Soft procedures for hard impacts

The ESSD regulatory capacity in theory

Social partners have been associated with the European decision-making process from the early decades of the European Community. Informal joint committees were established very soon in sectors affected by the first European regulations and the launch of the European Coal and Steel Community and the European Economic Community (Dufresne, 2006). Such bodies played at first a consultative role, but the role and status of committees evolved throughout the integration process, and the number of sectors having their own committee grew (Degryse and Pochet, 2011), until there was broad coverage of manufacturing and service activities with 43 committees by July 2016 (Degryse, 2015).

A major turning point occurred in the 1980s when Jacques Delors, then President of the European Commission, launched at Val Duchesse the European bipartite social dialogue (Didry and Mias, 2005). A few years later, the Social Protocol annexed to the Maastricht Treaty provided that ‘before submitting proposals in the social policy field, the Commission shall consult management and labour on the possible direction of Community action’ (Article 138 EC). Social dialogue included consultation by the Commission and, besides, a possibility for bipartite negotiation: dialogue between management and labour could ‘lead to contractual relations, including agreements’ (Article 139 EC). ¹

Therefore, social partners formally became neo-corporatist actors in European social politics, and gained the status of ‘co-legislators’ able to produce their own rules and norms (Falkner, 1998). Several scholars considered in a positive light the rights that were granted to social partners and thought that this would lead to the emergence of a form of Eurocorporatism (Didry and Mias, 2005; Falkner, 1998; Welz, 2008).

In 1998, the European Commission decided to formalize and to institutionalize further the formal or informal working groups that existed so far and the social partners were encouraged to sign joint texts (European Commission, 1998). Furthermore, the Commission promoted the creation of new sectoral committees. Among the most recent ones are the committees for the graphical industry and for ports, both set up in 2013. In terms of outcomes, the committees adopted, from 1978 to 2013, 734 joint texts. Yet, the agreements, joint texts converted into Directives or implemented in accordance with specific national social partner or Member State practices, account for only 2 per cent of the texts adopted (Degryse, 2015).

Joint texts mainly consist of three types of instruments: frameworks of action, guidelines and codes of conducts, and policy orientations. All belong to the sub-category of process-oriented texts: ‘this category consists of a variety of joint texts which are implemented in a more incremental and process-oriented way than agreements. In these texts the European social partners make recommendations of various kinds to their members for follow-up […]’ (European Commission, 2004). These instruments do not entail formal transposition of their provisions in the legal framework of each Member State, but rather provide guidelines and orientations to national social partners. In other words, these texts rely on a soft non-binding process for implementation. The European social partners make recommendations but have no constraining power towards their national member organizations. In order to follow the implementation process, they initiate regular evaluations of the progress made in the Member State and organize diverse follow-up procedures such as written surveys, periodic reports or conferences. However, the follow-up procedures are often piecemeal and patchy because these mainly depend on the ability of the European social partners for their organization and the good will of their national affiliates to provide feedback (Pochet et al., 2009).

The prevalence of soft texts, combined with limited information on implementation and impact of texts, raises questions about the ability of the ESSD to affect national employment relations (Keller, 2008; Smismans, 2008). So far there has been little evaluation of the implementation of the ESSD joint texts and previous studies have been rather sceptical in this regard (Keller and Weber, 2011; Pochet et al., 2009). Researchers highlight the difficulty in quantifying implementation and in monitoring the impact of the texts at national level (Weber, 2010). Relying on formal benchmarks such as collective bargaining and coverage rates, they come to the conclusion that soft instruments produce limited results (Keller and Weber, 2011).

Regulatory capacity in practice: political processes and multi-level strategies

The joint texts concluded at European level do not strictly correspond to what is called a collective agreement in the various Member States: they do not have a similar legal and constraining status that fits with the institutional specificity of each collective bargaining system. As Marginson and Sisson already noted in 2004, European-level processes should not be assessed against the characteristics of long-established industrial relations practices in national contexts (Marginson and Sisson, 2004). This also applies to the outcomes of the European social dialogue. Therefore it is not fully legitimate to evaluate the status and the intentions of texts resulting from the ESSD against the characteristics and effectiveness that collective agreements have within each industrial relations system.

Rather, one can consider that soft non-binding joint texts are close to new modes of governance (Pochet, 2007). Studies looking at soft EU instruments mainly refer to their learning potential and procedural impacts. These studies consider the domestic impact in terms of ideational influence, as well as according to the redistribution of power that they may foster (Jacobsson, 2004; Zeitlin, 2009). Yet, from a European policy-making perspective, researchers usually see new modes of governance as strongly dependent on ‘the shadow of hierarchy’, defined as legislative and executive decisions, and effective only when there is a ‘threat’ of possible legislation. Smismans (2008) considers that agreements and joint texts of the European social dialogue can be neither reached nor implemented effectively without the intervention of European institutions and the ‘threat’ of possible legislation.

From an industrial relations perspective, questions of legitimacy and efficiency relate to the diversity of collective bargaining systems in Europe, because the implementation of outcomes relies on a diversity of the statutory status of collective agreements, diverse processes of collective bargaining, and different rates of collective bargaining coverage. Diversified industrial relations not only lead to a high degree of complexity in tackling issues discussed in the European sectoral committees, they also complicate implementation processes because the notion of agreements has different legal meanings in different national contexts, and not all sectoral social partners have a similar capacity to act in their respective domestic system. Some scholars thus consider that the joint texts cannot guarantee implementation of common minimum standards in the Member States (Keller, 2008).

If one takes the nature and requirements of process-oriented texts into account, on the one hand the outcomes of the ESSD should not be appraised only according to their strict regulatory impact assessed in the terms used to evaluate traditional collective agreements. On the other hand, assessing the impact in terms as general and vague as ‘cognitive processes’ or ‘learning outcome’ raises conceptual and empirical difficulties: how to evaluate any ‘cognitive’ impact or how to measure learning processes? If being too narrow in the criteria used may lead to unjustified criticisms that do not reflect the very complexity and specificity of European-level social dialogue, being too broad bears the risk of insufficient conceptual precision and of seeing causality where there is only coincidence.

It is therefore relevant to use a wider notion of implementation of process-oriented texts in terms of the influence or impact that they have on initiatives undertaken by the national social partners. Broader and softer benchmarks than transposition into a collective agreement must be used to assess the impacts of the ESSD texts. This includes cognitive and procedural dimensions. Implementation, in this perspective, covers not only formal transposition into a given type of domestic text, but also other processes that relate to local activities initiated as a consequence of a given text.

This perspective, in turn, requires further conceptualization of relations between European-level texts and activities undertaken within domestic contexts, in order to understand why trade union and employer associations within Member States become committed to implementing a given text.

When examining the implementation of the European social dialogue agreements and joint texts, scholars have identified complex linkages between the national structures and the European level, which may then be an obstacle to attempts at European-level regulation (Léonard, 2008; Prosser, 2012; Weber, 2010). Systems of industrial relations differ widely among the 28 Member States. Procedural rules, levels of collective bargaining, hierarchies between the different levels, statutory status of collective agreements, and the parties entitled to conclude collective agreements vary from one country to another and, within each country, from one sector to another (Traxler et al., 2001). Each time that implementing EU agreements or joint texts requires any kind of action from the national constituencies, these characteristics shape different capabilities for the national actors.

In addition, the lack of interest of the national social partners in the topics addressed by the texts, as well as their unwillingness to implement the texts at the national level, have also been highlighted as explanatory factors of difficulties in implementation (Prosser, 2012). Such findings highlight the complexity and the diversity of aspects that should be taken into account.

Keune and Marginson (2013) propose to draw on multi-level governance perspectives to understand the complexity of industrial relations in Europe. In their view, the emergence of transnational levels in industrial relations provides new opportunities and simultaneously adds new limits and constraints to the national social partners’ ‘scope of action’. The approach can be applied to understand why national trade unions and employer associations participate in the European sectoral social dialogue committees, while doing so on a voluntary basis. In a multi-level environment, opportunities for action are increased, because there is some duplication of fields for potential action (Pralle, 2003).

Keune and Marginson (2013) also consider that complexity of multi-level governance cannot be fully grasped by traditional institutionalist approaches of industrial relations. For them, the vertical dynamics between the European social dialogue and the national level cannot be considered only on a top-down basis. The implementation of a given text resulting from the ESSD should not be seen as a descending process whereby national-level constituencies just adopt or reject the content of the text, and transpose it or not within their own institutional structure. Instead, it must be considered as a process in which national-level players and European players interact and, in these interactions, each player’s interests, preferences and resources intervene.

Implementation, in this view, is defined as initiatives undertaken by the national social partners as a consequence of a given text discussed at European level. And one can suppose that implementation takes place when national players see an interest in the text and have the capacity, in terms of resources and strength, to act in their domestic context in relation to this text. This, in turn, depends on the interest that they have in the specific topic covered by the text. Understanding implementation requires, therefore, looking at upstream processes whereby national constituencies give a mandate to European social partners to discuss that specific topic in the ESSD committee. In this perspective, the regulatory capacity of the European sectoral social dialogue depends on the involvement of national social partners in the committees and the strategies that these and the European players develop when interacting at both levels.

Implementation and impact of texts in Member States

In line with the conceptual discussion presented above, implementation is considered effective when national actors fulfil the requirements comprised in the European text by establishing new structures, developing new practices or organizing dissemination activities.

The interviews with European and national actors, analysis of European federation follow-up reports and internal documents indicate that national trade unions and employer associations undertake different types of action as an outcome of a joint text signed at European level. Table A1 in the Appendix presents the follow-up results of the implementation of three joint texts. The first text is the ‘Code of conduct on ethical cross-border recruitment and retention in the hospital sector’. The other two were concluded in the agriculture sector: the ‘European agreement on vocational training in agriculture’ and the ‘European agreement on the reduction of workers’ ‘exposure to the risk of work-related musculo-skeletal disorders in agriculture’. ² The table shows a diversity of actions ranging from the creation of new structures such as a national observatory for workers’ health and safety in agriculture to the development of collaboration for the exchange of information and practices.

At first, process-oriented texts aim at promoting and raising awareness about common standards and principles. In the context of these texts, national social partners are mainly expected to inform their members about the existence and content of each text and to raise their awareness about the issues at stake and the principles promoted by the texts.

In the four sectors, national sector-level organizations organized information campaigns, conducted studies and even monitored the follow-up at company level. Such dissemination activities mainly affect the national actors on a cognitive level by fostering an ideational and discursive change mainly based on the use of a common language and common principles (Jacobsson, 2004; Zeitlin, 2009). The European texts therefore lead the national employers and trade unions to adopt a new cognitive framework, shift their agendas and pass the information to their members and other national actors.

In a context of severe staff shortage in the hospital sector, the code of conduct on hospital recruitment aims to promote ethical practices and to stop unethical practices in cross-border recruitment of health workers. To this end, the German trade unions organized information campaigns as well as dissemination activities via their websites and newsletters, to inform their affiliates as well as foreign workers about their rights in terms of recruitment. In addition, one trade union organization provides individual advice to foreign workers and thus ensures that clear information about recruitment is given (interviews with German trade union organization in health sector, national internal documents, 2012).

Following the text on the prevention of musculo-skeletal disorders (MSD) in agriculture, one Bulgarian trade union reported having organized several meetings with the labour inspectorate department in order to foster the elaboration of measures in matters of risk prevention. In the same vein, the organization representing the Bulgarian farmers organized seminars for their affiliates in order to raise their awareness about the importance of risk prevention programmes in farms (interviews with Bulgarian representatives of the agriculture sector, 2012).

As a third example, following the 2005 Joint Statement of the EU social partners in the postal sector on CSR, Poste Italiane adopted, with the trade unions, a protocol of understanding on corporate responsibility. The objective was to inform and involve employees in social responsibility initiatives (interviews with Italian officials from social partners’ organizations in postal services, 2009).

Beyond dissemination activities, European joint texts can lead national constituencies to develop new practices or norms. Such practices are developed either by the means of joint or by unilateral actions, and sometimes they can be integrated into a collective agreement.

The analysis conducted in the hospital sector and agriculture shows that the three joint texts studied led the social partners to adopt new substantive practices in Belgium, Denmark, France, Germany, the Netherlands, Spain, Sweden and the United Kingdom.

In agriculture, the European joint text on vocational training aims at harmonizing national vocational training systems with a view to facilitate worker mobility. One of the provisions recommends that national social partners develop measures relating to the validation of skills acquired by vocational training. In this framework, the Swedish social partners in the sector entered into negotiation and reached an agreement intending to develop a system for validation of vocational skills, and they initiated the green card, a specific skills passport for jobs in agriculture (interview with Swedish social partners, EU and national internal documents, 2012).

In the hospital sector, the code of conduct led the Dutch social partners jointly to create a blacklist identifying recruitment agencies that do not respect the ethical principles promoted by the code (interview with the Dutch national social partners of the hospital sector, EU and national internal documents, 2011).

However, the Joint statement on corporate social responsibility in the commerce sector as well as the Joint statement on CSR in the postal sector did not lead to any substantial impacts at national level.

In addition, the empirical research highlighted other types of actions undertaken by the national social partners in the framework of the European joint texts. On the one hand, in some cases, national actors use a given European text as a lobbying tool, in an attempt to put pressure on their national authorities. Social partners in the Latvian health sector reported that they use the principles promoted by the 2008 code of conduct on recruitment to legitimize their demand for better funding of public health (interviews with Latvian social partners in the health sector, 2011). Lobbying activities were also found in Romania where trade unions with the help of their European federation tried to put pressure on the government to stop the massive emigration of medical staff (interviews with Romanian trade unions in the health sector, 2011).

On the other hand, European joint texts can initiate collaboration or exchange programmes between national actors. In agriculture, AGRITRANS is a joint project that brings together trade unions from nine countries with the aim of exchanging information and good practices on vocational training. The trade unions involved exchange information about the organization of their vocational training system and develop joint initiatives related to a Register of Agricultural jobs. Such initiatives helped the Bulgarian agriculture trade union to be informed on potential measures aimed at improving the vocational training system (interviews with social partners in agriculture, EU and national internal documents, 2012).

An official from the Portuguese Post reported that the ‘Joint statement of the EU social partners in the postal sector on CSR’ of 2005 was followed by a series of seminars dedicated to exchange of good practices. Such activities encouraged actors to build sustained relationships with other European partners. For instance, the Portuguese Post developed a plan on ‘equal opportunities’ directly inspired by a practice presented by the French La Poste (interviews with Portuguese officials from the postal services sector, 2009).

Empirical data thus show that soft joint texts have an impact on the national context. They foster exchange of knowledge or at the procedural level help national actors to gain increased bargaining power. They can even create an ideational and discursive change while in some cases they also directly affect the content of national regulations and practices of the national social partners. Yet, if this indicates that the ESSD joint texts have some regulatory value, these findings are only effective in some countries. An absence of implementation was also found in many cases. The question, then, is what processes lead to effective implementation and what prevents implementation.

Political processes and multi-level strategies at play

Why do sector-level trade unions and employers’ associations in the Member States become committed to implementing a given text resulting from the European social dialogue, especially when this text has no constraining power? In order to understand the multi-level process that is at play, it is necessary to examine players’ interests, resources and capacity, and their interactions with their European representatives.

A first sub-question is the extent to which the European social dialogue provides opportunities for national-level and for European-level social partners.

The first and main motivation expressed by respondents for being interested in the work done in their ESSD committee is a will to influence the European decision-making process and to avoid any initiative from the Commission or the Council that could exert a negative influence on their members. In addition, participants in committees are conscious that some issues require transnational coordination to be dealt with. These trade union and employer representatives therefore consider the European social dialogue to be a tool potentially useful for dealing with challenges related to internationalization or to European integration.

In the hospital sector, albeit that public health remains mainly rooted in the national welfare systems, health policies are increasingly Europeanized and challenged by trends associated with internationalization. In such a context, the code of conduct on transnational recruitment aims at dealing with the increasing mobility of health care staff and related problems such as staff shortages, the recognition of qualifications and working conditions of foreign workers. According to the respondents, issues related to transnational recruitment are present in each European country and therefore require a European answer. A Swedish trade unionist made the following statement:

it is a big issue for all the countries (…) because either you come from a country that recruits people from other countries or you come from a country that delivers staff to another country … so we are all affected. (interview, Kommunal, November 2011)

Yet, there are significant differences in participation from one Member State to another. Such differences can be explained, according to the strategies developed by the national actors towards the European social dialogue. In some cases, national sector-level organizations consider the European social dialogue as a superfluous level of decision-making because the issues that are dealt with do not fit with their national strategies, or because they consider the domestic level as the only relevant level to deal with such challenges. In such cases, national organizations are not involved or have a limited participation in the ESSD committees.

According to the French and Belgian trade unions in the hospital sector, the existing regulations set higher standards than the provisions recommended at the EU level. They also consider that transnational recruitment is not a solution to the shortage of staff and they prefer to deal with it through negotiations on employment conditions within their own domestic context (interview with Belgian and French trade unions in the hospital sector, 2011).

On the other hand, organizations from Central and Eastern countries are usually very keen to participate because they view the ESSD as a means to gain legitimacy and recognition at EU level. It is also a means to express their concern about low working and living conditions and to find solutions to improve these while their national authorities remain reluctant to act. A Bulgarian representative in the agriculture sector made the following statement:

[The ESSD] is very important for the exchange of information because we can improve our work and at the same time we can improve the work of EFFAT with our specificities and contributions and make them aware of the problems in the regions, in the small countries and in the new member countries. (interview with a Bulgarian trade unionist, 2012)

From the analysis of internal documents such as meeting minutes, it is possible to identify which organizations participate regularly in the ESSD committee meetings, and to delineate their profile according to their domestic institutional position and resources. In order to try to capture the institutional position of the social partners, several dimensions are taken into account: main level of collective bargaining, coverage rates of agreements, and role played by the sectoral social partners in collective bargaining and in public policy-making. In parallel, organization density and the total number of members provide an estimate of organizational resources (Glassner, 2008; Traxler, 2007, 2009).

Empirical findings suggest that the key players in the committees are representatives of significant organizations that possess high resources but also a strong position within their national industrial relations system. Representatives from Belgium, Denmark, France, Germany, the Netherlands, Spain, Sweden and the United Kingdom are closely involved in decision-making processes. They bring their national inputs and develop projects for the European social dialogue. As an outcome of their participation, when a joint text is reached, its content reflects these actors’ concerns and interests. On the other hand, due to a lack of resources, representatives from the eastern countries are usually limited to a more peripheral involvement.

When it comes to implementation, the national trade unions and employer associations who influenced the drafting of the texts are then more likely to undertake initiatives to fulfil the European requirements. Implementation of a given text also relies on the scope for action of the national social partners in their own domestic context. National sector-level trade union and employer representatives who are finally able fully to implement the text are the ones who possess a regulatory capacity within their domestic system of industrial relations, and specifically on the issue covered by the text in question.

The joint text on vocational training in agriculture recommends that national actors develop new measures to improve worker mobility. Such measures concern the validation of vocational skills, but also the involvement of social partners in skills assessment and in the organization of training. The role of trade unions and employer associations in defining and organizing vocational training vary widely from one country to another. While in Germany, Sweden and Belgium, the social partners conclude agreements and participate in joint committees in this field, in Spain, in Bulgaria or in Ireland, they play rather a consultative role. Their institutional ability to intervene on vocational training issues constrains their room for manoeuvre when they are supposed to implement a joint text stemming from the European sectoral social dialogue. When industrial relations are weak, respondents see the regulatory power of public authorities as a major limit to implementation, because social partners then have little or no say in the domain. In terms of follow-up, they are consequently limited to lobbying actions or collaboration.

The impact of the ESSD in Member States thus depends on the strategies, interests and resources of sector-level social partners from the different countries. When the national actors participate in the meetings of ESSD committees, they are able to weigh up the content of the texts, which then reflects their own national interests and challenges. When national sector-level trade unions and employer associations consider that the ESSD can be a potential tool to strengthen their action in their domestic context, they implement the joint texts issued, and they undertake actions at the domestic level.

Implementation cannot be considered as a one-way movement. It is rather the outcome of the multi-level strategy of the national actors. Yet, beyond the interests of trade unions and employer associations, it is also a question of resources and capacity. Participating in the ESSD committees requires that national sector-level organizations have important resources – in terms of people and finance – and a strong institutional position at the national level. Furthermore, implementing a given text strongly depends on the ability of national sector-level social partners to undertake action autonomously in the field covered by the joint texts. Consequently, if the European social dialogue grants new opportunities to the national actors, only national players who already possess capabilities in their industrial relation system can use them effectively.

The impact of texts resulting from the European sectoral social dialogue, in this view, results from a complex process in which national sector-level social partners see an interest in the topic covered by the text, have an interest and resources to become committed with their European counterparts in discussions that lead to the signature of a text on this topic in the European committee and, finally, have the capacity to act in the field covered by the text, and an interest to act, within their own domestic context.