The Pay Exclusion, Procedural Competence, and the Limits of Social Constitutionalism After the Ruling of the Court of Justice on the Adequate Minimum Wages Directive

Table of Contents: 1. Introduction: the constitutional fault line. – 2. The Directive on adequate minimum wages: ambition, architecture, and constitutional risk. – 2.1. Political origins and normative stakes. – 2.2. The architecture of constrained ambition. – 3. The annulment action: constitutional arguments and what they disclose. – 3.1. The Nordic model as constitutional protagonist. – 3.2. The legal architecture of Denmark’s case. – 3.3. Constitutional stakes beyond the litigation. – 4. The judgment: doctrinal architecture, critical assessment, and the question of fundamental rights. – 4.1. The ‘direct interference’ test and its antecedents. – 4.2. The annulment of Article 5(2): criteria for adequacy. – 4.3. The annulment of Article 5(3): non-regression and the logic of the acquis. – 4.4. The silence on fundamental rights: a methodological critique. – 4.5. Partial annulment and the severability question. – 5. Differentiated impact, structural limits, and the double standard of EU constitutional review. – 5.1. Countries with statutory minimum wages: adequacy without criteria. – 5.2. The Nordic systems: autonomy preserved, but at what price? – 5.3. The constitutional asymmetry: austerity, conditionality, and selective review. – 6. Conclusions: social constitutionalism at a crossroads.

Abstract: The Grand Chamber judgment of 11 November 2025 in Case C-19/23, in which Denmark successfully challenged selected provisions of Directive (EU) 2022/2041 on adequate minimum wages, constitutes one of the most constitutionally significant rulings in recent EU social law. This article examines the judgment from the perspective of EU constitutional law rather than labour law, focusing on four interconnected questions: the interpretation of the Article 153(5) TFEU pay exclusion; the annulment of the adequacy criteria of Article 5(2) and the non-regression clause of Article 5(3); the Court’s disengagement from the fundamental rights dimension, including Article 31 of the EU Charter; and the constitutional asymmetry between the Court’s deferential treatment of Troika austerity conditionality and its strict review of upward social harmonisation. The article argues that while the judgment preserves a meaningful space for EU social legislation, its minimalist approach – rejecting a rights-conforming interpretation and stripping ‘adequacy’ of any common methodological content – represents a missed constitutional opportunity. It concludes by identifying the interpretive resources – Article 3(3) TEU, the Charter, ILO Convention No. 131, and the European Social Charter – that could support a more ambitious social constitutionalism in future litigation.

Keywords: adequate minimum wages – Article 153(5) TFEU – EU social competence – pay exclusion – fundamental rights – social constitutionalism.

1.   Introduction: the constitutional fault line

On 11 November 2025, the Grand Chamber of the Court of Justice of the European Union delivered its highly awaited judgment in Case C-19/23.[1] The ruling resolved what had become one of the most politically charged pieces of litigation in the recent history of European social law: an annulment action brought by the Kingdom of Denmark, with the Kingdom of Sweden intervening in support, against the Directive on Adequate Minimum Wage in the European Union.[2] The Directive was adopted on the basis of Article 153(1)(b) TFEU, which grants the EU competence to set minimum requirements on ‘working conditions’, yet Article 153(5) TFEU explicitly excludes ‘pay’ from the EU’s social policy powers: this choice, according to some critical views, made the Directive constitutionally vulnerable from the outset.[3]

The case required the Court to navigate the structural tension encoded in Article 153 TFEU between the Union’s aspiration to promote fair working conditions and the categorical exclusion of ‘pay’ from its harmonisation competence. That tension reflects a constitutional bargain struck at Maastricht and confirmed at Amsterdam between labour-market traditions so different in kind that no single normative framework could embrace them all without remainder.

The procedural history of the case amplified its significance. Following the Advocate General’s Opinion of 14 January 2025,[4]which recommended the full annulment as claimed by the Kingdom of Denmark, speculation about the Directive’s fate intensified across the legal scholarship, the European institutions, and civil society. The labour law community mobilised rapidly and critically. When the Grand Chamber spoke, it produced what many commentators greeted as a judicious constitutional settlement, preserving the Directive’s core while excising the provisions that crossed the Treaty’s outer limits.[5] That characterisation is not wrong, but it is incomplete. While the CJEU ultimately preserved the Directive’s substantive core, the analytical framework it employed to distinguish permissible from impermissible interference with ‘pay’ under Article 153(5) TFEU raises significant concerns, as it seems that the only way for the Directive to survive was not to impose any obligation whatsoever on Member States. 

Read from the perspective of EU constitutional law rather than labour law, the judgment is better understood as a revealing episode in a longer struggle over the constitutional status of social rights within the Union legal order – a struggle that, in C-19/23, the advocates of a substantively robust Social Europe do not seem fully won. This article engages with that struggle. It approaches the ruling not only as a labour law matter – though its labour law implications are substantial – but mainly as a constitutional event bearing on the structure of Union competences, the interpretive role of fundamental rights, the legitimacy of competing regulatory traditions, and the contested relationship between economic governance and social protection in the EU order.[6]

The argument proceeds as follows. Section 2 reconstructs the Directive’s normative ambition and its architecture, showing how it sought to operate at the frontier between procedural facilitation and substantive influence on wage outcomes. Section 3 examines Denmark’s constitutional challenge and the deeper tensions it crystallised. Section 4 analyses the Court’s reasoning in detail, focusing on the ‘direct interference’ test, the annulments of Article 5(2) and (3), and the judgment’s studied disengagement from the fundamental rights dimension. Section 5 surveys the differentiated impact of the ruling across Member States. Section 6 situates the judgment within the longer arc of EU social constitutionalism and draws conclusions.

2.   The Directive on adequate minimum wages: ambition, architecture, and constitutional risk

2.1.  Political origins and normative stakes

Directive (EU) 2022/2041[7] draws its political mandate from Principle 6 of the European Pillar of Social Rights, proclaimed jointly by the Parliament, the Council, and the Commission in Gothenburg in November 2017.[8] That Principle, i.e. the right to fair wages sufficient to ensure a decent standard of living, had been an aspirational declaration, which the Directive aims to translate into justiciable obligation. Its recitals catalogued a Europe of stagnating low wages, in-work poverty touching disproportionately on women, migrants, and young workers,[9] pandemic-induced precarity in ‘essential’ but undervalued sectors,[10] and fragmented minimum wage enforcement.[11] The overall picture is of a single market whose integration logic had, in significant respects, operated to the detriment of the wage floors that any concept of social justice requires.

That diagnosis was inseparable from a broader political economy critique. The decade following the 2008 financial crisis had been characterised, across much of the EU, by wage compression and collective bargaining retrenchment – sometimes spontaneous, sometimes deliberately induced through austerity conditionality attached to financial assistance.[12] The Directive represents a considered institutional response to that experience: an acknowledgement that the Union’s credibility as a social project requires not merely the proclamation of rights but instruments capable of ensuring their material content. Its adoption was therefore a signal of political reorientation as much as a specific regulatory intervention, and this context matters for how its constitutional architecture should be read.

The Directive also bears constitutional significance as an attempt to operationalise the social dimension of the Treaties in a domain, wages, that had long been treated as beyond the Union’s reach. The horizontal social clause of Article 9 TFEU requires the Union to take into account requirements linked to the promotion of employment, the guarantee of adequate social protection, and the fight against social exclusion when defining and implementing its policies. The objectives of Article 151 TFEU – promotion of employment, improvement of living and working conditions, proper social protection, dialogue between management and labour – are framed as common objectives of the Union and the Member States. These provisions have long struggled to generate legally enforceable obligations of a genuinely progressive character; the social rights they invoke have tended to function more as teleological interpretive aids than as autonomous sources of competence. The Directive on Adequate Minimum Wages is, in this sense, a test of whether the Treaty’s social provisions could be mobilised to produce concrete improvements in the material conditions of low-wage workers – and the outcome of Case C-19/23 constitutes an answer to that test that is, at best, equivocal.

2.2.  The architecture of constrained ambition

The final text of the Directive is a document of elaborate self-constraint and great political compromise within the so-called Trilogy before final approval. Articles 1(3) and 1(4) state explicitly that the Directive creates no obligation to introduce a statutory minimum wage, no obligation to make collective agreements universally applicable, and no obligation to make a statutory minimum wage the exclusive mechanism of wage protection.[13] These clauses were the political price of adoption. 

Within those self-imposed limits, the Directive deployed a distinctive regulatory technique. Rather than setting wage levels – which obviously Article 153(5) TFEU placed beyond the Union’s reach – it structured the processes by which States must determine whether their minimum wages are to be considered adequate. Article 5 is, indeed, the centrepiece: it requires Member States with statutory minimum wages to establish clear and stable criteria for adequacy assessment, to update those wages at regular intervals, and, in the version before the Court, to apply a specified set of economic criteria to those assessments.[14] The 60%-of-median and 50%-of-average benchmarks, though formally indicative rather than binding, served as common reference points for a shared evaluative language across Member States.

This design belongs to the family of ‘new governance’ instruments that have characterised much of EU social policy since the Open Method of Coordination: coordination through process, benchmarking, and transparency rather than hard harmonisation of outcomes.[15] It instantiates what the literature on promotional labour law describes as a regulatory strategy of ‘positive sanctions’ – creating incentive structures that make substantive progress the path of least resistance, without mandating any specific result.[16] The elegance of the approach lay in its claim to escape the binary of harmonisation versus non-action: the Union would not fix wages, but it would discipline the reasoning by which wages are set – requiring that process to be transparent, periodic, evidence-based, and informed by a common set of economic benchmarks.

The Directive also contains a notable innovation in its approach to monitoring and enforcement. Articles 9[17] through 11[18]establishes a framework of monitoring, reporting, and data collection through which the Commission would track Member State compliance with the procedural and promotional obligations. Article 10[19] requires Member States to report annually to the Commission on minimum wage protection, coverage, and trends. This transparency architecture serves a dual purpose: it generates the comparative data needed to assess progress toward convergence, and it creates a public accountability mechanism through which the political costs of inadequate domestic action could be raised. That dimension of the Directive has survived the litigation intact, and its potential long-term influence on national wage policy should not be underestimated, even if it operates only through political pressure rather than legal obligation.

The constitutional wager inherent in this design was deliberate. The drafters understood that procedural and methodological obligations represent a grey zone between the lawful facilitation of working conditions and the prohibited regulation of pay. They calculated, on reasonable grounds, that prescribing how States must reason about adequacy is sufficiently distinct from prescribing what level they must reach. The Court’s eventual response to that calculation was, as Section 4 will show, affirmative in principle but restrictive in application.

3.   The annulment action: constitutional arguments and what they disclose

3.1.  The Nordic model as constitutional protagonist

To understand Denmark’s challenge, it is necessary to grasp the institutional architecture it was defending. The Danish as well as the Swedish labour market systems rest on a principle of virtually absolute collective autonomy: the normative authority over employment conditions, including, in particular, wages, belongs to the social partners, not to the state.[20] There is no statutory minimum wage in Denmark; sectoral collective agreements, negotiated between employer associations and trade unions, govern wage floors for some 84% of the workforce.[21] The legitimacy of this system derives not from constitutional mandate – unusually, neither Denmark nor Sweden anchors labour rights in constitutional texts or fundamental rights instruments in the way that most other EU Member States do – but from a long tradition of functional efficacy: the model delivers high wage coverage, strong purchasing power for workers, and social peace, with minimal state intervention.

The structural tension between this model and EU law is well documented and long-standing.[22] It pre-dates the Adequate Minimum Wages Directive: it surfaced in the aftermath of Viking[23] and Laval[24] – rulings in which the Court subordinated the right to collective action to the requirements of the free movement of services – and it runs through the entire history of the Posted Workers Directive and its successive revisions. Each time EU law has asserted a heteronomous standard over conditions previously regulated exclusively by collective agreement, the Nordic systems have pushed back, and with some juridical justification: their model operates on the premise that the social partners, not the legislature, are the legitimate agents of wage governance, and any external normative constraint threatens that premise at its root.

The Adequate Minimum Wages Directive provoked a particularly acute reaction because it was addressed, in formal terms, to all Member States, but its substantive obligations under Article 5 applied only to those with statutory minimum wages. For Denmark and Sweden, the operative provision was Article 4 on collective bargaining promotion, which set a target of 80% coverage and required action plans where coverage fell below it.[25] The threat was not immediate as both countries substantially exceeded that threshold, but it was structural: the Directive created a framework in which the Union would monitor, benchmark, and potentially sanction the adequacy of collective bargaining coverage, asserting a supervisory role over a domain the Nordic model had always treated as autonomous. That prospect was, for the Nordic social partners, intolerable regardless of present compliance.

Denmark’s legal challenge was precisely constructed on the assumption that the whole Directive crossed the boundary drawn by Article 153(5) TFEU between lawful support for ‘working conditions’ and prohibited interference with ‘pay’.[26] The annulment case also questioned the choice of legal basis for the entire Directive. Denmark contended that a measure whose practical effect is to regulate wage adequacy – however indirectly – cannot validly rest on Articles 153(1)(b) and (f) TFEU, because the ‘centre of gravity’ of the act relates to pay rather than to working conditions in the narrow sense.[27]

The Court addressed this argument by applying its standard legal basis methodology, asking what the ‘principal aim and content’ of the Directive are, and rejected it, finding that the measure was primarily directed at improving working conditions by ensuring adequacy of minimum wage protection, with only incidental effects on wage levels as such.

Denmark also raised subsidiarity and proportionality arguments, contending that the degree of Union action authorised by the valid provisions went further than necessary and paid insufficient regard to national regulatory diversity.[28] These arguments were largely dismissed by the Court, which upheld the proportionality of the provisions it validated on the basis that the objective of addressing in-work poverty and convergence within the single market has a cross-border dimension that Member State action alone cannot adequately address.

Finally, Denmark’s fallback position sought partial annulment of specific provisions, namely Article 4(1)(d) and/or Article 4(2) of the Directive, which relate to the promotion of collective bargaining. The Court confirmed that collective bargaining provisions related to the objective of promoting ‘working conditions’ were validly adopted, and that a qualified majority rather than unanimity sufficed for their adoption. This secondary claim was therefore also dismissed.

3.2.  Constitutional stakes beyond the litigation

What the Danish challenge articulated, beneath its technical legal form, was a question of constitutional design with implications extending far beyond wage regulation. How much procedural discipline may the Union impose on a policy domain that the Treaties have formally reserved to Member States? At what point does the structuring of national decision-making processes shade into the de facto determination of outcomes? And who – the Court, the legislature, or the political process – is the appropriate arbiter of that boundary?

These are not merely academic questions. The ‘joint-decision trap’ identified in the political science literature on European integration,[29] whereby national vetoes and unanimity requirements prevent the Union from acting on shared problems, has a judicial analogue: the capacity of a single Member State to challenge, through Treaty-based litigation, a legislative initiative adopted by a large majority of Member States and their democratically elected representatives. The C-19/23 litigation illustrates this dynamic with unusual clarity. The Directive was adopted by a qualified majority of the Council and a broad majority in the Parliament, following years of consultation and negotiation. Its annulment, even partial, illustrated the structural tension between the constitutional constraint of attributed competences and the democratic mandate of the Union’s legislative organs.

The case was also, in an important sense, a constitutional stress test conducted through judicial rather than political channels. Had Denmark wished to entrench the Nordic model’s autonomy through treaty amendment, it could have sought a revision of Article 153(5) TFEU to make its exclusions more precise or more comprehensive. It chose instead to litigate – a strategy that, regardless of outcome, produces a judicial construction of the existing text that constrains future legislative choices without requiring the political mobilisation that treaty change demands. This illustrates how competence litigation in the EU functions as a form of quasi-constitutional politics, with the Court serving as an arena in which the boundaries of the political settlement embedded in the Treaties are renegotiated through juridical means.

4.   The judgment: doctrinal architecture, critical assessment, and the question of fundamental rights

4.1.  The ‘direct interference’ test and its antecedents

The Court’s legal method in C-19/23 is disciplined and internally coherent. Its central analytical tool is the ‘direct interference’ test: Union measures that directly determine the level of minimum wages fall within the Article 153(5) exclusion; measures that bear on the procedures and processes through which wages are determined, without specifying outcomes, fall within the competence conferred by Article 153(1)(b) and (f). This test derives from a line of case law extending back to Del Cerro Alonso and Impact,[30]which had already established that the Article 153(5) exclusion must be interpreted strictly, as a derogation from the general rule, and that it cannot encompass all measures that might in some way bear on pay. The Court at paragraphs 71–72 restates and develops this principle with unusual explicitness, holding that interpreting the exclusion to preclude any Union measure with ‘positive effects on the level of remuneration’ would hollow out the social competences of Articles 153(1)(b) and (f), since remuneration is an ‘integral part of working conditions’.[31]

This is constitutionally important. The rejection of the Advocate General’s maximalist reading of the pay exclusion preserves a meaningful space for EU social legislation even within the TFEU’s constrained competence architecture. It confirms that the Union may enact measures whose purpose is to support Member States in delivering adequate working conditions, even when those conditions inevitably have a wage dimension, provided it stops short of specifying what wage level States must achieve. The validated provisions of the Directive – Article 4’s collective bargaining framework and Article 5(1)’s procedural requirement for regular, structured review[32] – constitute an important building block for future EU social legislation in this space, and the Court’s reasoning in upholding them is doctrinally sound.

It is also worth acknowledging what the Court’s analysis of the legal basis gets right. At paragraphs 56–62, the Court applies the standard ‘centre of gravity’ test for identifying the principal legal basis, asking what the essential aim and content of the Directive are, and concludes that the measure is primarily concerned with improving working conditions by promoting the adequacy of minimum wages, not with directly regulating their level.[33] This is a defensible reading of the Directive’s structure and purpose. The problem lies not in the identification of the competence base but in how strictly the Court then applies the Article 153(5) exclusion within that base.

4.2.  The annulment of Article 5(2): criteria for adequacy

The annulment of Article 5(2), however, warrants searching scrutiny. The contested provision required Member States to take account of at least four economic indicators when assessing the adequacy of their statutory minimum wages: purchasing power adjusted for cost of living; the general level and distribution of wages; wage growth rates; and long-term productivity trends.[34] The Court held at paragraph 132 that prescribing these criteria, even as minimum requirements rather than determinative formulae, constitutes a ‘direct interference’ with pay, because the criteria are ‘directly related to the determination of the level of minimum wages’ and constrain the substance of national decisions in a way that goes beyond procedural facilitation.

This reasoning is vulnerable to a principled objection that the Court does not adequately address.[35] The criteria of Article 5(2) are not wage levels, rather they are parameters of rationality for an adequacy assessment process. Requiring a government to take purchasing power into account when deciding whether its minimum wage is adequate is not the same as telling it what wage to set. It is specifying the epistemic conditions of a legitimate decision: a decision that ignores the cost of living, or abstracts from actual wage distributions, or disregards productivity trends, is not merely suboptimal, it is constitutionally unreasoned in the sense that EU law invests in the concept of ‘adequacy’. The distinction the Court collapses, between the quomodo of wage determination and its quantum, is constitutionally real and analytically workable. 

Sagan and Schmidt seem to reinforce this point:[36] they demonstrate, indeed, that the mandatory criteria left Member States with an extremely wide margin of discretion: States remained free to decide on the relative weight of each criterion, could choose between gross and net wages, determine the composition of cost-of-living baskets, select the relevant time horizons for productivity measurement, and factor in local or regional conditions. The criteria did not prescribe a particular wage level or even a methodology for arriving at one; they prescribed the informational inputs of a rational deliberative process. That each criterion – purchasing power, wage distribution, wage growth, long-term productivity – admits of radically different operationalisations by different Member States only confirms that Article 5(2) was structurally incapable of directly determining the level of pay. The provision constrained neither the outcome nor the institutional form of wage-setting; it required only that certain macroeconomic realities not be ignored entirely when governments exercise their sovereign prerogative to set minimum wages.

Consider the analogy with pay transparency legislation. The Directive on Pay Transparency,[37] not challenged on the same grounds, requires employers to disclose pay information, prohibits pay secrecy clauses, and mandates pay assessments where gender pay gaps exceed 5%. These requirements directly affect the process by which wages are set and revised within employing enterprises, arguably a more intrusive intervention in wage determination at the firm level than prescribing macro-economic criteria for assessing minimum wage adequacy at the national level. The contrast reveals that the Court’s reasoning in C-19/23, if applied consistently, would produce anomalous results across the existing EU social acquis. Anti-discrimination provisions of the Equal Treatment Directive[38] and the Part-Time Work Directive[39] that require employers to review and adjust pay arrangements could, by the same logic, be characterised as direct interference with remuneration. The Court does not engage with these analogies, and the absence of engagement is telling.

The practical consequence of this annulment is constitutionally troubling. The concept of ‘adequacy’, the cornerstone of the Directive’s normative architecture, is now formally unanchored at Union level.[40] Member States retain the word but lose the shared framework for its interpretation. ‘Adequate’ in one Member State may mean 40% of median wages and 65% in another. The Union has recognised the problem of in-work poverty as a shared concern warranting common action, nonetheless it has been deprived of the instrument needed to ensure that the common standard means anything common. This is a structural limitation on the Union’s capacity to function as a social community rather than a mere economic arrangement.

4.3.  The annulment of Article 5(3): non-regression and the logic of the acquis

The annulment of the non-regression clause embedded in Article 5(3) is, from the standpoint of EU social constitutionalism, the most constitutionally alarming feature of the judgment. The provision at issue prohibited Member States that used automatic indexation mechanisms from implementing those mechanisms in a way that produced a nominal reduction in the minimum wage.[41] The Court held at paragraph 151 that this constitutes a direct interference with pay determination because it prevents States from using permitted indexation techniques in a manner that would reduce nominal wage levels.

The non-regression principle is not merely a technical drafting device. It is a structural pillar of EU social law, consistently upheld by the Court across the fields of fixed-term work (Adeneler), collective redundancies, and working time.[42] Its function is to prevent the Union from being deployed as a ratchet operating in the wrong direction: to ensure that engagement with EU law does not provide a pretext for dismantling existing national protections. By annulling the non-regression clause in Article 5(3), the Court implicitly signals that Member States retain a near-unrestricted discretion to reduce nominal minimum wage levels, at least where they do so through automatic mechanisms, without any EU-level constraint. In a period of high inflation, when the purchasing power of minimum wages requires active defence against automatic nominal erosion, this signal is not constitutionally neutral. It suggests that the constitutional architecture of the Union permits more interference in wages flowing downward than flowing upward: a structural asymmetry whose implications for EU social constitutionalism should not be minimised.

The Court’s reasoning on Article 5(3) is also internally inconsistent in a way the judgment does not acknowledge. The Directive permitted explicitly the use of automatic indexation mechanisms. It merely required that where such mechanisms are used, they should not produce nominal wage reductions. If the permission to use indexation mechanisms does not constitute interference with pay – and the Court accepts that it does not – it is difficult to understand why a minimal constraint on how those mechanisms may operate constitutes a direct interference of the kind prohibited by Article 153(5). The Court asserts the conclusion without engaging with the structural logic that would be needed to justify it. That cannot be a constitutionally coherent account of how Article 153(5) operates.

4.4.  The silence on fundamental rights: a methodological critique

The most distinctive and consequential feature of the judgment, one that sets it apart from what a more constitutionally ambitious ruling might have looked lie, is its systematic disengagement from the fundamental rights dimension of the case. The Court analyses C-19/23 as a competence dispute, applies the direct interference test, and reaches its conclusions without once invoking Article 31 of the EU Charter of Fundamental Rights.[43] Article 31(1) guarantees every worker the right to ‘working conditions which respect his or her health, safety and dignity’. Read in the light of the Charter’s general provisions, and of the Court’s own jurisprudence establishing the Charter as binding primary law since the Treaty of Lisbon, there was a strong interpretive basis for reading Article 153(5) TFEU in a manner consistent with, rather than destructive of, this guarantee.

A rights-conforming approach would have proceeded as follows. Article 153(5) excludes ‘pay’ from harmonisation because the Treaty reserves wage levels to national determination. But this exclusion cannot be interpreted so broadly as to deprive Article 31(1) of operational content. The right to working conditions that respect dignity implies at minimum that the Union can establish a common framework for assessing whether wage levels are consistent with dignity, that is, whether they are ‘adequate’ in the legally meaningful sense. The criteria of Article 5(2) are clearly not wage levels; but the epistemological conditions for a dignity-conforming assessment of adequacy. To annul them is to remove the only instrument through which the right of Article 31(1) can be given any common content at Union level.[44]

This kind of rights-conforming interpretive approach is not novel in the Court’s own jurisprudence. In the working time field, the Court has consistently interpreted the scope of its dedicated Directive[45] in the light of the right to annual paid leave as a ‘particularly important principle of EU social law’. In CCOO v Deutsche Bank[46] the Court held that the right to limitation of maximum working hours in Article 31(2) of the Charter required Member States to establish systems for objectively measuring each worker’s daily working time, a procedural obligation grounded in the fundamental rights dimension of the working time rules. The parallel with the Adequate Minimum Wages Directive is instructive: just as the right to working time limitation requires objectively defined measurement systems to give it practical effect, the right to fair working conditions, including adequate remuneration, requires some common definitional framework for what ‘adequate’ means to give it any content beyond national discretion. The Court’s failure to draw this parallel signals a reluctance to extend the rights-informed interpretive approach to the politically more sensitive domain of wage governance.

International instruments that the Court regularly invokes as interpretive resources, Article 4 of the revised European Social Charter on the right to fair remuneration, ILO Convention No. 131 on minimum wage fixing,[47] specify precisely the kind of criteria annulled by the Court as constitutive of the concept of a fair and adequate wage. By declining to mobilise these sources, the Court forgoes the opportunity to give substantive content to the Charter’s social rights, treating them as aspirational declarations rather than as interpretive principles capable of shaping the scope of Treaty competences. The result is a jurisprudence that confirms the structure of social rights in the Union legal order as permanently subordinate: to the market in the post-Viking era,[48] to attributed competences in the post-C-19/23 era.

Paragraphs 71–72 of the judgment are, in this respect, revelatory in their incompleteness. The Court acknowledges that ‘working conditions’ and ‘pay’ overlap, and that remuneration is integral to working conditions. It thereby creates the doctrinal foundations for a more robust recognition of the Union’s social competence. But it immediately proceeds to apply the direct interference test in a manner that forecloses exactly the use of that competence which would give it practical content. A court more attentive to the purposive dimension of the Treaty’s social objectives, and more willing to treat Article 3(3) TEU’s commitment to a ‘social market economy’[49] as an operative constitutional principle rather than a rhetorical flourish, would have drawn different conclusions from the same normative premises.

4.5.  Partial annulment and the severability question

The Court’s decision to partially annul the Directive, rather than uphold or strike it down in its entirety, required a judgment on severability: whether the annulled provisions were separable from the rest of the act without distorting its essential substance.[50]The Court held, at paragraphs 158–172, that the Directive could function without Article 5(2) and the contested part of Article 5(3), since the remaining provisions – the procedural framework, the collective bargaining promotion obligations, the monitoring and reporting requirements – form a coherent regulatory scheme capable of independent operation. This conclusion is formally defensible. But it raises a deeper question: if the criteria for adequacy assessment were central to the Directive’s normative architecture, and they were, as the legislative history makes abundantly clear, then what remains is not just the Directive without its most intrusive provisions, but the Directive without its substantive spine. The form is preserved; the transformative ambition is diminished.

5.   Differentiated impact, structural limits, and the double standard of EU constitutional review

5.1.  Countries with statutory minimum wages: adequacy without criteria

For Member States with statutory minimum wages, which represent the majority of EU countries, the practical effect of the judgment concentrates on the loss of the methodological framework that Article 5(2) would have provided. For countries with historically low minimum wages, such as Bulgaria, Romania, Latvia, Lithuania, the Directive’s surviving provisions impose a procedural discipline that may in time have meaningful effects. The requirement of regular and structured review, consultation of social partners, and annual reporting to the Commission creates a framework of public accountability that did not previously exist at EU level. Even without binding adequacy criteria, the political pressure generated by this transparency, the regular visibility of disparities between national minimum wage levels and the indicative reference benchmarks, may prove a durable instrument of convergence. The analytical challenge is to assess whether this form of ‘governance by embarrassment’ can substitute for the legal traction that Article 5(2) would have provided.

Italy provides a revealing case study. Italy is unusual among major EU economies in lacking a statutory minimum wage altogether.[51] Its labour market is regulated by a dense and increasingly fragmented landscape of sectoral collective agreements, many negotiated by unrepresentative organisations, so-called contratti pirata, and establishing wage floors widely considered inadequate.[52] Italian courts had begun to bridge this gap through constitutional interpretation, invoking Article 36 of the Italian Constitution alongside the Directive’s reference benchmarks as indicators of constitutionally adequate remuneration.[53] After C-19/23, those benchmarks survive in the Directive only as indicative values – politically useful but legally stripped of their Union-level anchoring. The Directive, in its surviving form, creates obligations of process without constraining outcomes: it requires Italy to have a procedure for assessing whether its collectively agreed wages are adequate, without providing the common definition of adequacy that would give that procedure any upward-convergence traction.

5.2.  The Nordic systems: autonomy preserved, but at what price?

For Denmark – and partly for Sweden – the judgment can be received as a success: the claim was not fully rejected, also the autonomy of social partners over wage determination is formally protected, and the threat of prescribed adequacy criteria does not materialise. But this vindication of collective autonomy is narrower, and more precarious, than it might appear. The judgment protects the Nordic model against this particular legislative initiative, but it does not resolve the structural tensions that generated the conflict in the first place. Emerging low-wage sectors within Nordic labour markets,[54] such as cleaning, home care, domestic services, digital platform work, remain inadequately covered by collective agreements, and the high aggregate coverage figures mask significant sectoral variations. The constitutional wall the Court has built around the pay exclusion will prevent not only this Directive but any future EU instrument from providing a common framework for addressing these weaknesses, even where they arise from cross-border competitive dynamics that no single Member State can address on its own. In protecting autonomy, the ruling also entrenches isolation.

There is also an internal trade union dimension. The Nordic confederations’ opposition to the Directive created significant strain within the European Trade Union Confederation, where affiliates from Southern and Eastern European countries saw the Nordic position as a form of structural solidarity deficit.[55] The post-judgment rebuilding of that solidarity, and the development of a common European trade union strategy for minimum wage governance within the constraints now established by the Court, will require political investment that the legal conflict has made more, not less, difficult to mobilise.

5.3.  The constitutional asymmetry: austerity, conditionality, and selective review

Any honest assessment of C-19/23 must confront the elephant in the room: the Court’s historical treatment of austerity measures that directly and coercively interfered with national wage sovereignty. The Troika conditionality programmes of 2010–2015 required States receiving financial assistance to reduce statutory minimum wages, suspend indexation mechanisms, and structurally dismantle collective bargaining systems.[56] These were binding conditions for the disbursement of loans. The interference with national wage autonomy was far more direct, far more coercive, and far more socially damaging than anything contemplated by the Adequate Minimum Wages Directive.

The Court’s response to that interference was systematically deferential. It either held that the memoranda, though negotiated and supervised by EU institutions, fell outside the scope of EU law as instruments of international public law, thus escaping review for compatibility with Article 153(5) and the social rights of the Charter, or it upheld the contested measures as serving a legitimate general interest objective.[57] The constitutional shield of the pay exclusion, which today prevents the Union from prescribing that States must take purchasing power into account when setting their minimum wages, was simply not deployed when the same Union’s institutions were requiring those wages to be cut.

This asymmetry is constitutionally indefensible, and it has not been adequately theorised.[58] It produces a normative hierarchy in which the economic and financial architecture of the Union is effectively insulated from the social rights constraints that the Treaties nominally impose, while the Union’s capacity to act constructively in the social sphere is hemmed in by a formalism that treats methodological criteria for adequacy assessment as equivalent to directly setting wage levels. The Directive, considered procedural, facilitative, non-prescriptive as to outcomes, was subjected to a searching constitutional examination that the Troika’s substantive wage-cutting programmes never faced. One does not need to be a committed Eurosceptic to find this pattern constitutionally troubling.

6.   Conclusions: social constitutionalism at a crossroads

Judgment C-19/23 is a constitutionally competent ruling. The Court navigates a genuinely difficult Treaty landscape with formal precision, produces a workable demarcation between permitted procedural facilitation and prohibited substantive interference, and avoids the institutional crisis that a wholesale annulment of the Directive would have caused. Its contribution to the interpretation of Article 153 TFEU, particularly its rejection of a maximalist reading of the pay exclusion and its affirmation at paragraphs 71–72 that remuneration is an integral part of working conditions, will prove useful for future EU social legislation seeking to occupy the available constitutional space.

The judgment’s contribution also extends to the legal basis methodology. At paragraphs 56–62, the Court confirms that Articles 153(1)(b) and (f) TFEU together constitute a meaningful competence base for measures promoting adequate minimum wages. This confirmation forecloses the argument, advanced by the Advocate General in more expansive terms, that the entirety of minimum wage legislation falls outside the Union’s reach. The Court draws a usable map of the available constitutional territory, even if it then draws its boundaries narrower than a rights-sensitive approach would have required. That map will guide both the Commission’s next legislative steps in the social field and the interpretive choices of national courts applying the surviving provisions of the Directive.

But a constitutionally competent ruling is not necessarily a constitutionally ambitious one. And it is ambition, of a disciplined, rights-grounded kind, that the moment demanded.[59] When fundamental rights are at stake, rulings like the one in case, deeply inspired by a minimalistic approach,[60] risk looking conservative, or at the very least producing conservative effects.

A Union navigating geopolitical fracture, democratic erosion, and widening inequality cannot afford a constitutional jurisprudence that treats fundamental social rights as aspirational declarations, that confines the concept of ‘adequacy’ to an empty signifier whose content States may determine arbitrarily, and that annuls the only provision protecting workers against nominal wage reduction through automatic mechanisms, while remaining silent on a decade of coercive interference with national wage sovereignty under Troika conditionality.

The limitations of the judgment also carry implications for the Union’s legislative strategy going forward. If the Union cannot define common criteria for assessing minimum wage adequacy, because that constitutes ‘direct interference’ with pay, the Directive’s monitoring and reporting architecture will produce data without producing standards, annual transparency without annual accountability in any meaningful normative sense. The Commission may respond by developing soft guidance, recommendations, and country-specific assessments that operationalise the indicative benchmarks without claiming binding force; or it may seek to develop new legislative instruments through different Treaty bases, the internal market provisions, the horizontal social clause of Article 9 TFEU, or targeted employment policy instruments under Article 147 TFEU. Neither path is without risk, and both will be shaped by the constitutional constraints that C-19/23 has now definitively established.

What the Court might have done, and what remains available for the future, is to develop a jurisprudence of social rights that takes seriously the Charter’s standing as binding primary law. Article 31(1)’s guarantee of working conditions respecting dignity, Article 27’s information and consultation rights, and Article 28’s collective bargaining protection together constitute a constitutional grammar for social rights that the Court has so far deployed only fitfully and asymmetrically. A rights-conforming interpretation of Article 153(5), one that reads the pay exclusion in the light of the Union’s fundamental rights commitments rather than in isolation from them, would not have required the Court to fix wages at the European level; it would have required it to recognise that procedural criteria for assessing the adequacy of wages fall on the permissible side of the Treaty line because they are the minimum condition for giving the right to fair working conditions any common European content.

The constitutional grammar for such a development exists. Article 3(3) TEU’s social market economy clause – still awaiting activation as a genuine interpretive principle rather than decorative treaty language – would support a reading of the social competences that treats adequacy as a threshold concept with minimum common content. The ILO and Council of Europe instruments that the Court routinely invokes in other contexts provide the substantive reference points for that concept. And the doctrine of rights-conforming interpretation, well established in the Court’s own jurisprudence on civil and political rights, offers the methodological pathway for integrating these sources into the competence analysis.

There is, finally, a question of constitutional honesty that the judgment avoids. The Union’s institutional actors chose to adopt the Adequate Minimum Wages Directive with the stated aim of ensuring that minimum wages in the EU are adequate to provide a decent standard of living. The Court has confirmed that the Union may legislate toward that aim but has removed the only provision that would have given the shared concept of ‘adequacy’ a common normative content. The result is a legal instrument that formally pursues a goal it is constitutionally forbidden from defining. This ruling fails to establish a sustainable constitutional equilibrium on which further legislative and judicial activity in the years ahead would be taken on social rights.

Judgment C-19/23 defers this constitutional reckoning without resolving it. The Directive survives in attenuated form, procedurally demanding but substantively toothless on the critical question of what ‘adequate’ means. The Union has been confirmed in its power to speak about decent wages and to structure the processes by which States discuss them. It has been told, once again, that it cannot guarantee their common substance. The judgment is, in this sense, a faithful mirror of EU social constitutionalism as it currently stands: sophisticated in form, constrained in ambition, and still searching for the constitutional vocabulary that would allow the promise of Social Europe to match the scale of the social challenge. Whether C-19/23 represents the outer limit of that vocabulary, or merely a point on the arc of a jurisprudence still in formation, is the question that the Court’s future docket will, in time, answer.