Rekindling the ultra-vires debate?
Most lawyers in the field of European Union law seem to have reached an (unspoken) consensus that breaches of the principle of conferral of powers (Art. 5(2) TEU) either do not occur or can not be successfully challenged before the ECJ anyway. Contrary to this, Advocate General (AG) Emiliou’s Opinion from the 14 January 2025 in Case C‑19/23 Denmark v Parliament and Council – surprisingly unmentioned on the Court’s press release site – argues in remarkably clear language that the Directive on Adequate Minimum Wages exceeds the Union’s competences and, in particular, that the ECJ should not apply an overly narrow interpretation of the relevant exclusion of ‘pay’ from the social policy competences of the Union under Article 153(5) TFEU. He has thus proposed to the ECJ that the Directive should be annulled in full.
The purpose of this post is to root AG Emiliou’s Opinion on this specific EU Directive in the broader discussion on a potential ‘competence creep’. In light of this, this post will (1) contextualize, (2) delve into a detailed analysis of AG Emiliou’s Opinion, and (3) comment on its implications within the framework of the Court’s case law.
The context
Directive 2022/2041 on Adequate Minimum Wages (AMW Directive), adopted in October 2022, aims at ‘improving […] in particular the adequacy of minimum wages for workers in order to contribute to upward social convergence and reduce wage inequality’ (Article 1), by tackling the statistical gap in monthly minimum wages across Member States.
However, since its proposal by the Commission in October 2020, the AMW Directive has been under criticism for a potential breach of competences. From the outset, Denmark and Sweden opposed the legal act and subsequently voted against its adoption in the Council. All other Member States with the exception of Hungary, which abstained, were in favour of the Directive. Already during the legislative process, the Danish parliament questioned compliance of the AMW Directive with the principles of subsidiarity and proportionality by means of a reasoned opinion, advocating that wage conditions were best regulated at national level.
Following the adoption of the Directive, in January 2023, Denmark swiftly brought an action for annulment of the entire Directive under Article 263 TFEU before the ECJ, and as an alternative claim asking for annulment of Articles 4(1)(d) and 4(2) of the Directive. Denmark argues that the Directives breaches the principle of conferral of powers and Article 153(5) TFEU. The latter stipulates that the EU may not legislate in the areas of ‘pay’ and ‘the right of association’. Denmark has brought forward that both exclusions embedded in Article 153(5) TFEU are encroached upon by the AMW Directive.
The Opinion of AG Emiliou
The most crucial provision of the AMW Directive for AG Emiliou’s reasoning is Article 5. Article 5(1) of the Directive obliges Member States with statutory minimum wages – and only those – to create a transparent ‘procedure for setting adequate statutory minimum wages’. This procedure shall be guided by criteria in relation to national levels, including at least the purchasing power of statutory minimum wages, the general level of wages, the growth rate of wages and the long-term national productivity levels (Article 5(2)). Statutory minimum wages shall be updated every two years or if the Member State follows the indexation mechanism set out in Article 5(4) at least every four years (Article 5(5)).
According to AG Emiliou, the legislator was knowingly ‘walking on thin ice’, or, in the words of Sacha Garben, a ‘tightrope’ when adopting the AMW Directive (para 50). His central argument is that the Directive as a whole and specifically Article 5 fall under the scope of the exclusion in Article 153(5) TFEU as they indeed set requirements for ‘pay’ in the Member States and therefore ultimately violate Article 5(2) TEU.
Understanding ‘pay’ according to AG Emiliou
It is generally accepted and not denied by the AG, that exclusions have to be interpreted strictly and that only a direct interference constitutes an infringement, Article 153(5) TFEU posing no exception. However, he argues, ‘exclusions […] must not be interpreted so strictly as to be deprived of their effectiveness’ (para 55). Following this principle, AG Emiliou considers it a ‘fallacy’ to limit the scope of ‘pay’ to measures that harmonise wages – read as: the level of wages – a position advocated by all parties to the conflict, except Denmark and Sweden (para 50). In his assessment, the ‘pay’ exclusion covers – but is not limited to – measures that harmonise the level of pay. For his analysis, he uses two main justifications.
First, since Article 153(5) TFEU, like its previous versions, uses the broad term ‘pay’ in contrast to merely ‘level of pay’, other aspects of the Member States wage-setting systems also fall within the scope of said Article (para 59). He cautions that otherwise the EU legislature could harmonise all aspects of the wage-setting system, provided it stopped short of harmonising wages (para 55).
Secondly, the narrow interpretation of Article 153(5) TFEU established in existing case law is not applicable in the present scenario, and has to be reconsidered in general, as it originated from the evaluation of instruments with objectives that differ significantly from those of the AMW Directive. The most prominent case in which the question of the interpretation of ’pay’ was raised, is Case C‑268/06 Impact. According to AG Kokott’s Opinion in that case, ‘only the level of pay […] is removed from the [EU] legislature’s competence by [Article 153(5) TFEU]’ (para 176). Turning to the Court’s judgment in Impact, AG Emiliou admits that the wording might seem to suggest that the Court was emphasising solely the exclusion of the ‘determination of the level of wages’ and the ‘level of pay’ (para 123 of the judgment), as opposed to ‘pay’ more generally (para 52).
Notwithstanding, the AG argues that the wording of the judgment in Impact (‘covering measures – such as […]’) (para 124 of the judgment) leaves the door open for a broader interpretation in the present case (para 53). Additionally, the decisions of the Court that led to the narrow reading of Article 153(5) TFEU, as outlined in the paragraph above, (Case C-307/05 Del Cerro Alonso; Impact; Joined Cases C-395/08 and C-396/08 Bruno and Others; and Joined Cases C‑501/12 to C‑506/12, C‑540/12 and C‑541/12 Specht and Others) all concerned instruments that, unlike the AMW Directive, had as their objective to regulate a matter other than ‘pay’, such as the equal treatment in employment, the principle of non-discrimination to fixed-term workers or the removal of discrimination against part-time workers.
Thus, the AG proposes that the Court’s historically strict interpretation should be understood for what it is: a safeguard to ensure that Article 153(5) TFEU does not prevent the adoption of a legal act which only indirectly affects pay (para 58),as opposed to limiting its applicability to only the level of wages – ultimately depriving it of its effectiveness.
Does the AMW Directive directly interfere with ‘pay’?
To support his conclusion, the AG must demonstrate that the AMW Directive directly interferes with ‘pay’ as he interprets it. Emiliou’s argument revolves around the theory, that the decisive factor in determining direct interference is whether a provision seeks to regulate an area that falls within the scope of an exclusion (para 62). Consequently, the fact that the Directive’s requirements are loosely worded and could be seen as only a minor interference with ‘pay’ is irrelevant—no direct interference is permitted at all (paras 62-64).
AG Emiliou convincingly argues that, while some EU directives have indirectly affected pay – such as the Working Time Directive (2003/88/EC) in establishing entitlements to paid annual leave (para 61) – no prior EU legislation has sought to impose procedural requirements for setting minimum wages. He illustrates how the instruments introduced in Article 5 of the AMW Directive are specifically designed to regulate pay and therefore infringe Article 153(5) TFEU (paras 78-87). Particular attention should be given to the criteria outlined in Article 5(2), as these criteria will play a decisive role in determining the level of wages, potentially outweighing priorities of the national legislator.
Moreover, the AG rejects the notion that the AMW Directive merely sets procedural requirements, arguing instead that it functions as ‘a substantive obligation in disguise’ (para 84). For instance, linking the statutory minimum wage to the national growth rate inevitably guides wage levels (para 84). Additionally, under Article 5(3), the Commission could initiate infringement proceedings if a Member State were to lower its minimum wage after introducing an indexation mechanism (para 85). However, this is ultimately not even relevant: as established earlier, mere procedural obligations can already constitute an infringement of Article 153(5) TFEU (para 86).
Following a well-reasoned analysis, AG Emiliou concludes that the AMW Directive encroaches on the Member States’ exclusive competence to regulate ‘pay’, as the instruments it introduces have as their objective to regulate pay through wage-setting mechanisms (para 87).
And what about the right of association?
The question of a potential breach of the right of association is commented on relatively briefly by the AG (paras 97-111). Articles 4(1)(d) and 4(2) require Member States to take measures to protect trade unions and employers’ organisations from each other’s interference and to provide a framework for collective bargaining if collective bargaining coverage in the Member State is less than 80%.
Contrary to Denmark’s position, AG Emiliou considers these provisions lawful. He recalls that the exclusion in Article 153(5) TFEU must be interpreted narrowly and thus only covers the right to form associations and not the right to collective bargaining (paras 102-103). This is systematically reflected by the fact that Article 156 TFEU distinguishes between these two distinct rights. Additionally, if 153(5) TFEU were to be understood as covering all areas of collective bargaining, the competence laid down in Article 153(1)(f) in the field of ‘representation and collective defence of the interests of workers and employers’ would be rendered meaningless (para 105). Again, applying the direct interference test, Emiliou concludes that Articles 4(1)(d) and 4(2) do not have as their object the regulation of the right of association, since they impose no conditions on the establishment of or membership of an association (para 108).
A special treat for enthusiasts of competence doctrine
With its second plea in law (the first being the infringement of Article 153(5) TFEU), the Danish Government reinvigorates a fundamental question discussed in EU law on competences: what is the correct legal basis for legislature falling within the scope of multiple competence norms? The choice of legal basis is important, because Article 153(1)(f) TFEU requires unanimity in the Council, whereas Article 153(1)(b) TFEU is subject to the ordinary legislative procedure.
Denmark claims that the AMW Directive could not be validly adopted merely on the basis of Article 153(1)(b) TFEU, considering it gives equal importance to both the regulation of ‘working conditions’ (Article 153(1)(b) TFEU) and the ‘representation and collective defence of workers’ interests’ covered by Article 153(1)(f) TFEU.
The AG sides here with the EU legislator and rejects Denmark’s claim (paras 115-120). By reviewing the relevant case law (such as Case C-244/17 Commission v Council and Case C-178/03 Commission v Parliament and Council), he assesses that an act which pursues several objectives must be based on a single legal basis if one of those objectives can be identified as predominant (para 115). In the present case, even if it must be conceded that the obligation, set out in Article 4 of the AMW Directive, to ‘progressively increase the rate of collective bargaining coverage’, by adopting and implementing an ‘action plan’ that has to be notified to the Commission, is certainly not entirely peripheral, it can be assumed that the central objective of the Directive is the establishment of a framework for the adequacy of minimum wages (para 119) and thus predominantly regulating ‘working conditions’.
Contrary to this, I believe that despite its name, the AMW Directive puts a lot of emphasis on improving collective bargaining (the word appears 60 times in the text of the Directive). Furthermore, it can be argued that Article 4, if a Member State does not have a standing tradition of promoting collective bargaining, is just as relevant as Article 5,in terms of the potential impact on national labour law. Nevertheless, a lenient approach by the ECJ is to be expected. This is because determining whether there is a predominant objective of a legal act is often ambiguous, leaving judicial leeway that is unlikely to be used to the detriment of a Union measure.
Last stance against the competence creep in Luxembourg?
As Paul Craig urges, one should be wary of a ‘low-intensity review’ when analysing the multi-faceted relationship between the ECJ and the distribution of powers between Member States and the EU, and that ultimately, it is also a matter of mere ‘perception and feel’, whether an EU act exceeds its ascribed limits. Despite this, or perhaps precisely because of this, the issue of a perceived competence creep, i.e. the gradual and mostly unnoticed usurpation of more and more competences of the Member States by the European Union through legislative and judicial (in-)action has fuelled a long-running debate (see also here and here).
The ECJ can influence the way Union competences are expanded in two ways: the first, which was not relevant in the present case but is of importance to the broader discussion, is through a broad application of free movement, state aid and non-discrimination rules, or even general principles of (EU) law, declaring national laws inadmissible and thereby expanding the area in which existing EU norms play a decisive role.
The second way, pertinent to the case brought by Denmark, is the (overly) Union-friendly interpretation of competence norms and the compliance with such by the European legislator. A sober analysis shows that the ECJ had little opportunity in its case law to deal with infringements of competences by Union bodies, as this objection is rarely raised. Conflicts over competence are generally settled on the political stage through the involvement of the Member States in the lawmaking process. However, there are (admittedly, dating back some time) as far as I can see three examples in which the ECJ, during the 70 years of its activity, has declared European legal acts to be ultra vires: Case 294/83 Les Verts , Joined cases 281, 283, 284, 285 and 287/85 Germany and Others v Commission, and Case C-376/98 Germany v Parliament and Council.
In light of this, the present case provides a welcome opportunity for the Court to clarify the doctrine (or rather: develop it) to be employed when interpreting competence exclusions. The pending judgment may also help to further refine the test to be applied when assessing the correct legal basis for acts with multiple objectives; although it seems probable that the judges will endorse Emiliou’s reasoning on this point. Nevertheless, the Court should seize this opportunity to signal that competence boundaries, even with the best of intentions regarding the material effects of a legislative act in mind, should be taken seriously by the EU legislature and that the flexibility of the Treaties has its limits. However, it seems unlikely that the Court will annul the Directive, as more restrictive interpretations of the ‘pay’ exclusion or the test of direct interference are within the bounds of what is reasonable in the light of existing case law. Another argument against the Court intervening, although not strictly legal, may be the resounding support of Member States other than Denmark and Sweden for the AMW Directive, recognising that the judicial evaluation of competences and their limits is, after all, also a matter of politics.
Lars Allien is a PhD candidate and research assistant at the Chair for Public Law (Prof. Ellerbrok) at Freie Universität Berlin.
