30+ Years of European Social Dialogue: What Way Forward After the EPSU-Case?

Abstract

For 30+ years the European Social Dialogue (ESD) has been an important part of European Union law-making in the field of social policy, which makes it a good moment to look back, but mostly to look forward. The latter, especially after the rulings in the EPSU-case, which were criticised as meaning the end of the European Social Dialogue. The aim of our contribution is therefore to interrogate and assess the future of the ESD through an empirically grounded analyses of ESD and its dynamic relationship to the evolutionary process of European integration. Our conclusion is that in terms of input and output legitimacy, the ESD can improve, especially by building coalitions, and by improving the quality of their output by applying the Better Regulation principles to their agreements. Both activities would lead to the further maturing of their autonomy and might lead to a greater respect for horizontal subsidiarity by the Commission.

1. INTRODUCTION

The European Social Dialogue (ESD) has been an important part of EU law-making in the field of social policy since the adoption of social agreement with the Treaty of Maastricht in 1992. With the ESD running for over 30 years, it is a good moment to look back on what has been achieved and to look forward to what could be good paths to continue. The latter particularly after the EPSU-rulings, have been perceived as highly unsatisfactory, especially by labour law scholars.¹ Comments include that these rulings ignore the importance of social dialogue in the field of social policy since they deny the existence of horizontal subsidiarity, disrespect the autonomy of social partners, and therefore will have a huge detrimental impact on the ESD.² What strikes us in these comments is that most of them seem to be based on certain interpretations of horizontal subsidiarity, autonomy and the Commission’s role in law-making that may exist in theory, however, little account is given of how these concepts work in practice. The aim of this contribution is therefore to interrogate and assess the future of the ESD based on more empirically grounded analyses of the ESD and its dynamic relationship to the evolutionary process of European integration.

Thereto, this article is structured as follows. In section 2 we will briefly elaborate on the development of the ESD in general and on the concepts of horizontal subsidiarity and autonomy of social partners in the EPSU-rulings. Section 3 will focus on the autonomy of social partners from the perspective of the ESD’s capacity to enhance the input and outcome legitimacy of EU legislation and policies. This is an interesting angle because enhancing the legitimacy of EU legislation was one of the reasons the ESD was introduced. Horizontal subsidiarity and the social partners’ autonomy are key aspects in enhancing the legitimacy of the EU’s legislation and policies. Section 3 turns to the changed position of the European Commission in EU policy- and law-making (3.A), including Better Regulation and REFIT (3.B), and how this has affected ESD (3.C). In section 4 we offer some reflections on what seem to be interesting ways forward for the ESD after the EPSU-case based on how it has functioned so far in practice rather than how it should or is ought to operate. Section 5 concludes.

2. THE EUROPEAN SOCIAL DIALOGUE AS A GOVERNANCE MECHANISM AND THE EPSU-RULINGS

The ESD is regulated in articles 154 and 155 of the Treaty on the Functioning of the European Union (TFEU). In short, the ESD enables social partners organised at the EU level to take over the legislative process from the Commission. Thus, instead of the Commission sending a proposal for legislation in the field of social policy to the European Parliament and the Council, social partners can negotiate an agreement (art. 154 TFEU). Once they have reached an agreement, social partners can choose to either implement the agreement themselves ‘in accordance with the procedures and practices specific to management and labour and the Member States’ or for matters covered by article 153 TFEU, by a Council decision on a proposal from the Commission (art. 155(2) TFEU).³ As such, the ESD is one of the governance mechanisms of the EU in the field of social policy.⁴ Other mechanisms are the legislative procedure (article 153(2) TFEU) and the open method of coordination.⁵

The second implementation option for the agreement of social partners, via a Council decision, has raised several issues over the course of time. In two occasions those issues resulted in cases with the Court of First Instance (CFI) or General Court (GC) since the Lisbon Treaty. In the first case,⁶ the employers' organisation UEAPME argued that the 1996 EU Parental Leave Directive, negotiated and agreed by the social partners, should be annulled because UEAPME and therewith the voices of small and medium-sized enterprises had not been consulted. The CFI dismissed the case but set some guidelines on formal issues, including the representativeness of the social partners involved with the agreement and the legal status of the decision of the Council. So far, these Council decisions have taken the form of minimum standard directives.⁷

In the second case,⁸ trade union EPSU claimed that based on article 155(2) TFEU, the Commission was obliged to send a proposal to the Council to implement the agreement on information and consultation for central government workers and civil servants, as was requested by all social partners involved with the negotiation of the agreement. To substantiate their claim, EPSU stated that based on the principle of horizontal subsidiarity, as granted to the social partners by the Commission in its 1993 Communication,⁹ ‘social partners are best placed to assess whether an agreement must be implemented at the level of management and labour and the Member States or at EU level’.¹⁰ Furthermore, they argued that the scope of autonomy of social partners would be reduced, if the Commission would be permitted ‘to refuse, on the ground that it was not appropriate, to submit a proposal for a decision implementing an agreement concluded by the social partners, the latter would in practice be induced to negotiate the content of that agreement in advance with the Commission in order to enable its implementation later’.¹¹

These arguments are based on perceptions and interpretations provided by the Commission itself in its policy documents about the ESD. For example, horizontal subsidiarity is recognised by the Commission in its 1993 Communication on the implementation of the agreement on social policy.¹² Autonomy of social partners has been recognised as a fundamental right in article 28 of the EU’s Fundamental Rights Charter, as something to be respected in article 151 TFEU,¹³ and more generally as a strength of social partners in the 2002 communication in which the Commission recognised the social partners’ ability to ‘undertake genuine independent social dialogue, that is to negotiate independently agreements which become law’.¹⁴

The arguments presented in the EPSU-case seem also to be based on general understandings of horizontal subsidiarity and autonomy.¹⁵ Schmitt, one of the strongest criticisers of the EPSU-case and a leading scholar on the ESD, for example, bases her interpretation on what could be called a European philosophical tradition on the interpretation of subsidiarity.¹⁶ More precisely, for Schmitt the principle of subsidiarity is based on a proximity that reflects ‘the conviction that social groups, and after them, local authorities are best placed, compared to distant public authorities, to regulate relationships and activities of people whose interests they represent’.¹⁷ As such, Schmitt considers subsidiarity as a two-pronged principle: central public authorities, like the EU institutions, are supposed to act only when actions from social groups (horizontal subsidiarity) or local authorities (vertical subsidiarity) are presumed insufficient to achieve the legislative goals. Thus, when it comes to issues such as employment, working conditions, occupational health and safety, working time, information and consultation rights, this interpretation of subsidiarity means that social dialogue and collective bargaining take precedence over public initiatives.

Additionally, for social partners to be able to have a genuine social dialogue and collective bargaining, these activities need to be free from any interventions by public institutions. In other words: social partners need to be able to operate autonomously.¹⁸ This reflects an interpretation of autonomy based on constitutional ideas of collective bargaining as found in the EU’s Member States,¹⁹ an interpretation that Welz presents as a corporatist concept.²⁰ Bogg and Dukes typify it as ‘a kind of “collective freedom of contract” idea’.²¹ In fact, Bogg and Dukes’ analysis of autonomy in the context of ESD underlines that we do not really know what ‘autonomy’ means. And, although everyone favouring ESD seems to agree that more autonomy in the ESD is a good thing, Bogg and Dukes argue that different actors ‘might have intended to defend rather different—and potentially, even, conflicting—sets of values’.²² In total they identified four different uses of autonomy that have been dominant in the political and academic discourse about the ESD.²³

The above leads to the conclusion that both, horizontal subsidiarity and autonomy, are rather elusive concepts in the context of the ESD. Consequently, it has left the EU courts much room for its own interpretations of these concepts. When these concepts are to be interpreted within the context of the ESD only, this may not be problematic, unfortunately, from a labour law perspective, this is not the case, since the ESD is part of the wider context of EU governance. Thus, as was the situation in the EPSU-case, these concepts must be interpreted in relation to other provisions of EU lawmaking, especially articles 5(3) TEU (principle of subsidiarity) and 17 TFEU. Not surprisingly, the Commission based its arguments on these provisions. Based on article 17(1) TFEU the Commission shall ‘promote the general interest of the Union and take appropriate initiatives to that end’. Combined with article 17(3), which stipulates that the Commission is to carry out its responsibilities without taking instructions from any other entity, article 17(1) TFEU implies that the Commission has a discretionary power to decide whether an agreement of social partners is appropriate to be implemented as EU legislation, that is by a Council decision.

In the EPSU-rulings the courts do acknowledge the autonomy of the social partners, however, the courts also argued that none of this compels the EU institutions to implement such an agreement of social partners at EU level.²⁴ With regard to horizontal subsidiarity, the GC was rather strict in its interpretation which it based on article 5(3) TEU solely, thus not also taking into account articles 154 and 155 TFEU. According to the GC, article 5(3) TEU governs the exercise of the EU’s shared competences with its Member States and therefore expresses only a vertical subsidiarity.²⁵ That the Commission made statements about horizontal subsidiarity in, among others, its 1993 communication on the implementation of the social agreement,²⁶ holds no stand in court since these are ‘devoid of any binding legal force’. This means that such documents ‘cannot be successfully invoked to preclude the interpretation of a provision of the treaties that follows from its wording, context and the purpose of that provision’.²⁷

With regard to the autonomy of social partners, the GC concluded that based on article 152 TFEU this means that the stage of negotiating and concluding an agreement is exclusively the terrain of social partners, meaning that they ‘may engage in dialogue and act freely without receiving any order or instruction from whomsoever’.²⁸ It follows that the institutions, including the Commission, ‘must refrain from any conduct aimed at directly influencing the course of the negotiations or imposing the principle or content of an agreement on the social partners’.²⁹ As such the GC acknowledged the importance of the autonomy of social partners. However, the GC continues, social partners have not been given ‘the power of adopting acts that produce binding legal effects as regards third parties, or even of directly submitting a proposal for a decision implementing an agreement to the Council’.³⁰ Consequently, the autonomy of social partners does not include that the institutions, particularly the Commission and then the Council, are bound to give effect to a joint request of social partners to have their agreement implemented by a decision of the Council.³¹

The ruling of the General Court and was upheld by the Court of Justice of the European Union (CJEU), as it found that the GC made no errors in its interpretations as claimed by EPSU.

3. AUTONOMY OF SOCIAL PARTNERS LEGITIMISING EU LEGISLATION AND POLICIES

Since we already concluded that the concept of autonomy of social partners is elusive, which has not changed with the EPSU-rulings, it is important to specify how we understand autonomy. We do this in two steps. First, building on the work by Bogg and Dukes on autonomy of the ESD, we present a scale of autonomy from weak to strong. Second, building on the works of Verbrugge and Smismans, we assess autonomy by its capacity to legitimize EU legislation and policies. Before doing this, we must make one caveat. We are aware that the ESD comprises a panoply of activities, including tripartite EU policy-making activities, for example in the context of EU strategic programmes like Europe 2020, NextGenerationEU, and the EU’s recovery and resilience plans.³² However, our assessment will be limited to the ESD as a governance mechanism, meaning its role is to conclude agreements as laid down in articles 154 and 155 TFEU.

A. ESD Scale of Autonomy

The agreements that social partners negotiate and conclude based on articles 154 and 155 TFEU reflect a certain level of autonomy in the sense that they can be the result of intensive EU guidance (low autonomy of social partners) or of fully autonomously operating social partners. The ESD is strongly guided when the agreement is the result of two consultations (one on the direction and one on the content of a proposal), strong interference by the Commission during the negotiations,³³ and the agreement is implemented via a council decision (read: directive). In such a situation, autonomy of social partners is considered weak. On the other hand, the autonomy of social partners is strong when the agreement was initiated by the social partners themselves (so no consultation took place), there was no interference by the Commission at all during the negotiations, and the agreement was implemented by the social partners themselves (autonomous). Then there is a group of agreements that fall in between because there was consultation, but only on the direction, the Commission was somewhat involved with the negotiations, but not fully, and the agreement is implemented either by a directive or autonomously. See Table 1 for a more schematic overview of these “types of ESD by their level of autonomy.

B. ESD’s Capacity to Legitimize EU Legislation and Policies Based on Autonomy

Verbruggen³⁵ analysed the ESD as a form of co-regulation as defined in the EU’s 2003 Interinstitutional Agreement for Better Law-making (IIA).³⁶ Co-regulation in the IIA is understood as ‘a mechanism by which the Community legislative act entrusts the attainment of the objectives defined by the legislative authority to parties which are recognised in the field (such as economic operators, social partners, non-governmental organisations, or associations)’.³⁷ As co-regulators, the involvement of private parties, such as social partners, compensates for democratic deficit problems which should enhance the EU’s regulatory legitimacy.³⁸ To assess whether the ESD enhances the EU’s legitimacy, Verbruggen uses Scharpf’s criteria for input and output legitimacy.³⁹ EU policies are input legitimate ‘if and because they reflect the will of the people’ and output legitimate ‘if their outcomes serve the citizens’ interests’.⁴⁰ Input and output legitimacy are related since the latter cannot be achieved without the former, because ‘it is impossible to identify the citizens’ interests without having in place input-legitimacy processes’.⁴¹

The results of Verbruggen’s assessment are interesting in the context of horizontal subsidiarity and social partners’ autonomy on a few points. First, the assessment highlights once again issues regarding the representativeness of EU social partners. While their internal structures and mandates may have improved over the course of time,⁴² it remains disputable whether they are representative enough in terms of absolute numbers of workers and employers they represent due to consistent and rather deep decreasing numbers of members of their national affiliates.⁴³

Secondly, the decreasing level of representation of social partners, makes them heavily dependent on the Commission to obtain the necessary financial means to carry out their tasks in the social dialogue.⁴⁴ Consequently, it is questioned whether they can actually act truly independently, or autonomously.⁴⁵ A conclusion that Verbruggen links to Lo Faro’s argument that social partners should be regarded as quasi-agents of the European institutions in the process of adopting European social regulation.⁴⁶

Thirdly, and last, the fact that the level of representativeness of social partners is falling raises concerns about their capacity to influence the European legislative agenda in the field of social policy. In this context, Smismans⁴⁷ argues that with the Commission as the principle agent of EU law-making it holds a strong hand in what social partners do at all times.⁴⁸ To substantiate this claim, he refers to the fact that many of the framework agreements that have been negotiated by social partners, both, those implemented by a Council decision and autonomous agreements, are in fact following from the Commission’s consultation process or the ‘threat’ of a Commission initiative (the shadow of the law);⁴⁹ have been negotiated with strong Commission interference;⁵⁰ are part of EU policies and strategies that are so prominent that social partners also had to act;⁵¹ or social partners have acted in the EU institutions’ ‘cognitive framework’⁵² on those issues.

(i) Input Legitimacy

Smismans’ observation is interesting as it helps us to understand to what extend social partners have been operating autonomously and therewith enhanced the EU’s input legitimacy, therefore we have updated Smismans’ research. Our assessment includes all social partner agreements (autonomous and implemented via a directive) which have been adopted until July 2023.⁵³ The assessment is based on the official EU documents that were part of the adoption process of the agreements or could be related to the agreement’s topic.⁵⁴  Table 2 provides a summary of the results of our assessment.

What is interesting in this summary table is that it shows that there is not one agreement that cannot be linked to activities employed by EU institutions. At the same time, it is not surprising, in fact, it confirms the results of an empirical study conducted in 2005 by De Boer and others who found that ‘the presence or absence of either European sector-specific policy or European general policy with sector-specific consequences is [ … ] an important explanatory factor behind the establishment of an ESD committee and the achievements of results’.⁵⁵

However, our results show that there may be a slightly changing trend since 2012 indicating that social partners start to act more autonomously. Thus, rather than acting in response to consultations and with strong interference by the Commission, social partners are acting within the ‘lighter’ context of the EU set by its policies and strategies and its more general cognitive frame. This might partly be the result of a decline in the flow of proposals coming from the Commission since Barroso took office which was rather inactive in the field of social policy,⁵⁶ as well as a move from binding to non-binding EU social policy.⁵⁷ Partly it could also be contributed to the fact that social partners have matured. For example, the 2015 agreement on information and consultation in central government administrations, which led to the EPSU-case, was negotiated after the first round of consultation (thus only about a possible direction). Social partners picked up on it, even though the Commission concluded after this round of consultation to not further pursue the topic.⁵⁸

Since there seems to be a slight change in the autonomy with which social partners operate, we tend to conclude that social partners are increasingly contributing to the legitimacy of EU policies with their input. However this remains within the context of the EU, especially its cognitive framework. The latter conclusion can be further substantiated by the results of a study we conducted on the content of the work programmes of social partners which they have been adopting since 2003. These work programmes are interesting in this context, as they give an impression of what social partners will be focussing on in their activities. About a quarter of the planned activities in the work programmes deal with the capacity building of social partners themselves. Most of the material topics addressed in the work programmes are concerned with digitalisation, lifelong learning, training, education, skills, youth employment, and topics related to EU governance such as the 2010 Lisbon Agenda. None of these topics are the traditional topics found at the national level of collective bargaining, such as wages and annual leave days. Instead, these are all topics that fit very well with the Commission’s agenda. It becomes more interesting though when we focus on the type of activities social partners planned in their work programmes, since about one-third of the activities are lobbying activities that are directed at the EU institutions, including the Commission. This suggests that, even though social partners follow the Commission’s agenda, they have an ambition to influence this agenda⁵⁹ and as such provide the EU institutions with input which could further legitimize the policies of the EU.⁶⁰

(ii) Output Legitimacy

The assessment of the ESD’s output legitimacy on the EU’s social policies is not overly positive either, however, opinions in academic writing differ. As Verbruggen notes, when the outcomes of ESD are assessed on their quantity, it can be qualified as positive since social partners have produced many joint texts, such as guidelines, declarations, codes of practice, and action frameworks.⁶¹ However, when a qualitative assessment is made, for example by assessing to what extend the output of ESD has added value for the social protection secured by the EU and national legislation, the conclusions are less positive.

Some scholars argue that added value can be found in the mere fact that social partners have been able to negotiate framework agreements on topics the EU institutions could not, ie parental leave, part-time work and fixed-term work.⁶² At the same time, comments have been made that the level of protection offered by these agreements is not impressive.⁶³ The requirements of the parental leave agreement, for example, have been analysed as being lower than those of the International Labour Organisation’s convention on this topic.⁶⁴ The agreement on information and consultation for civil servants and employees of central government administrations, which triggered the EPSU-case, has also been analysed as having little added value.⁶⁵ On the other hand, the Commission recognised that the autonomous framework agreement on harassment and violence at work has added value to the relevant EU and national legislation.⁶⁶ Based on our own assessment, the same can be said about the agreement on active ageing and an inter-generational approach, which seems to make a valuable addition to the Guidelines of the EPSCO and EMCO,⁶⁷ since it contains an action-oriented approach directed at the employment relationship. Added value can also be found in the agreement on Telework and a couple of sectoral agreements, among others on working time in aviation, and the agreements implementing the ILO Maritime Labour Convention (2006) and the Work in Fishing Convention (2007), all topics on which the Commission—the EU—did not act.⁶⁸

In the literature, there have also claims been made that the ESD outcomes lack added value because they create no obligations and the content is often rather vague and therefore their impact is weak.⁶⁹ However, a study we conducted analysing what we call the ‘legal quality’ of the content of the autonomous agreements, shows that this varies greatly among the agreements. We have analysed the legal quality of the agreements based on an analytical framework that has been used before to analyse the legal quality of (other) EU soft law instruments, including recommendations, resolutions, declarations, action programmes, and the open method of coordination⁷⁰ and that of transnational private initiatives, in particularly CSR codes of conducts and international framework agreements.⁷¹ This analytical framework covers three features of legal documents: legality; substance or negotium; and structure. For this part, we have focussed on the second (substance), which is comprised of two main elements that each are broken down into sub-elements (see Table 3). The analytical technique applied is that of coding the agreements on the sub-elements. The higher the overall score, the clearer its content is in terms of obligations it creates and expected behaviour to be compliant with those obligations. The results of this analysis are presented in Figure 1. ⁷²

A couple of things stand out from Figure 1. First, six out of the thirteen agreements (including the much-debated hairdressers’ agreement) scored maximum on this feature, which means that the legal quality of the content of these agreements is high. Secondly, the three agreements score extremely low, since they formulate no clear obligations, at best they hold political or moral obligations and when it comes to precision, they only score a point on internal coherence. Thirdly, the agreement on information and consultation for central government administration scores somewhat in the middle. This is mainly because it holds mostly hortatory provisions which leave it up to the Member States and social partners at the national level to further work out rules that guarantee a minimum of topics to be subjected to information and consultation procedures. For the delegation of further rulemaking legal language has been used, however, leaving the Member States and social partners a large margin of appreciation, which has been one of the main critics of this agreement in terms of lack of added value.⁷³

Another point that stands out in this figure, although it cannot be directly derived from it, is that it is the agreements that are action-oriented,⁷⁴ rather than setting rights and obligations,⁷⁵ that score low on the legal quality because they use mainly verbs as ‘promote’, ‘endeavour’, ‘can include’, ‘aims’, ‘recognise’, ‘should’ and ‘could’. These are all words that in general are considered by lawyers as legally weak. However, when we read the full texts, these agreements are very clear in the sort of activities that could be developed in the workplace to address the topics of those agreements. As such these agreements hold a rather high practical added value. Therefore, our conclusion is that when it comes to output legitimacy, the added value of agreements lies not always in the sort of obligation it formulates and how precisely it is doing that. Instead, a distinction should be made between agreements that aim to set a (minimum) standard and those that are action-oriented. The legal quality of the former can be fully assessed by the sort of obligation and precision, whereas the latter should be assessed on to what extend the proposed activities have been formulated precisely enough to achieve their aims.

Another often expressed critique in the literature is that many autonomous agreements lack indications on implementation and follow-up monitoring of the implementation which weakens the output legitimacy.⁷⁶ We have checked this also based on our analytical framework, since this is part of the third feature of legal documents—structure. More particularly, it can be assessed by the second and third elements: monitoring compliance and further rule-making (see Table 4 for the sub-elements and coding). Figure 2 shows the results of this analysis.

What stands out in this figure firstly, is that none of the agreements scores the maximum. The highest scores are four out of six, and this is for only three out of the thirteen agreements. Second, eight out of the thirteen agreements have no score at all on ‘further rule-making’. We admit that this especially is somewhat loosely related to the implementation or monitoring of the agreement, nevertheless, it is an indication to what extend the agreement pays attention to compliance with its content in general. Thirdly, all agreements indicate that the implementation will be monitored via a report, however, only five do this on a regular, cyclical basis. Three of these are among the more recently adopted agreements. The 2021 agreement on women in rail sets up an elaborate system of review: regular evaluation of the agreement, including adaptation of the content based on the national experiences and (evolving) problems (Article 7 of the agreement). It is too early to tell whether there is a trend to include regular monitoring and follow-up in the agreements, but the more recent agreements could be an indication of it. More generally, all agreements are concerned with the monitoring of its implementation, but there is room for improvement.

4. THE COMMISSION, INTERESTS, BETTER REGULATION AND REFIT

In this section, we will turn to the role of the Commission in the ESD and how this has changed since the introduction of the ESD. A change, which we think explains why the Commission became more explicit about conducting an appropriateness test before sending a proposal to implement the social partners’ agreement to the Council. Therefore, we will first address the position of the Commission as guardian of the interests of the EU (Article 17 TEU). Second, we will turn to the role of the Commission as an initiator of EU law, including in the field of social policy and how this has changed with the Better Regulation Agenda and REFIT.⁷⁷ We will conclude with some reflections on what this broadly means for the ESD.

A. The Commission as Guardian of the Interests of the EU

Article 17(1) states that the Commission promotes the general interests of the EU and that it takes appropriate initiatives to that end. What is to be understood with ‘general interest’, is not further explained. In her comment on the EPSU-rulings, Velluti concluded that the courts considered ‘general interest of the EU’ too narrowly in the context of Article 17(1) TEU only, instead of considering it in the wider context of Article 13(1) TEU.⁷⁸ The latter includes also the values of respect for human rights and solidarity under Article 2 TEU, the social objectives of Article 3 TEU, fundamental rights as per Article 6 TEU and the ‘Solidarity’ title of the EU’s Charter of Fundamental Rights.⁷⁹ According to Velluti, such a broader reading of ‘general interest of the EU’ could also include the interests that social partners represent in the field of social policy. Such an acceptance could have given the courts the space to rule that the Commission has an obligation under Article 155(2) TFEU to submit a proposal holding the agreement of social partners to the Council.

This is an interesting interpretation of ‘general interests of the EU’, which we warmly support, however, the reality is that the European courts did not see it in this way. This raises the question of why they didn’t. More particularly, we will argue that the European courts did not do this, because of the changed position of the Commission in the governance of the EU. Particularly, since the inception of the ESD with the signing of the Treaty of Maastricht, the nature of European integration has fundamentally changed. The process of European integration has deepened into core state powers, such as Economic and Monetary Union and migration and security, domestic political actors have mobilised to contest EU supranationalism, thereby contributing to processes of politicization.⁸⁰ Procedurally, this change is reflected in the more stringent requirements of subsidiarity and proportionality, offering states another means of control over EU activities, since it basically puts a higher ‘legislative burden of proof’ on the Commission that Union activity is necessary.⁸¹ The call for more autonomy by social partners at the turn of the century has also been evaluated in this context.⁸² Afterall, the route of more autonomous agreements is a way of preventing often more stringent and more widely applicable EU law (negotiating under the shadow of the law).

The fundamentally changed nature of EU integration has important consequences for EU actors in all stages of the policy-making process as EU policy is both challenged and contested from across increasingly divergent interests within the EU’s system of multi-level governance. Increased political contestation within the EU has several consequences, such as the growth of populism and the move towards more technocratic forms of policy-making, which lead to the slowing of the decision-making process.⁸³ Within the European Parliament, for example, the overall trend has been for the two main centre-groups, the European People’s Party on the right and the Socialists and Democrats on the left, to secure lower numbers of seats for their MEPs with support for those of the far-left and far-right of the political spectrum increasing.⁸⁴

In addition, with the progressive transfer of policy-making from the national to the European level, the network of interest groups in Brussels has intensified over time.⁸⁵ Not only does this erode the ability of national governments to set their own policy agendas but also EU-level interest groups are also contributing towards the EU’s agenda-setting and policy-making process. A further important dimension to this transferring of lobbying is the EU’s openness and permeability to interest lobbying compared with even the most pluralistic Member States. The openness of the Commission and the European Parliament to interest groups enables them to become an independent source of European integration.⁸⁶ Although lobbying is recognised as an activity that enhances the legitimacy of EU policies and legislation, since it can be considered as a form of participatory democracy, it raises also challenges. Some lobby groups, like businesses, can also undermine the objectivity that should be maintained,⁸⁷ since their private impact assessments are often more thorough than those of the (understaffed) Commission.⁸⁸ At the same time, it leaves the position of social partners (thus business and workers' organisations together) ambiguous. And even though social partners are consulted as part of the ESD and given a single voice to be considered by the Commission, it is surrounded by a larger group of voices the Commission is listening to which in the end may result in a decision on the proposal that is contrary to the opinion or interests of social partners.⁸⁹

Within this evolving political landscape, the Commission is in something of a unique position, because as guardian of the EU’s interests and sole initiator of proposals for EU legislation (Article 17(1) TEU) it determines its activities. While doing so, the Commission can either play to the politics by siding with a political faction or go with the politics by occupying the centre-ground. The two Barroso Commissions (2004–2009; 2009–2014) are noted for their consistency to play to the politics and siding with more centre-right leaning Governments within the Council and the EPP in the Parliament. Not only was there a slowing of the social legislation and political deadlock over agreements but also Commission proposals for directives were often either unambitious or weakened existing provisions.⁹⁰ This led to claims by Crespy and Menz of the Barroso Commission ‘debasing’ social Europe.⁹¹ An effect experienced by social partners since they noticed a less supportive attitude from the Commission towards the ESD.⁹²

By contrast, the strategies of the Juncker Commission (2014–2019) and Von der Leyen Commission (2019–2024) go with the politics. The Juncker Commission is notable for its ability to be a centre-ground or catch-all Commission in which it aimed to both appease and unite the EU’s divergent interests to deliver a renewed momentum within the process of European integration. The backdrop to this is the negative reputational consequences of the EU-driven austerity stemming from the Eurozone crisis.⁹³ As such, within the social dimension, the Juncker Commission built broad coalitions, both vertically and horizontally. For example, vertically, the Commission built a strong working relationship with the European Parliament via the MEP Maria João Rodrigues, who had been one of the chief architects of the Lisbon Strategy and became an ally to the Juncker Commission in social Europe.⁹⁴

Horizontally, the Commission strategically used the firm consultation process set by the Better Regulation Agenda and impact assessment, to engage with as many actors as possible from outside the EU institutions. In fact, the Juncker Commission used it more extensively and deliberately than any previous Commission.⁹⁵ For example, in its far-reaching public consultation for the European Pillar of Social Rights (EPSR), launched in March 2016, there were some 200 position papers received from institutions, social partners, civil society organisations and citizens, over 16,500 responses to an online questionnaire and over 60 events organized throughout Europe.⁹⁶ The Von der Leyen Commission seems to continue on this path, as for example expressed in the 2021 Porto Social Summit, with the signing of the Porto Social Commitment (PSC).⁹⁷ The PSC is a joint effort of the Portuguese Presidency of the Council of the EU, the Commission, the European Parliament and social partners, ‘to consolidate the commitment already made with the European Pillar of Social Rights and create the synergies for an inclusive, sustainable, fair and job-creating recovery’.⁹⁸ Furthermore, the PSC ends with

‘We stress the importance of engaging and debating how to strengthen the social dimension of Europe with all citizens in the context of the Conference of the Future of Europe and beyond, involving the social partners and mobilizing the European society.’

Which is a further indication of the intention of the Von der Leyen Commission to continue building broad coalitions.

B. Impact of Better Regulation Agenda and REFIT on the Commission

Since the turn of the century, the EU’s policy-making process has been underpinned by the Better Regulation Agenda. The Better Regulation Agenda has its origins in the Lisbon Strategy (2000) which requested that the Commission, the Council and the Member States define ‘by 2001 a strategy for further coordinated action to simplify the regulatory environment, including the performance of public administration, at both national and Community level’.⁹⁹ Following recommendations formulated by the Mandelkern Group¹⁰⁰ the Commission has continuously developed the Better Regulation Agenda.¹⁰¹ The aim of the Better Regulation Agenda is not about regulation or deregulation, instead, it is

‘a way of working that allows political decisions to be prepared in an open and transparent manner, informed by the best available evidence, including via the comprehensive involvement of stakeholders. This is to ensure that the EU acts in line with the overarching principles of subsidiarity and proportionality, i.e. only where necessary and in a way that does not go beyond what is needed to address the problem at hand.

‘Better regulation’ is also a means of guaranteeing that EU legislation has the broad support of EU citizens and remains fit for purpose, future-proof and open to innovative solutions in a context of ever more rapid technological, societal and environmental change.’¹⁰²

One of the underlying reasons for the Better Regulation Agenda can be found in perceptions of the Commission being a self-interested bureaucratic monster,¹⁰³ and, from the perspective of the Commission to free itself of being a ‘scapegoat for poor legislation’.¹⁰⁴ The latter is interesting in the context of the EPSU-case, as well as outcome legitimacy, since Better Regulation means for the Commission, among other things, that it ‘has to have the courage to block unjustified regulatory initiatives from other actors’.¹⁰⁵ In order to enable the Commission to do so, impact assessments are an important tool for Better Regulation. These are not only about costs and administrative burdens, and they generate not ‘clear-cut conclusions or recommendations regarding the final policy choice’, on the contrary, a good impact assessment functions

‘As an important aid to decision-making the results and the alternative options considered – in all cases – need to be presented in a transparent and understandable way to provide the basis for a political discussion on the relative advantages and disadvantages of the relevant options. This allows political decision-makers to examine the trade-offs between affected groups and/or between the impacts on the social, economic and environmental dimensions.’¹⁰⁶

Better Regulation is complemented by the Regulatory Fitness and Performance Programme (REFIT), which began in 2010. The purpose of REFIT is to review existing legislation in selected policy fields through ‘fitness checks’ in order to keep current regulation ‘fit for purpose’.¹⁰⁷ Pilot exercises began in 2010 in four areas: employment and social policy, environment, transport and industrial policy.¹⁰⁸ Initially, REFIT was greeted with scepticism with claims it would pursue a de-regulatory agenda of EU regulation, especially in the field of social policy.¹⁰⁹ In other words, the liberalisation and de-regulation of national economies as part of the Single European Act (1986), which was accompanied by re-regulation at the European level, would succumb EU regulation to a process of de-regulation.¹¹⁰ A consequence of this process would be a truly deregulated European economy. This interpretation of REFIT, which on the one-hand has its roots within Thatcherism, is also tied to the political positioning of the Barroso Commissions and its ability to play with the politics¹¹¹ in the context of the fallout surrounding the Commission’s 2004 proposal for the Services Directive, the re-launching of the Lisbon Strategy in 2005, as well as the EU’s response to the Eurozone crisis.¹¹²

The reality of REFIT has been quite the opposite. In her study about REFIT, Laulom explored the process for the Social Acquis in which the fitness check exercise was launched in three areas: (1) informing and consulting workers on the national level (three directives); (2) the legislative body of Occupational Safety and Health (OSH); (3) and the Written Statement Directive. The analysis conducted by Laulom reveals that the Commission’s approach and evaluations did not lead to deregulation. On the contrary, the reviewed Directives were evaluated as being ‘fit for purpose’, ie as fulfilling their objectives and as being coherent and effective. In some instances, gaps within the existing Directives were identified, which required further review, and fed into the Juncker Commission’s legislative agenda, including the revision of the Written Statement Directive¹¹³ and the Work Life Balance Directive.¹¹⁴ The latter is especially interesting, because it replaces the Parental Leave Directive which implements the agreement of social partners, which was evaluated as having little added value in both versions: 1997 and its 2009 recast.¹¹⁵

More generally, according to Garben, in an EU context, the EU law-making process has become increasingly institutionalised and the EU’s legislative process is increasingly conditioned by it, through the way it structures and informs the Commission’s exercise of its monopoly on legislative initiative, and in its application to the Parliament and Council in their capacity as the European Legislator.¹¹⁶ Better Regulation tools such as regulatory impact assessments, consultations with stakeholders and ex-post evaluations of regulatory interventions, are used by governments to ensure input, throughput and output legitimacy for their policymaking and to address perceived failures of regulation such as extensive administrative burdens on business, insufficient use of scientific evidence in preparing public policies, and biased interest presentation in decision-making.¹¹⁷ Consequently, the nature of the EU’s policy-making process has fundamentally changed with Better Regulation and REFIT, in that EU Directives need to be legitimised at different stages of the legislative cycle from a broad range of interests.¹¹⁸

REFIT has further contributed to a changed regulatory playing field since it requires the Commission to review the fitness of its regulation not by a one-on-one standard, but by a coordinated governance approach. Following developments within the Lisbon Strategy (2000), Europe 2020 (2010), and Von der Leyen’s current strategy programmes set by the European Council, piece meal regulations seem to have become a feature of the past.¹¹⁹ An effect that, in the realm of social policy, seems to have been strengthened with the adoption of the European Pillar of Social Rights in 2017.¹²⁰ As such, the Commission is forced to review its legislative activities in a broader, concerted, context. The Written Statement Directive (considering developments in employment relationships, such as platform work and remote work) and the Work-Life Balance Directive (which extended the parental leave directive to a wider context) are good examples of such coordinated approaches. Other examples are the first consultation on information and consultation rights in the restructuring directives¹²¹ and the proposal for a platform work directive and AI act which are part of a broader set of regulatory activities which are employed under the heading of ‘digital decade’.¹²² More generally, having strategies set by the European Council, also means that the room for manoeuvre for the Commission has increasingly become limited to these political strategic agendas. Consequently, the room of manoeuvre for social partners to have their agreements implemented via a Council decision has also become smaller. Or at least, it makes it for the Commission a lot more difficult to convince the Council to implement an ESD agreement not related to a wider programme.

C. ESD, Commission, Better Regulation and REFIT

As part of building broad (horizontal and vertical) coalitions, the Juncker Commission initiated a number of activities to rebuild its ties and relations with social partners and to revive the ESD.¹²³ A key initiative in this is the Commission’s declaration A New Start for Social Dialogue,¹²⁴ (adopted in 2015 and again in 2016) in which the Commission and Social Partners agreed that there is a need for:

• ‘- a closer involvement of the social partners in the European Semester,

• - stronger emphasis on capacity building of national social partners,

• - increased involvement of social partners in EU policy and law-making,

• - a clearer relation between social partners’ agreements and the Better Regulation Agenda’.¹²⁵

This is repeated in the 2016 quadripartite statement on A New Start for Social Dialogue, in which various intentions have been formulated for different actors involved with the ESD. For example, the cross-industry social partners (or peak organisations)¹²⁶, intent to inform the Commission at the beginning of their negotiations of all framework agreements and frameworks of action; inform the Council at the beginning of their negotiations of framework agreements for which they have agreed to ask implementation by a Council decision; and organise fact-finding seminars at the beginning of their negotiations on autonomous agreements and /or frameworks of action.¹²⁷ The Commission, for example, will endeavour, ‘in line with the Better Regulation Agenda and taking into account the specificities of the social partner agreements’ to support social partners during the negotiations, ‘including, upon request, technical and legal information’.¹²⁸ Member States are recalled to ‘take the necessary steps to […] with full respect for the social partners’ autonomy and upon request by social partners, provide information and advice […] on implications and practicalities for Member States of transposing’ their agreements at Union level.¹²⁹ At the same time, social partners are advised to keep national public authorities informed.¹³⁰ As such, it seems in our view that social partners are, likewise the Commission is doing as part of Better Regulation, expected to keep ‘warm’ (vertical) relationships with the Member States.

The Von der Leyen Commission clearly aims to continue building broad (horizontal and vertical) coalitions. Especially with its recognition of the need for ESD to manage fair transitions. While the Juncker Commission also acknowledged the need to strengthen national level social partners, the Von der Leyen Commission seems to be on a path to take the vertical coalition building to the next level.¹³¹ For example, by organising, on relevant topics, ‘ad-hoc/extraordinary tripartite discussions involving European and national social partners and Member State representatives’.¹³²

When focussing more on Better Regulation, it is interesting that the Commission recognises in its declaration A New Start for Social Dialogue that there is a need for more clarity between ESD agreements and the Better Regulation Agenda. Such is also emphasised by the social partners in the ‘Guiding Principles/Key Messages’ of their joint declaration, which was approved at the Thematic Group meeting on 26/27 January 2016.¹³³ Section II of the ‘Guiding Principles/Key Messages’, deals with the clarification of the relationship between ESD agreements and Better Regulation. It starts with the statement that Better Regulation should genuinely respect the specific nature of ESD agreements.¹³⁴ Unfortunately, it is not specified what is meant by this. Point 20 provides a bit more information though, as it states that:

‘For those agreements for which EU social partners request the implementation by EU law, but have the specificity of following a Commission consultation, the Commission should not revisit the need for EU action, as this was already determined.’

This is at odds with what the Commission put in its declaration A new Start for Social Dialogue, which explicitly includes an appropriateness test before submitting a proposal to implement an ESD agreement to the Council.¹³⁵

In the follow-up report on A New Start for Social Dialogue,¹³⁶ short reports have been provided on the results of the works of two Thematic Groups that were established to substantiate commitments and to monitor the follow-up. Thematic Group 2 dealt with social dialogue, policy-making and law-making, including the special role of social partners in the Better Regulation Agenda.¹³⁷ The Group stressed ‘the need for the social partners to contribute to the building up of the evidence base sustaining their agreement, while keeping the national public authorities informed of the preparation of the agreement’.¹³⁸ In itself these findings are not surprising, since Welz, in his study, already indicated that from a ‘good governance’ perspective, there is much space for improvement for the ESD in general.¹³⁹ Still, the conclusions of Thematic Group 2 are interesting. For example, it indicates that it is not enough for social partners to explain in the preamble why the agreement is important, what it takes into account, and what its aim is. Moreover, to us, it seems that Thematic Group 2 is advising social partners to apply the principles of Better Regulation, such as making their own impact assessment (evidence-based) to support their joint request to have the agreement implemented by a Council decision, when negotiating an agreement.

5. CONCLUSION: WHAT WAYS FORWARD FOR THE ESD AFTER THE EPSU-CASE?

To conclude, we will reflect on our findings and indicate some ways forward for the ESD. Claiming that social partners can operate completely autonomous within the context of EU governance is not possible. The influence of the Commission seems to be omnipresent and although there seems to be an emerging trend that this influence is becoming less strong, at the moment true autonomy seems to exist only as a constitutional idea about collective bargaining as part of the freedom of association and contract.

One might wonder whether within the context of the EU this is problematic. First, the role of social partners and their agreements is a different one than that at national level, which is also visible in the topics of the work programmes of the social partners, which consists largely of activities to strengthen the social dialogue, lobby the Commission, and topics that can be related to EU policy programmes and strategies. The latter is clearly a consequence of the EU’s Better Regulation Agenda and REFIT, which leave less room for piecemeal initiatives by the Commission or social partners. Second, the more social partners get involved with governance mechanisms, besides the law-making process, also the European semester and the implementation of the European Pillar of Social Rights, the more they get institutionalised and as a result their activities will align more with those of the Commission. That does not mean though that the social partners cannot operate autonomously; it only means that it should not be approached in the same way as autonomy that is expected at the national level.

The autonomy of social partners does matter though. It matters at the initial and negotiation phases of the ESD. This is also recognised in the EU treaties, especially Article 154 TFEU which allows social partners to initiate negotiations after the first round of consultations by the Commission, and by the Commission in its 2023 Communication. Autonomy also matters in the phase of consultation for which more space seems to be created due to a change in attitude of the Commission from playing to the politics (Barroso Commissions) to going with the politics (Juncker and Von der Leyen Commissions). The latter includes building broad partnerships, which on the one hand leads to the institutionalisation of social partners as indicated above, but also to more space for social partners to lobby the Commission, which is an important way of influencing EU policies (in put legitimacy). The Von der Leyen Commission seems to take such partnerships as reciprocal: the Commission informs, consults and supports social partners, but expects social partners to inform the Commission, European Parliament and Member States about their activities, including about initiatives for agreements.

Such a reciprocal partnership seems to fit with the changed nature of EU integration which has deepened and includes core aspects of national policies, including in the field of social policies, with examples such as the work-life balance directive, the adequate minimum wage directive, and the platform work directive, which all have a deep impact on labour law aspects of the Member States which before were considered as not possible.¹⁴⁰ Consequently, the principles of subsidiarity and proportionality have become more strict putting more pressure on the Commission to prove that adopting a measure at EU-level is the best way to address the issue. The Better Regulation Agenda and REFIT can be considered as instrumental in this, among others, through the broad consultations which enhance the input legitimacy. It seems only logical, that if more is expected from the Commission by, especially the Member States (represented in the Council), this is also reflected in the proposals the Commission puts forward to the Council to implement an agreement of social partners. Which well explains why the Commission is of the opinion that it needs to perform an appropriateness test before it sends an agreement to the Council and refuses to do so when the outcome of the test is negative.

From the original idea behind the ESD this is odd because social partners are supposed to know what is needed and should best be regulated at EU-level, however, those times were different and holding on to this does injustice to how EU integration and governance have developed. Our conclusion is therefore that if social partners want to regain their (feeling of) autonomy, they should adapt their game to the new reality. In particular, by (further) building and strengthening coalitions that are appropriate for the setting of ESD, especially with their affiliates,¹⁴¹ but also with the Member States and the European Parliament.¹⁴² Secondly, following the suggestion that seems to have been made by Thematic Group 2 (Follow-up report on A New Start for Social Dialogue), social partners could apply the principles of Better Regulation to their agreements, especially those they wish to implement via a Council decision. Since Better Regulation, including Impact Assessments, are strongly embedded in the EU’s ‘DNA’ of law-making, it seems to us unavoidable for social partners to not play by the principles of Better Regulation. Of course, not in the exact same way as they apply to the Commission, but on certain points, such as broad consultations among their affiliates, greater transparency, an impact assessment light to show they have considered the possible administrative burden for SMEs, etc. This is not completely alien for social partners since similar, evidence-based studies, have been part of their lobbying activities. Playing in the spirit of these rules, may possibly reduce the Commission’s need to conduct its own appropriateness test, which would effectively give social partners the autonomy they wish for. Or at least, when the Commission conducts its own test, this might decrease unpleasant surprises, since the test of social partners should capture the most relevant issues. Additionally, the test of social partners, may strengthen the position of the Commission towards the Council (read: member states) and the European Parliament in accepting the proposal holding the agreement of social partners. As such, in ‘playing the game of the EU’, social partners may have a lot to win.

Footnotes

Author notes