailing concept of a company when employee representative bodies (instances représentatives du personnel, IRP) were set up in France, from works councils (comités d’entreprise, CE) in 1946 to the extension of their privileges under the Auroux Laws of 1982.[1] Since the 1990s, in keeping with the decentralization of collective bargaining, there has been a succession of reforms (Giraudet, 2018), the most recent of which concerned the merger of IRPs within establishments into the Social and Economic Committee (Conseil social et économique, CSE) in 2017.
While the decentralization of collective bargaining at the company level has been widely promoted in France and internationally (Visser, 2016), the traditional conception of a company has been largely undermined by a dual process of productive and organizational disintegration on the one hand, and economic and financial concentration on the other (Marchington et al., 2005; Weil, 2014; Perraudin et al., 2014; Appelbaum, 2017). What, then, is the company level when the work collective and centres of power are “fragmented” and embedded in multiple networks of dependency? A number of studies have shown the deterioration in working and employment conditions for employees who are “far removed” from the centres of power, particularly in the companies dependent on order givers. According to Bernahrdt et al. (2016), the development of outsourcing in the United States is at the root of an increase in wage inequalities, as well as pressure on wages and working conditions that particularly affects the least qualified and lowest paid.
For France, Perraudin et al. (2014) show that order takers concentrate execution of the work and that their median wages by qualification are considerably lower than among order givers. Overexposure of subcontracted employees to occupational hazards also raises the question of order givers outsourcing their risks and the more difficult and dangerous work (Thébaut-Mony, 2008; Perraudin and Thèvenot, 2022). This fragmentation of work is therefore at the root of an unprecedented deterioration in working conditions, which cannot be addressed by the traditional concept of the company as a relationship between employer and employees (Weil, 2014). According to Doellgast, Bidwell and Colvin (2021), the growth of employer power and the decline of institutional protections since the 2000s have reached such a scale that they call into question analyses of the employment relationship based on a clear delimitation of the actors involved (unions, employers, states), and call for the process of fragmentation of work and productive processes to be embedded in renewed theories of industrial relations.
Because the relationship of subordination goes beyond the legal boundaries of the company, when it is “fragmented” by the presence of order givers or financial groups, there is a need to rethink the institutional framework of collective employee representation and the associated power resources, for both employers and employees. In France, the local level of representation (the establishment or the company) can be articulated at the inter-establishment level within the framework of the central works council (comité central d’entreprise, CCE), or at the inter-company level with the group works council (comité de groupe, CG) or the economic and social unit (unité économique et sociale, UES). This institutional design thus formally creates a framework for worker representation that could correspond to the organizational structure of the “fragmented firm.” But is it a resource that enables workers to make their voices heard?
This article links the theory of power resources (Refslund and Arnholtz, 2022) to the concept of institutional toying (Benassi and Kornekalis, 2021) to show that employers use institutional shortcomings to limit the scope of power that these institutions could represent for employees. Indeed, through their strategy of fragmentation, managers gain structural, institutional and associational power (in the terminology of Wright, 2000), endowing them with a scope of action that can affect the very configuration of employees’ power resources. In other words, fragmentation gives managers the ability to (re)configure collective representation to their own advantage. This is a case of “institutional toying” where, through fragmentation, managers gain institutional power resources in such a way that keeps workers in a weakened position. The formation of centralized representative bodies, involving different components of the “fragmented” company, bears similarities to network bargaining in that they are intended to unite a fissured workforce in the face of corporate fragmentation strategies (Anner et al., 2021). Our study of collective representation bodies in France, rather than of the negotiation content,[2] leads us to conclude that associational resources can be shaped by employer strategies. The ability of employers to shape these centralized bodies as well as their interactions with local representative bodies becomes an institutional resource for them, limiting the associational resources available for employees.
We have identified three forms of fragmentation, each of which is a source of dependency between companies or establishments. “Productive” fragmentation refers to the processes of vertical disintegration and outsourcing of ancillary or core activities, which lead order givers to organize and control the activity and workforce of order takers, who find themselves in a de facto situation of controlled autonomy (Appay, 2005). “Capital-based” or “financial” fragmentation, which is the subject of a wealth of literature, is partly linked to the financial restructurings carried out by organizations since the 1990s (Beaujolin-Bellet and Schmidt, 2012). The result is a loss of autonomy for subsidiaries in terms of strategic direction, and a restrictive budgetary framework for decision making, particularly with respect to employment. Finally, “geographical” or “site” fragmentation refers here to the presence of several “establishments” belonging to a same company, with partial “autonomy” in managing employment and working conditions, but dependent on decisions made at head office. Empirically, we rely on the REPONSE 2017 (DARES) survey to quantify fragmented companies in France, as well as the type of IRP deployed within them. We also have six case studies of fragmented companies to illustrate the diversity of employers’ institutional toying strategies in setting up centralized bodies.
After first highlighting the prevalence of the three forms of fragmentation studied, and the limited role played by centralized bodies, we reveal the institutional limits of centralized bodies, through an analysis of the latitude available to management. We then show that these difficulties are consubstantially linked to fragmentation, through the processes of division it feeds. We conclude with proposals likely to help restore associational resources for employees in the fragmented company and restrict management’s room for manoeuvre.
2. Quantitative weakness of centralized bodies in a production landscape dominated by fragmented companies
Establishments are largely affected by various forms of fragmentation (2.1). While French legislation provides for centralized representative bodies covering fragmented companies (2.2), many establishments do not have any (2.3).
2.1 Dominant forms of fragmentation
The REPONSE (Relations professionnelles et négociations d’entreprise, professional relationships and business negotiations) survey, carried out every six years by the Directorate for the Coordination of Research, Studies and Statistics (Direction d’animation de la recherche et des études statistiques, DARES) of the French Ministry of Labour, provides information on labour relations in establishments in France with at least 11 employees, in the commercial and not for profit sectors, excluding agriculture.[3] It enables us to identify which establishments are part of a fragmented company (through subsidiaries, outsourcing, or membership in a multi-site company), as well as the IRPs present at the local and centralized levels. In establishments with at least 11 employees and belonging to companies with 50 or more employees, and for which, on the date of the survey, employers were required to organize elections for the establishment of a works council (CE),[4] the survey results show that fragmentation has increased since the 2005 survey, even becoming the norm (see Table 1). In 2017, 88% of the establishments studied were involved in an outsourcing chain as an order giver and/or order taker.[5] More than four out of five establishments belonged to a multi-site company, and almost two-thirds were subsidiaries of a group.
Beyond the widespread trend towards fragmentation, it is the combination of forms that stands out. More than half of all establishments are involved in outsourcing, belong to a multi-site company and are part of a corporate group, and only 2.5% of establishments are not affected by any of these three forms of fragmentation.
Table 1
Forms of fragmentation and their combination
* The question on the use of outsourcing was modified in 2011 to include all forms of outsourcing, which may explain the lower proportion of establishments concerned by outsourcing and the three forms of fragmentation in 2005.
Scope: establishments with 11 or more employees, belonging to a company with 50 or more employees, in the commercial and not for profit sectors, excluding agriculture.
Faced with the various forms of fragmentation, French legislation has, since instituting the works councils, encouraged the emergence of new bodies for social dialogue that cover not only an employer’s direct employees, but also the “social reality” where employer power is exercised (Giraudet, 2018) by promoting interaction between the local level, where the concrete conditions of employee life are lived, and the central level, where economic decisions are made.
2.2 Three types of centralized bodies in the fragmented company
Created in 1946, the central works council (CCE) was designed to bring together several levels of representation in companies with complex structures, i.e. local (embodied by the works council) and the central level (embodied by the central works council). Twenty-five years later, in 1970, with the jurisprudential creation of economic and social units (UES), the possibility of collective representation bringing together employees from different employers appeared. Then in 1982, the Auroux laws gave the works council its current role,[6] stating that a group works council (CG) must be set up within corporate groups formed by a dominant company and its controlled companies, thus recognizing that employer power can be exercised over the employer itself through its relations of financial dependence. The 2017 ordinances reforming the French Labour Code, which merged the various IRPs existing within establishments into a Social and Economic Committee (CSE), did not alter the various possibilities offered for setting up collective representation within the complex structures of fragmented companies: establishment CSEs and central company CSEs can be set up, UES works councils become UES CSEs, and group works councils (CGs) remain unaffected.[7] They have, however, opened up the field of negotiation to the actual structuring of the body, and in particular to the number of distinct establishments.[8]
Thus, there are three types of centralized bodies aiming to adapt social dialogue to changes in companies’ scope. The group works councils are a response to the growing number of subsidiaries, while the central committees are designed to adapt to the needs of multi-site structures. Finally, if there is no centralized body specifically designed to supervise outsourced work, the economic and social units (UES) can play this role. By definition, economic and social units are made up of several legally distinct companies with close ties. The economic unit refers to the complementarity or similarity of their activities, as well as the concentration of management powers in one of the companies. The social unit refers to the existence of a homogeneous working community Accordingly, a UES can be formed in the context of outsourcing relationships, but it can also be formed, for example, between certain subsidiaries of a group. In this way, it transcends all forms of fragmentation, as long as the close relationships are recognized by court decision or collective agreement.
2.3 Little centralized presence
The REPONSE survey enables us to assess the extent to which employee representative bodies have been set up, at both the local and central levels. In 2017, before the labour code reform, three-quarters of the establishments studied had a works council at the local level, and the number was even higher (nearly 80%) when the establishment was part of a multi-site company or a corporate group: in such cases, local bodies are very common (Table 2).
Table 2
Local and centralized bodies in fragmented companies (% of establishments)
Scope: establishments with 11 or more employees, belonging to a company with at least 50 employees, in the commercial and not for profit sectors, excluding agriculture.
In contrast, although French legislation requires a central works council to be set up in multi-site companies, only 19.8% of the establishments concerned have one. Similarly, only 24.5% of establishments belonging to a group subsidiary have a group works council (CG), and only 19.1% of establishments are covered by an economic and social unit (UES).[9] Even though the forms of fragmentation that make centralized bodies necessary are extremely widespread, these findings indicate that only a small proportion of establishments are covered by such a body, even where this is provided for by law.
3. Centralized bodies—a manifestation of employers’ power resources?
According to Refslund and Arnholtz’s (2022, p. 1962) re-reading of Korpi’s work, employee representative bodies can be viewed as a tool for expressing the state of power relations through the mediation of a series of “routinized operations,” making it possible to avoid deploying other power resources, while still keeping them available. These authors call for a dynamic approach to understanding the evolution of these institutions and the balance of power between categories of actors. However, while centralized bodies were created to provide employees with a means of counter-power, the provisions defining their set-up and structure can transform them into a power resource for employers, in a way that is particularly visible in the context of the fragmented company.
We demonstrate here that the legal framework for centralized collective representation makes it possible to turn these bodies into institutional resources for employer power in fragmented companies. In fact, this framework allows management to refuse or avoid these bodies (3.1), or to prevent them from being set up at the local level (3.2). This gives management considerable leeway in their configuration (3.3). We have drawn on our fieldwork (see inset) to highlight and analyze the diversity of possibilities for “institutional toying.”
3.1 Refusing or avoiding the creation of a central body
As we saw earlier (Table 2), fewer than one in four establishments in fragmented companies in France has a centralized body corresponding to its form of fragmentation. Yet the law stipulates that the employer has both the responsibility and the obligation to organize elections to set up such a body. However, as there is no penalty for failing to meet this obligation, the effectiveness of this legal rule depends on the vigilance of employee representatives.
Three of our cases have no centralized bodies. The LABO case illustrates the power of company management to refuse their creation. Two requests by employee representatives for recognition of an economic and social unit (UES) were rejected by management. While LABO management’s refusal required the support of a law firm, the absence of a group works council or UES at FAÇONNAGE and AERO is more a case of avoidance. The interviews revealed that the possibility of creating such bodies had been discussed at management level. However, the human resources managers or site managers are often not the ones who make the decisions. This is demonstrated by the LABO employee representative’s comments: “We also talk to managers who are not the decision-makers, so they just do what the group tells them to do.”[10] He adds that the decision is “at some point always validated by someone in the United States.” The story is similar at FAÇONNAGE: “It’s really hard to work like this. We really get the impression that the site managers have no room for manoeuvre. We’re always telling them they’re useless to us, we say ‘Introduce us directly to the group HR managers so that they can negotiate with us.’ As soon as we ask them for something, it’s ‘ah, well, we’ll take it to the group to find out if it’s okay.’” (FAÇONNAGE ER).
There is a huge gap between managers who have been able to rebuild unity through often informal means (Jalette and Barton, 2021) and employees whose collective power is still scattered. Collective representation bodies should enable employee representatives to be informed and consulted on all decisions affecting them, so as to make their situation less asymmetrical.
3.2. Obstructing the establishment of local bodies
The framework for collective representation in fragmented companies is designed to ensure interaction between local and central employee representative bodies. In establishments with centralized bodies, this interaction remains insufficient (Table 4). For those where it is compulsory to set up a works council, more than one establishment in ten—and even one in five for groups—does not have a local body.
Table 4
Absence of local bodies
Scope: establishments with 50 or more employees, in the commercial and not for profit sectors, excluding agriculture.
Our analyses of the case studies show that managers can position the social dialogue space at a level that enables them to reduce the cost of coordination between establishments or sites. This choice can be particularly advantageous given employees’ limited power at the central level. In other words, it’s all the easier for management to formally approve a centralized body when employees lack the associational power resources to act within it. In sectors with a relatively strong union presence and high skill levels, companies may tend to prevent access to a central forum for social dialogue. Conversely, when trade union organizations are poorly organized or have to deal with a fragile (e.g. low-skilled) workforce, companies may seek to centralize social dialogue. The contrast between the pharmaceutical and cleaning sectors is particularly enlightening. To use Hirschman’s vocabulary: the more potential voice employees have, the less management encourages the creation of a place where it can be expressed (Doellgast, 2010).
For instance, NETTOIE set up a UES which, on the face of it, appears to be the most appropriate level for exchanges within the group. However, the interviews revealed that by creating this body at the centralized level, the group was able to avoid setting up Social and Economic Committees (CSEs) at the local level, i.e. in each of the branches, which employ well over 50 people. This is not an exception, as 13.6% of establishments with 50 or more employees belonging to a UES do not have such a committee (Table 4). The example of NETTOIE’s UES shows not only that the decision to create it was a matter of employer prerogative (“Management decides on the UES . . . It doesn’t do anything special,” NETTOIE ER), but also that it is not driven by social dialogue concerns. As Grumbach pointed as early as 2012, legal recognition of the UES, which for a long time was systematically initiated by employees, is now frequently the result of requests from employers, who are prepared to go to court to obtain it when employees have refused the creation of this new structure, deeming it unsuitable for negotiating relevant agreements. Some explanations for employers’ interest in the UES include the reduced number of employee representatives resulting from the introduction of a centralized body, the destabilization of existing bodies as a result of elections following its creation, and the fact that recognition may lead to a review of collective agreements.
3.3. Configuring the entity to suit management needs
Outside the context of the fragmented company, the structuring of the collective representation body seems to be codified in a way that leaves less room for “institutional toying.” As the centralized body is intended to structure a social dialogue straddling several sites or companies, this framework still requires adaptation and even, since the 2017 reform, partial negotiation as the employer retains the final say in the absence of agreement. These margins for manoeuvre have been widely documented (Farvaque, 2019), but fragmentation gives them a particular significance.
BANQUE, made up of two sites, provides a first example. The sites have not been recognized as “distinct” establishments, and a single Social and Economic Committee (CSE) represents the collective workforce. In practice, the CSE focuses mostly on the main site, where management and support functions are also located. The absence of a committee at each site is not due to the number of employees at the secondary site (250), but to the company’s decision not to grant autonomy in human resources matters to the site manager, a criterion that provides an opportunity not to recognize the establishment as “distinct” and therefore not to set up a CSE at each site. Elected representatives from the main site may reserve places on the CSE electoral lists for representatives from the remote site, as they see fit. However, their specific interests tend to be partially overlooked, despite their position being more economically fragile: the “secondary” site is a call centre whose activity is implicitly in competition with another call centre to which the group entrusts services of the same type, whereas the employees at the main site are both more qualified and belong to the core banking business.
The way in which collective representation operates in the cleaning sector sheds further light on the ability of managers to choose the form of centralized body. Indeed, the two case studies from this sector show that, when it comes to collective representation, the options chosen can be diverse. NETTOIE created a UES, while JAVEL is structured with establishments and has a central CSE. But more than that, these two cases demonstrate that managers have the latitude to influence the choice of representatives who will sit on the centralized body. More generally, the creation of a centralized body raises the question of how different production sites will be represented: which establishments will have representatives on the central CSE? Which companies will be represented on the CSE of the UES or CG?
While the employer’s interference is obvious for NETTOIE, where only one branch (out of 12) is represented because the head office has built the UES CSE list, the difficulty appears to be purely technical[11] for JAVEL given the number of branches. Indeed, the rules governing the formation of JAVEL’s central CSE mean that only 25 sites (out of 113) can be represented. However, far from being purely technical, this situation raises the question of which branches should be represented on the centralized body. The criterion recommended by the legislation, which seems more consensual, is the size of the establishment: the largest establishments (or companies, for a UES) are represented. Yet, while the average size of a branch is generally a result of the organization of production specific to the sector, its exact size depends on the division of activity within the territory, which is fairly flexible and clearly in the hands of management. According to employee representatives, JAVEL will not hesitate to subdivide a large entity in which the representatives are too combative, so as to exclude them from the central committee, or to merge several branches displaying reduced activity, so as to trigger new elections.
Our case studies allow us to go beyond a one-dimensional approach to corporate fragmentation (i.e. companies reject centralized bodies) by highlighting the diversity of strategies implemented to configure these bodies: refusal, avoidance, choice of institutional form and location. These are, in fact, dependent on the economic and social context, and sometimes the acceptance of an entity becomes a means of isolating certain employees by attaching them to a “centre” that does not correspond to the place where their interests are most visible or easily defensible.
4. Fragmentation as a means of weakening employees’ power resources
The latitude that managers seem to have in setting up a collective representation space is made possible by employees’ low level of interest in this matter. Indeed, while the creation of a UES may come from a court decision, failure to take the initiative or refusal to organize elections exposes the employer to criminal prosecution. Having the right to a centralized body is therefore not sufficient for it to exist.
This glaring discrepancy between the opportunities offered by the legal framework and the “non-use”[12] of bodies could lead to criticism of union inaction, or point to employees’ inability to get organized and make demands. Our qualitative analyses reveal that the situation is both more complex and more diverse. Indeed, centralized bodies do not have the capacity to reduce employee fragmentation (4.1). The low level of demands from unions may therefore be the result not only of discouragement processes, but also of workforce divisions, which can sometimes be explicitly brought about by company management through functional divisions and encouraging competition between sites (4.2).
4.1 Centralized bodies incapable of reducing fragmentation
In the cleaning sector, the existence of a centralized body does nothing to alleviate the fragmentation of cleaning staff, who are dispersed among a multitude of order givers. The scope covered by centralized bodies does not appear to be conducive to improving employees’ situation: the most significant working conditions are scarcely discussed within the body, their definition being referred instead to the branch when it concerns wages, for example, and to the order giver when it concerns the most concrete points of working conditions (Nizzoli, 2015). Thus, cleaning agents’ representatives cannot act on the aspects that have real economic impacts, because the competitive conditions (and the effects of the legal framework for awarding commercial contracts in the sector) foster an agreement on this point between employers to transfer them to the sectoral level; nor can the representatives act on the day-to-day aspects of working conditions, which depend on another company (Devetter and Valentin, 2021). These difficulties are compounded by the distance between employee representatives and the “rank and file,” which is particularly detrimental given the fragility of the workforce. In this sense, the creation of a central CSE or a UES CSE is a legal response to a much broader problem.
Finally, even though centralized bodies have been set up in formal terms, they do not provide a forum for collective expression and consideration of employees’ interests: a forum for discussion may have been created within the scope of the fragmented company, but employee representatives do not seem able to take advantage of it. The fragmentation of companies gives rise to a problem of identifying the place and the (employer) players who are supposed to participate in social dialogue (Denis, 2018), but above all it reinforces the inequality between the employer and employee positions. While particularly visible in the cleaning sector, this situation can be observed on a much wider scale: the BANQUE example discussed earlier shows that the existence of a body shared by both sites is not sufficient to create a space that genuinely integrates the remote site.
4.2 Lack of entities reflecting competition between employees
Beyond the lack of means at their disposal to bring social dialogue to life at the central level, employees and their union organizations seem to struggle to build a community of interests. At first glance, employees may appear to have no desire for a centralized body. However, this apparent lack of demand can be understood as a result of the fragmentation strategies implemented by employers. This is illustrated by the two case studies from the pharmaceutical sector.
In this sector, high profitability and the branch’s role in wage matters, which is limited to setting minimum wage levels rather than eliminating competition in hiring, should motivate employees to join forces in wage negotiations. Yet this is not what our two field examples show. Despite marked differences between the two corporate groups,[13] they illustrate Freyssinet’s (2005) assertion that “groups are very adept at placing their various units in competition with each other” (p. 325) [Translation]. In both cases, employees and their representatives feel that their interests are best protected by remaining “each in their own corner.” At FAÇONNAGE, the main aim is to preserve some of the employment conditions acquired when the sites belonged to a multinational pharmaceutical laboratory. Indeed, since each group site came from a different divesting laboratory, each site has its own specific heritage and culture, whether in terms of work organization or certain social agreements carried over at the time of divestment. In this context, and despite the fact that the group’s configuration should allow for the creation of a group works council (CG), the legal segmentation is sufficient to extinguish this potential request from representatives. The interviews highlighted the extent to which employees, fresh from bitter negotiations to maintain some of the benefits they enjoyed before being transferred to the group, believe they have obtained better conditions than those in other companies belonging to the group.
As for LABO, the way in which management is working to build divisions between employees can be seen in the restructuring steps that followed the representatives’ first request for recognition of a UES. It was immediately after this episode that management carried out an activity-based reorganization, structuring the group by socio-professional categories, then by trade union organizations thanks to functional separation. Whereas head office is staffed 90% by managers, the plant is staffed mostly by blue-collar workers and technicians, and this is directly reflected in the union organizations present. Employee representatives from both LABO and its head office are somewhat reluctant to introduce a UES as it could destabilize the equilibrium between unions. As LABO’s ER explained: “The problem is that you have to reach an agreement with the elected representatives from the other sites . . . on which expert to call in, on this, on that, and we already know that we won’t agree with the elected representatives from head office.” The head office ER added: “Basically, all the union organizations are saying ‘yeah, but I’m not represented here, so I’ll probable lose.’ So it’s like we’re all just standing there staring each other down, thinking ‘we’ll just stay put.’” This comment is in keeping with the finding highlighted by Giraud and Ponge (2016), who maintain that when faced with this situation, employee representatives prefer to remain local, to avoid the difficult coordination between central and local elected representatives.
5. Conclusion
The growing prevalence and overlapping of these three forms of fragmentation has not been accompanied by changes in centralized employee representation bodies to ensure they are capable of structuring effective collective representation. In France, centralized employee representation bodies—known as group works councils (CGs), social and economic unit (UES) committees and central works councils (CCE)—have been designed as a form of protection or institutional resource for employees in the face of the fragmentation of the production process due to corporate groups, outsourcing and geographical dispersal. Drawing on the concept of “institutional toying” has enabled us to gain a better grasp of the diversity of strategies used by employers to shape collective representation to their liking, and, in this way, to better understand why a significant proportion of establishments are still not covered by such bodies—or conversely, when they are present, why they are partial or ineffective. At the same time, we’ve shown that the potential for institutional toying may itself be the result of employer strategies to split production processes. The fragmentation of work and the division of labour thus produced offer levers for multiplying the possibilities of institutional toying. In this way, company fragmentation becomes an instrument for weakening employees’ power resources, because it reduces their interest and ability to participate in social dialogue, and divides and creates competition within a fragmented workforce.
The expression of employees’ interests therefore requires a framework that promotes the working community involved in the production process, as Delahaie and Perez (2021) have also shown in their analysis of negotiations in times of crisis. This renewal concerns two fronts: associational resources to overcome the divisions created by fragmentation, and institutional resources to rebalance the ability to set the framework for employee consultation.
From the point of view of associational resources, in addition to consultation times common to the entire “company” (and away from management), a common place (real or virtual) and access to expertise would seem essential to overcoming the complexity of the situations generated by remoteness. In the absence of a centralized body, the right to expertise makes it possible to access information on the entire “company” scope.[14]
In terms of institutional resources, a rebalancing between the options available to employers and those available to employee representatives could be achieved by modifying the suppletive law (which applies in the absence of an agreement). The latter is in the hands of managers and allows them, for example, to set the number of distinct establishments or the number of representatives. While it is difficult to influence the production infrastructure, it would be logical to give employees the means to determine, at least as a last resort, the manner in which they are represented. Similarly, management refusals (or “oversights”) to set up these bodies should prompt a reaction from the administration.[15] It appears difficult for employee representatives to demand the creation of these bodies.
More generally, the question of access to representative bodies at the centralized level can be likened to that of access to a social right. Asserting a right is a first step, but it is not enough to make it systematically effective. The beneficiaries of this right also need to be provided with the legal and informational resources needed to claim and assert it. But this second step is also insufficient, as a growing body of literature on the notion of non-use is now emphasizing: a significant proportion of obstructions do not stem from explicit opposition from the opposing party, but rather from forms of self-censorship, varying in their degree of complexity. A lack of visibility of the benefits achieved, psychological costs and the absence of mechanisms for expressing demands are just some of the elements that feed the non-use of social rights. Yet this non-use has a cost. In failing to apply measures designed to improve social dialogue, whether deliberately or through avoidance, organizations will suffer from asymmetrical and inefficient management, and fall short of the objective enshrined in the Labour Code: “to ensure the collective expression of employees so that their interests are constantly taken into account.”
Parties annexes
Notes
-
[1]
In particular with the creation of health, safety and working conditions committees (comités d’hygiène, sécurité et conditions de travail, CHSCT).
-
[2]
See Peskine (2021) on negotiation issues in networked companies.
-
[3]
The REPONSE survey is the French equivalent of the United Kingdom’s “Workplace Employment Relations Study” (WERS) (see Amossé and Coutrot, 2008). It is based on a sample of over 4,000 establishments, representative of French establishments with 11 or more employees in the commercial and associative sectors, excluding agriculture. It provides a description of how employee representative bodies operate and interact within organizations, in relation to human resources management policies, work organization methods, economic strategies and company performance.
-
[4]
The Macron ordinances of 2017 led to the introduction of social and economic committees (CSE) in establishments with 11 or more employees, with the CSE replacing the three bodies that existed at the time, i.e. the works council or establishment committee (CE), staff delegates (DP), and health, safety and working conditions committees (CHSCT). This article, however, focuses solely on the works council, which remains the “matrix of the CSE” (Giraudet, 2018) in its operating rules. The quantitative data, taken from the REPONSE survey, the latest edition of which dates from 2017, relate to works councils (including single delegations that include at least one works council) and the case studies, carried out in 2019-2020, on CSEs.
-
[5]
For order givers, this may involve outsourcing not only activities related to their core business, but also support activities. Among establishments with 50 or more employees, 55.8% were order givers without being order takers, 29.9% were both order givers and order takers, and 2.7% were order takers without being order givers in 2017.
-
[6]
Since 1982, the purpose of the works council has been “to ensure that employees express their views collectively, so that their interests are constantly taken into account in decisions concerning the management and economic and financial development of the company, the work organization, professional training, and manufacturing techniques.” [Translation]
-
[7]
Nevertheless, the merger of bodies has had a significant impact on social dialogue (see special issue 107-108 of La Revue de l’IRES, 2022, coordinated by Delahaie and Fretel).
-
[8]
With the 2017 ordinances, recognition of sufficient autonomy in personnel management at the establishment level, in order to be considered “distinct” and therefore lead to the creation of a CSE at that level, requires a collective agreement negotiated with union delegates or, in certain cases, between the employer and the CSE. In the absence of agreement, the employer may choose the distinct establishments. This places the scope of employee representation under a twofold dependency on the employer, who must sign the agreement and agree to delegate power.
-
[9]
There is no legal obligation to recognize a UES, which comes about either through a collective agreement or through legal action. Its scope is equally complex. A UES can be set up between companies with a subcontracting or financial relationship, or a community of sites, provided that their economic and social unity can be proven.
-
[10]
All interview quotes in this article are translated from French.
-
[11]
The number of individuals that can be elected to a CSE is limited. When a company has two or more distinct establishments, a “central” CSE must be set up, as well as a CSE for each establishment. The number of members elected to the central CSE may not exceed 25 full members and 25 alternates.
-
[12]
To echo the numerous works on non-use of rights and social policies. See in particular Warin, 2016.
-
[13]
LABO and FAÇONNAGE fall into both the group/subsidiary-type of fragmentation and outsourcing-type of fragmentation. However, the group to which FAÇONNAGE’s two establishments surveyed belong is specifically a pharmaceutical subcontractor whose fragmentation stems from external growth, while LABO, which still belongs to a large pharmaceutical laboratory of a multinational whose head office is abroad, finds itself in a dual subcontracting situation: internally to the group head office, and externally to other laboratories (Geymond, 2022).
-
[14]
This was LABO’s primary motivation for requesting recognition of a UES: access to group-level information.
-
[15]
In France, the Directions régionales de l'emploi, de l'économie et des solidarités (Regional departments for employment, the economy and solidarity) are responsible for monitoring labour legislation. It is this authority, for example, which currently receives any objections from representatives to the options chosen by managers when they make use of rights under suppletive law.
References
- Amossé, Thomas and Thomas Coutrot. (2008). “2. Genèse et réalité d’une enquête.” In: Thomas Amossé Ed., Les relations sociales en entreprise: Un portrait à partir des enquêtes “Relations professionnelles et négociations d’entreprise” (pp. 38-68). Paris: La Découverte. https://doi.org/10.3917/dec.bloch.2008.01.0038
- Anner, Mark, Fischer-Daly, Matthew, and Michael Maffie. (2021). “Fissured employment and network bargaining: Emerging employment relations dynamics in a contingent world of work.” ILR Review, 74(3), 689–714.
- Appay, Béatrice. (2005). La dictature du succès, le paradoxe de l’autonomie contrôlé, L’Harmattan.
- Appelbaum, Eileen. (2017). “Domestic outsourcing, rent seeking, and increasing inequality,” Review of Radical Political Economics, 49(4), 513–528.
- Beaujolin-Bellet, Rachel, and Géraldine Schmidt. (2012). Les restructurations d’entreprises. La Découverte, Coll. Repères.
- Benassi, Chiara, and Andreas Kornelakis. (2021). “How do employers choose between types of contingent work? Costs, control, and institutional toying.” ILR Review, 74(3), 715–738.
- Bernhardt, Annette, Rosemary L. Batt, Susan N. Houseman, and Eileen Appelbaum. (2016). "Domestic Outsourcing in the United States: A Research Agenda to Assess Trends and Effects on Job Quality." Upjohn Institute Working Paper 16-253. Kalamazoo, MI: W.E. Upjohn Institute for Employment Research. https://doi.org/10.17848/wp16-253
- Delahaie, Noélie, and Coralie Perez. (2021). “Workforce adjustment strategies and concession bargaining in times of crisis: A qualitative approach based on French case studies.” Industrial Relations Journal, 52(5), 406–422.
- Delahaie, Noélie, and Anne Fretel. (2022). “La dynamique des relations professionnelles en France sous l’influence des réformes institutionnelles? Apports de l’enquête REPONSE et de ses post-enquêtes.” La Revue de l’IRES, 2022/2-3 (N° 107-108), 3-12. DOI: 10.3917/rdli.107.0003.
- Denis, Jean-Michel. (2018). “Asseoir la représentation sur la communauté de travail. À la recherche de la communauté perdue dans la branche de la propreté.” Sociologie du travail [online], Vol. 60 - n° 1 |January-March 2018, published online February 23, 2018, accessed October 2, 2023. URL: http://journals.openedition.org/sdt/1748 ; DOI : https://doi.org/10.4000/sdt.1748
- Devetter, François-Xavier, and Julie Valentin. (2021). Deux millions de travailleurs et des poussières, Les petits Matins, Paris.
- Doellgast, Virginia. (2010). “Collective voice under decentralized bargaining: a comparative study of work reorganization in US and German call centres.” British Journal of Industrial Relations, 48(2), 375–399.
- Doellgast, Virginia, Matthew Bidwell and Alexander J.S. Colvin. (2021). “New directions in employment relations theory: understanding fragmentation, identity, and legitimacy.” ILR Review, 74(3), 555–579.
- Farvaque, Nicolas (coord.). (2019). “Appropriation et mise en oeuvre des Ordonnances du 22 septembre 2017 réformant le droit du travail.” Rapport d’études DARES, N° 1, 110 pages.
- Freyssinet, Jacques. (2005). “Quels acteurs et quels niveaux pertinents de représentation dans un système productif en restructuration?” La Revue de l’IRES, 47(1), 319-336.
- Geymond, Maé. (2022). “Fusions-acquisitions, restructurations et externalisations: quand les mouvements stratégiques récurrents percutent les relations sociales.” Socio-économie du travail,2021(10), 95-127.
- Giraud, Baptiste and Rémi Ponge. (2016). “Des négociations entravées.” La nouvelle revue du travail [online], 8 | 2016, published online June 1, 2016, accessed October 2, 2023. URL: http://journals.openedition.org/nrt/2591; DOI : https://doi.org/10.4000/nrt.2591
- Giraudet, Carole. (2018). “Nécrologie juridique du comité d’entreprise: transformation et succession d’une institution juridique.” La Revue de l’IRES, 94-95(1), 29-56.
- Grumbach, Tiennot. (2012). “La généralisation de l’ouverture de l’appel pour contentieux de l’UES: un effet collatéral de loi du 20 août 2008.” Revue du droit du travail, p. 168.
- Jalette, Patrice and Ruth Barton. (2021). “Stratégies des multinationales pour reconfigurer le périmètre de l’entreprise et l’aire de la négociation collective en Tasmanie.” Négociations, De Boeck Supérieur, 1(35), 89-105.
- Korpi, Walter. (1985). “Power Resources Approach vs. Action and Conflict: On Causal and Intentional Explanations in the Study of Power.” Sociological Theory,3(2): 31–45.
- Marchington, Mick, Damian Grimshaw, Jill Rubery and Hugh Willmott (Eds.). (2005). Fragmenting work: Blurring organizational boundaries and disordering hierarchies. Oxford University Press.
- Nizzoli, Cristina. (2015). C’est du propre! Syndicalisme et travailleurs du « bas de l’échelle » (Marseille et Bologne), Presses universitaires de France, Paris.
- Perraudin, Corinne, Héloïse Petit, Nadine Thevenot, Bruno Tinel and Julie Valentin. (2014). “Inter‐firm Dependency and Employment Inequalities: Theoretical Hypotheses and Empirical Tests on French Subcontracting Relationships.” Review of Radical Political Economics, 42(2), 199‐220.
- Perraudin, Corinne, and Nadine Thevenot. (2022). “Exposition aux risques professionnels et prévention dans le travail en sous-traitance.” Socio-économie du travail, 2021(10), 193-230.
- Peskine, Elsa. (2021). “Établissement, entreprise ou groupe – La négociation d’entreprise en quête de ses espaces.” Négociations, n° 335, 2021/1, 39-51.
- Refslund, Bjarke, and Jens Arnholtz. (2022). “Power resource theory revisited: The perils and promises for understanding contemporary labour politics.” Economic and Industrial Democracy, 43(4), 1958–1979.
- Thébaut-Mony, Annie. (2008). Travailler peut nuire gravement à votre santé, La Découverte/Poche.
- Thevenot, Nadine (coord.), Geymond, Maé, Perez, Coralie, Perraudin, Corinne, Valentin, Julie and François-Xavier Devetter. (2021). “Entreprise ‘éclatée’ et périmètre de la représentation collective.” Rapport d’études DARES, n° 9.
- Thevenot, Nadine, Geymond, Maé, Perez, Coralie, Perraudin, Corinne, Valentin, Julie and François-Xavier Devetter. (2022). “Face à l’éclatement des entreprises, une représentation collective mise en défaut: une analyse à partir de l’enquête REPONSE 2017.” La Revue de l’IRES, vol. 107-108, no. 2-3, 13-40.
- Visser, Jelle. (2016). “What happened to collective bargaining during the great recession?”, IZA Journal of Labor Policy, 5(1), 1–35.
- Warin, Philippe. (2016). Le non-recours aux politiques sociales. Presses Universitaires de Grenoble.
- Weil, David. (2014). The Fissured Workplace. Harvard University Press.
- Wright, Erik Olin. (2000). “Working-class power, capitalist-class interests and class compromise.” American Journal of Sociology, 105(4), 957–1002.
Liste des tableaux
Table 1
Forms of fragmentation and their combination
* The question on the use of outsourcing was modified in 2011 to include all forms of outsourcing, which may explain the lower proportion of establishments concerned by outsourcing and the three forms of fragmentation in 2005.
Scope: establishments with 11 or more employees, belonging to a company with 50 or more employees, in the commercial and not for profit sectors, excluding agriculture.
Table 2
Local and centralized bodies in fragmented companies (% of establishments)
Table 3
Summary of the characteristics of the establishments surveyed
Table 4
Absence of local bodies