How has the Court of Justice changed its management and approach towards the social acquis?

Abstract

Social Europe – Case law of the Court of Justice – ‘Real’ and ‘apparent’ displacement – Court’s interpretive task more complex and contested – Directives based on ‘flexicurity’ policy – Cases in which workers have competing interests, e.g. age discrimination – Viking and Laval – Re-framing of employers’ interests as fundamental rights under Article 16 EU Charter – Declining relevance of the Court in labour law – Challenges for EU labour lawyers

Introduction

The ‘displacement of Social Europe’ is a powerful and provocative title, and has the potential to serve as a rallying-cry for scholars, trade unionists, activists and others concerned with workers’ rights. A range of recent developments, from the decisions in Viking and Laval, ¹ via the flexicurity narrative, to the ‘bailout’ conditions, have all cast doubt on the EU’s commitment to ‘Social Europe’, in the sense of protecting the fundamental social rights of workers as traditionally defined. Rights to engage in collective bargaining and collective action are clearly under threat, as are rights to job security.

I want to explore two ways in which a ‘displacement’ of Social Europe may be occurring within the Court’s case law, in the second of the two senses (‘threatened’) set out by Kilpatrick in her introduction to this special issue. First, while most directives and the cases concerning them continue to be about protecting workers in relation to their employers, some more recent developments have concerned directives which combine protective elements with deregulatory ones, or which address (or purport to address) conflicts between groups of workers with different interests. As I have argued elsewhere, mediating these conflicts is an important secondary function of labour law, but it can also be misappropriated to attack workers’ rights. ² The problem for the Court is that it is much more difficult to construe these directives to give effect to a worker-protective purpose when their objectives are more complex, and when protecting one group of workers might come at the expense, or appear to come at the expense, of protecting another group of workers. This gives rise to what we might term ‘apparent’ displacement: the Court is still striving to give Europe a social face, but it is faced with more polycentric and politically-contested cases. Second, there has been an increasing tendency to elevate employers’ interests in employment cases to a specially-protected status, or even to present them as rights. This has developed partly from the Court’s own case law, particularly in relation to the internal market, and partly from some of the rights recognised in the EU Charter of Fundamental Rights. These cases then appear to involve a clash of rights, which can seem difficult to solve without compromising workers’ rights. But this ignores the underlying power imbalance inherent in the very idea of labour law: that workers are in need of the law’s protection because of employers’ superior bargaining power. It has the potential to lead to a ‘real’ displacement or downgrading of Social Europe.

Understanding the particular legal forms that ‘apparent’ or ‘real’ displacement takes within the Court’s case law is, in my view, a crucial first step in the formulation of any strategy to reinvigorate Social Europe, a point to which I return in the conclusion.

Worker protection

Labour lawyers typically think about labour laws as pursuing worker-protective objectives in an imperfect market: safeguarding workers with weaker bargaining power against exploitation by employers. Of course, this is an oversimplification: worker-protective objectives are never pursued at all costs, and there is always a balance to be struck between the interests of workers and those of employers. What I want to suggest in this section is that at least some relatively recent EU directives in the field of labour law have pursued more complex objectives, making the Court’s interpretive task more difficult, and leading to an ‘apparent’ displacement of Social Europe.

First, some directives have been enacted in pursuit of the EU’s flexicurity agenda, which focuses on encouraging non-standard or ‘flexible’ forms of working and on employability in the labour market rather than security in a particular job. ³ Flexicurity is not a straightforwardly worker-protective set of policies: in some instances, the pursuit of flexicurity may involve removing or limiting some elements of worker protection in order to create jobs or to reduce the differences between standard and non-standard forms of work. The Court cannot be expected to interpret these directives using a straightforwardly worker-protective frame of reference. Second, some directives have brought into sharper focus a secondary function of labour law, to resolve conflicts between groups of workers. In some instances, a decision that protects one group of workers may harm the interests of another group, forcing the Court to make a difficult choice. Some of the Court’s decisions on the Posted Workers Directive have that polycentric character, as do some decisions on age discrimination, which will be the focus of my discussion in this section. ⁴

What does this mean for the displacement of Social Europe? First, although the Court could clearly have handled some of these cases more effectively, it is not entirely responsible for the difficulties, which also reflect decisions of the political actors in drafting legislation. Second, although there is some evidence of displacement, particularly in the flexicurity context, there is also evidence that the ‘social’ is being redefined. This makes the task of analysing and responding to the case law much more complex and, of course, politically contested.

Flexicurity

The directives on part-time, fixed-term and agency work (where the worker is supplied to an ‘end user’ via an employment agency) are, in part, straightforwardly worker-protective, in that they require member states to provide protections for workers engaged in these forms of work, principally in the form of equal treatment with workers employed in ‘standard’ employment relationships. ⁵ However, they also seek to promote the availability of these forms of work to a greater or lesser extent. In relation to part-time work, this is relatively uncontroversial, because it is widely accepted that freely-chosen part-time work can be of significant benefit to the worker where he or she wishes to combine paid work with other activities. In relation to fixed-term or agency work, it is more troubling, since these forms of work are generally insecure and may afford fewer legal rights to workers. ⁶ Controversy has arisen where member states have either been forced to dismantle restrictions on these forms of work or where member states have seized the opportunity presented by the directives to dismantle restrictions on these forms of work and have met with opposition from workers and trade unions at the national level.

Non-regression clauses

Most labour law directives contain ‘non-regression’ clauses. ⁷ Since directives have traditionally been envisaged as laying down minimum standards for protection, rather than a uniform level of protection, non-regression clauses have been seen as a mechanism for ensuring that member states in which workers already enjoyed a higher level of protection than that laid down by the relevant directive would not be able to reduce that to the minimum when implementing the directive. However, the interpretation of these clauses has proved to be controversial in relation to the directives on non-standard forms of work.

The non-regression clause in the social partner agreement annexed to the Directive on Fixed-Term Work is typical: ‘implementation of this Agreement shall not constitute valid grounds for reducing the general level of protection afforded to workers in the field of the Agreement’. ⁸ One issue of contention in the case law has been the proper scope of the non-regression clause. Does the clause apply to the member state’s regulation of fixed-term work generally, or does it apply only to the specific issues addressed by the directive (discrimination against fixed-term workers and abuse of successive fixed-term contracts)? This was addressed in Angelidaki. ⁹ The Court took a broad view, holding that even though the alleged regression had taken place in relation to the employer’s first use of a fixed-term contract (and therefore not a matter addressed by the Directive) the non-regression obligation still applied. ¹⁰ However, there are limits. In Bulicke, the Court held that there was no regression problem where Germany implemented a shorter limitation period for bringing age discrimination claims than the one that already existed for sex discrimination claims. ¹¹ In part, this was because age and sex discrimination were regarded as different fields of law. ¹²

What about the meaning of ‘non-regression’? In Angelidaki, the Court rejected an argument that the non-regression clause required the member state to keep all its existing law in place. Instead, it held that a change in national provisions would only be barred by a non-regression clause if two conditions were met. ¹³ First, the change must be linked to the implementation of the directive in question. If the member state could show that the change is part of an entirely separate domestic policy initiative, the clause would not bite. ¹⁴ Second, the change must significantly reduce the overall level of protection for workers in the member state in question. ¹⁵ This condition requires the national court to consider how many workers are adversely affected by the change and also to assess whether any reduction in protection is outweighed by increases in protection elsewhere. So, for example, it was relevant in Angelidaki that the change only affected a small group of public sector workers and that they benefited from other aspects of the new legislation.

Finally, there is an issue about the enforcement of non-regression clauses. If they had direct effect, individuals would be able to rely on them in the national courts to challenge their member state’s implementation of directives. However, in Angelidaki, the Court confirmed that the non-regression clause in the Directive on Fixed-Term Work did not meet the conditions for direct effect. ¹⁶ Another possibility would be for the national court to tackle the problem using its interpretive powers, but this has been restricted by the decision in Sorge. ¹⁷ In that case, the national court tried to reinstate limits on the use of fixed-term work which had been repealed by the legislature, but the Court held that this went beyond what the duty of compatible interpretation permitted. Thus, even if an individual can establish a breach of the clause – which seems unlikely, for the reasons just given – it may be difficult to identify a means of enforcing it in national law.

The problem the Court faces is that the concept of ‘non-regression’ makes sense where the legislative objective is worker protection, but is much harder to interpret and apply in the context of the ‘flexicurity’ agenda, where legislation may pursue a mix of worker-protective and deregulatory ends. ¹⁸

Review of restrictions

This brings us neatly to a second, but related, issue: the review of restrictions on the availability of non-standard forms of work. Clauses requiring a review feature in the social partners’ agreement on part-time work, implemented by the Directive on Part-Time Work, ¹⁹ and in the Directive on Temporary Agency Work, ²⁰ though (interestingly) not in the social partners’ agreement on fixed-term work.

Article 4 of the Directive on Temporary Agency Work requires member states and the social partners to review prohibitions or restrictions in national law on the use of temporary agency work, and states that such prohibitions or restrictions ‘shall be justified only on grounds of general interest’. ²¹ The proper interpretation of this provision came before the Court in the AKT case. ²² The Advocate General favoured a substantive interpretation of the provision: it prohibited the maintenance or introduction of restrictions on temporary agency work that could not be justified on general interest grounds. He further held that the provision could be enforced horizontally, so that an employer’s association could, in principle, rely directly on Article 4 to challenge a restriction on the use of temporary agency work contained in a collective agreement in litigation against the relevant trade union before the national court.

The Court of Justice chose not to follow the Advocate General and ruled instead that Article 4 merely imposed procedural obligations on national governments. As a result, it could not be invoked by a private party before a national court. The Court’s reading of Article 4(1) sought to locate it in the context of the remaining provisions of Article 4. These include the obligation to conduct a review of national law under Article 4(2), and to report the results to the Commission under Article 4(5), by the transposition deadline for the Directive. On this view, Article 4(1) was ‘addressed solely to the competent authorities of the member states’ and did not impose any obligations on national courts. ²³ The Court was keen to indicate that its interpretation of Article 4(1) did not deprive it of effect, noting that member states were ‘required to comply in full’ with their obligations under the provision, and ‘could have been obliged to amend their national legislation on temporary agency work’. ²⁴ Presumably, the Commission would have been able to bring infringement proceedings had a member state simply opted to maintain in force disproportionate restrictions on the use of temporary agency work. However, the Court emphasised the discretion afforded to member states to decide how best to reform their national legislation:

… the fact remains that the Member States are, to that end, free either to remove any prohibitions and restrictions which could not be justified… or, where applicable, to adapt them in order to render them compliant… ²⁵

Article 4(1) restricted ‘the scope of the legislative framework open to the member states’ but did not require ‘any specific legislation to be adopted’. ²⁶ This nuanced result could not be achieved by permitting scrutiny by the national courts.

The Court’s approach in AKT contrasts with its approach to a similar clause in the Directive on Part-Time Work interpreted in the Michaeler case. ²⁷ In Michaeler, a firm was fined a substantial sum of money by the Italian authorities because it had failed to comply with a statutory requirement to send a copy of each part-time worker’s contract to the labour inspectorate within 30 days of signing it. The firm argued that the notification requirement was an ‘obstacle’ to part-time work and was therefore contrary to the Directive. The Italian government argued that it was justified as a proportionate means of combating the problem of undeclared work. The Court held that the requirement was incompatible with the Directive: ‘The combination of that administrative formality and that system of penalties acts to discourage employers from making use of part-time work.’ ²⁸ It based its conclusion on the fact that there was no equivalent requirement for full-time contracts (making the requirement discriminatory) and on the existence of alternative methods for tackling undeclared work. In contrast to AKT, it was assumed in Michaeler (without a detailed discussion either in the Advocate General’s Opinion or the Court’s ruling) that the requirement on the member state to review and remove disproportionate ‘obstacles’ to part-time working was not simply a matter for the national authorities’ discretion but could be enforced directly by the national courts through the disapplication of the offending ‘obstacles’. ²⁹

At first sight, there are clear parallels between the provision in the Directive on Part-Time Work at issue in Michaeler and Article 4 of the Directive on Temporary Agency Work at issue in AKT. The most plausible explanation for the difference in the two decisions seems to be the textual difference between the two Directives. Article 4(2) of the Directive on Temporary Agency Work places a clear obligation on the member states to conduct a review of restrictions on temporary agency work by a deadline, and Article 4(5) requires that the review be reported to the Commission. These procedural obligations have no equivalent in the Directive on Part-Time Work, which may support the conclusion that the Directive on Temporary Agency Work is intended only to produce procedural effects. A possible textual counter-argument is that Article 4(1) of the Directive on Temporary Agency Work is much more specific than the parallel provision in the Directive on Part-Time Work about the circumstances in which restrictions may be justified – suggesting a substantive obligation – but, as we saw above, the Court’s response to this was to treat Article 4(1) as applying only to the review required by the remaining provisions of Article 4, and not as a free-standing provision. ³⁰ More generally, it is possible that the Court’s tougher stance on the Directive on Part-Time Work reflects the broader normative point that part-time work may be beneficial to workers (and is expressly promoted by the Directive ³¹ ) whereas the same is not true of agency work.

Conclusion

What I have sought to demonstrate in this section is the complexity of the interpretive task now facing the Court in dealing with the directives on atypical work, given that they are not straightforwardly about laying down a set of minimum rights for workers. In pursuit of the flexicurity agenda, these directives may require member states to review restrictions on the use of non-standard working in order to make these forms of work more readily available, which in turn may require the repeal of pre-existing measures in national law that would traditionally have been seen as worker-protective. This presents a challenge for the Court in identifying ‘Social Europe’. If it takes a strict line on non-regression and limits the scope of the review requirements, it risks accusations that it is protecting (already well-protected) workers in standard employment arrangements at the expense of those in non-standard work or the unemployed, and is not reflecting the full range of aims in the directives in its interpretation. If it relaxes its stance on non-regression and allows review requirements to be the subject of litigation in the national courts, it risks accusations that it is abandoning traditional notions of worker protection altogether. This is a much more difficult set of considerations for the Court to navigate and it is not surprising that it has done so with varying degrees of success.

Age discrimination

EU law on age discrimination offers another illustration of the increased complexity of the Court’s interpretive task, for two reasons. ³² First, in a significant departure from the treatment of other protected characteristics, there is a general justification defence available in respect of direct discrimination on grounds of age. ³³ This has generated a large body of case law as the Court has been called upon to give guidance on the application of the proportionality test to some of the many examples of age-related criteria commonly in use in the labour market, with little or no guidance from the EU legislature. ³⁴ Second, in some cases at least, it is possible to portray the issue as a clash between the competing interests of different groups of workers, making it more difficult for the Court to discern what would constitute a worker-protective decision.

The argument that there is a ‘clash’ between workers in some age discrimination cases is particularly apparent in relation to retirement ages. A number of different justifications for compulsory retirement feature in the case law but of particular interest for present purposes is the argument that encouraging older workers to exit the labour market will free up jobs for younger workers. This was accepted by the Court in Georgiev, a case concerning university professors, ³⁵ and in Fuchs, in relation to civil servants and specifically prosecutors. ³⁶ In both cases, the Court assumed that by forcing older workers to retire, job opportunities would be created for younger workers. As well as benefiting those younger workers, the Court appeared to accept that there would be benefits for the organisation from having a broader range of ages represented among the workforce, because of the possibility of ‘exchange of experience’ between different groups. ³⁷ Of course, it is for the national court to determine whether the argument is made out on the facts, so there is some scope for claimants to argue, for example, that there is no evidence that the retirement of older workers in the particular job or sector does in practice lead to the hiring of younger workers, but the important point for present purposes is that this justification is in principle available to employers.

The argument that compulsory retirement creates job opportunities for younger workers is much disputed in general terms. As Sargeant has explained in reviewing the empirical data on the question, the argument rests on the assumption that the number of jobs in the labour market is somehow fixed. ³⁸ In fact, having older people in employment with higher spending power may generate greater demand across the economy as a whole, promoting rather than stifling job creation. There is no strong evidence to support the claim that taking older people out of the labour market improves employment rates among the young. Of course, matters may be different in particular kinds of workplace or sector where the number of jobs is fixed (Supreme Court justices, for example) or relatively static, but these are relatively unusual in practice. In this regard, it may be worth noting that the two leading cases on this matter in the European Court of Justice are, as we have seen, about quite specific professions in the public sector and not about employment in private firms.

In terms of the displacement of Social Europe, the example of retirement ages is interesting because it presents different groups of workers as competing with each other for jobs, in a competition that only one group can win. The Court is being asked to make a politically-contested choice between these groups. Of course, both of the options facing the Court can be presented as worker-protective in their different ways. If its ruling points towards upholding the retirement age by accepting the ‘freeing up’ argument, it protects younger workers but at the expense of older workers, and if it rejects that argument, it appears to be protecting older workers at the expense of younger ones. Thus there is an ‘apparent’ displacement at work in the sense that the Court may still be trying to give Europe a social face, but it can no longer do so in a way that is (relatively) uncontroversial and clear for all to see.

Reconceptualising the interests of employers

So far, I have sought to demonstrate how the task of identifying a worker-protective perspective has become more challenging for the Court, given the emergence of directives with more mixed objectives and (apparent, or sometimes real) clashes between the interests of different groups of workers. I now turn to the other side of the coin: the interests of employers. Here, there has been a tendency in the case law to reconceptualise employers’ interests in a way which presents them as at least as weighty as, if not more so than, the fundamental social rights of workers. The most obvious example of this is Viking and Laval and related cases, in which the Court reasoned that workers exercising their right to engage in collective action needed to be able to justify their ‘infringement’ of employers’ interests in free movement as guaranteed by the Treaty. ³⁹ Perhaps more worryingly, Alemo-Herron and a few subsequent cases have elevated employers’ interests to the status of a fundamental right, by invoking the freedom to run a business in Article 16 of the Charter. ⁴⁰ While, on one view, this does not add very much to the rulings in Viking and Laval, my concern is that invoking the rhetoric of rights serves to reinforce the idea that employers’ and workers’ interests carry equal weight. This is problematic because it downplays the traditional starting-point of labour law, that workers need the law’s support because of the inequality of bargaining power between them and their employers. The decision in late 2016 in AGET Iraklis draws these various strands of case law together and is a neat illustration of the problems. ⁴¹

The decisions in Viking and Laval need no introduction and have, of course, been the subject of a large and generally critical literature. ⁴² What both cases have in common is a shift in what labour lawyers would regard as the usual reasoning structure in a labour law case. The employer’s interests (freedom of establishment in Viking and freedom to provide services in Laval) were described as ‘fundamental’, and the trade unions were required to justify the exercise of their right to strike as a proportionate means of pursuing the legitimate aim of protecting workers’ interests. ⁴³ This displaced the right to strike itself as the usual starting-point, and shifted the focus away from a requirement to justify any interferences with that right. Of course, it is arguable that this followed logically from the structure of EU law, which has, over time, elevated the core principles of the internal market to a fundamental status and is highly alert to possible instances of protectionism, so the Court was left without much choice in the matter. However, the requirement to justify particular exercises of the right to strike on worker-protective grounds is potentially problematic, as I have argued elsewhere, not because the right to strike is somehow incapable of justification, but because it is important for judges evaluating the justificatory arguments to have a good understanding of industrial relations and the role of strikes. ⁴⁴ In particular, an exercise of the right to strike is intended to cause some economic damage to the employer – that is the point – so while it can be highly effective, it is unlikely to be the least harmful option open to the trade union and its members. ⁴⁵

A further problem, though one that was not very clearly articulated in the Court’s judgments themselves, is the polycentricity underlying the cases, so that they also have elements of the problems identified above. In both cases, the right to strike was being invoked by workers in states with generally higher levels of wages and employment protections of various kinds (Finland and Sweden) but it is at least arguable that the employers’ actions would have benefited workers in other member states with lower costs. This is most apparent in Laval, where the employer was seeking to use posted workers from Latvia to perform the contract, who stood to lose out if the employer was unable to proceed. It is less obvious in Viking, since the employer had guaranteed the terms and conditions of employment of the existing workers, but in general terms ship-owners tend to reflag to states with lower regulatory requirements, enabling them to hire workers with less favourable terms and conditions of employment in the future. ⁴⁶

There are, of course, some signs that the Court’s jurisprudence has softened a little since Viking and Laval themselves, and the other two related cases decided at around the same time, Commission v Luxembourg and Rüffert. ⁴⁷ One interesting example of this is the ruling in RegioPost, which departed to some extent from the Court’s earlier decision in Rüffert. ⁴⁸ In RegioPost, a German state government invited bids for a postal services contract, and required bidders to sign an undertaking to the effect that they would pay the statutory minimum wage applicable to public servants in the state. RegioPost refused to give the undertaking and was therefore excluded from the tendering process. It argued that the requirement to pay the wage infringed the freedom to provide services and could not be justified because it applied only in the public sector and could not therefore be regarded as worker-protective in nature. The case fell to be decided under Article 26 of the 2004 Public Procurement Directive, which has since been superseded. ⁴⁹ This laid down two requirements for ‘special conditions’ in public contracts: transparency, which was clearly met, and compatibility with EU law, which the Court considered in the light of the Posted Workers Directive and Article 56 TFEU on the freedom to provide services. The Court found that the minimum wage requirement was compatible with Article 3(1) of the Directive because it was a ‘law’ laying down a ‘minimum rate of pay’. More interesting for present purposes was its discussion of the position under Article 56. It found that the law was a restriction on the freedom to provide services because it was an ‘additional economic burden’ for firms established in states with lower rates of pay, but that it pursued a legitimate aim of protecting workers. ⁵⁰ The Court then turned to the question of justification. Following the ruling in Rüffert, it might have been expected that the Court would, at this point, have found against the state on the grounds that the minimum wage, although a well-recognised mechanism for protecting workers, applied only to public sector workers and was therefore not an effective means of achieving this end. However, the Court simply ignored this issue, enabling it to find that the wage requirement was justified under Article 56 TFEU. While this may have reflected the wording of Article 26, which refers to ‘special conditions’ in public contracts, it also suggests some, albeit rather limited, softening in the Court’s approach towards social justifications for infringements of the free movement rules.

Another, perhaps more promising, development was the Opinion of Advocate General Trstenjak in the Commission v Germany case, ⁵¹ which has been very helpfully analysed by Barnard. ⁵² The case concerned the compatibility with EU law of the practice of many German public authorities of setting up pension funds with providers nominated in collective agreements with their employees rather than selecting a provider after a competitive tendering process in accordance with the EU public procurement regime. The Advocate General noted that the approach taken by the Court in Viking Line had appeared to rank the ‘fundamental freedoms’ – the free movement principles of the EU’s internal market – as occupying a higher position in the legal hierarchy than fundamental rights, such as the right to strike. ⁵³ She suggested instead a ‘double proportionality’ approach for cases of conflict in which a balance would have to be struck between the right and the freedom in question:

A fair balance between fundamental rights and fundamental freedoms is ensured in the case of a conflict only when the restriction by a fundamental right on a fundamental freedom is not permitted to go beyond what is appropriate, necessary and reasonable to realise that fundamental right. Conversely, however, nor may the restriction on a fundamental right by a fundamental freedom go beyond what is appropriate, necessary and reasonable to realise the fundamental freedom. ⁵⁴

Of course, it is arguable that this does not go far enough, in that it does not allow fundamental rights to occupy their usual position at the top of any hierarchy, but given the structure of the Treaty, which does place particular value on free movement, it may be the best that can be expected. ⁵⁵

Perhaps paradoxically, while a better understanding of the implications of describing workers’ rights as fundamental human rights may be helping to counteract the displacement tendencies of decisions like Viking and Laval, there is another side to the EU’s greater focus on fundamental rights that is contributing to displacement. This is the recognition of employers’ interests as fundamental rights, such as the right to run a business or the right to enjoy freedom of contract. The starting point for this aspect of the discussion is Alemo-Herron, a UK case arising under the Acquired Rights Directive. ⁵⁶ In that case, a local authority contracted out its leisure services to a private firm. The contracts of employment of the transferred employees were expressly subject to the collective agreement applicable to local government employees. Importantly, the relevant contractual term was ‘dynamic’ in nature, referring to the collective agreement in force at any given time. After the transfer, therefore, any changes in the local government collective agreement would also apply to the transferred employees. The transferee objected to this because as a private-sector employer, it was not able to participate in the collective bargaining process for local government.

The Court found in favour of the transferee on two grounds. First, it held that the Acquired Rights Directive was intended to strike a balance between protecting workers’ rights and allowing transferees to make operationally-justified changes in terms and conditions of employment. ⁵⁷ It noted that this was particularly important in public to private sector transfers because of the likelihood of greater differences in employment practices. Second, it found that there was an infringement of Article 16 of the EU Charter, which protects the freedom to run a business, because ‘the transferee’s contractual freedom [was] seriously reduced to the point that such a limitation [was] liable to adversely affect the very essence of its freedom to conduct a business’ by virtue of its inability to participate in the collective bargaining process. ⁵⁸ It rejected an argument on behalf of the transferred employees that the dynamic approach to the collective agreement was more favourable to them and was thus protected by the general provision in the Directive allowing a member state to provide higher levels of protection than those afforded by the Directive, on the grounds that (echoing an argument discussed earlier) the purpose of the Directive was not just to protect employees but to balance their interests with those of employers. ⁵⁹

Rather like Viking and Laval, there have been some subsequent rulings in which the Court has refined its approach in Alemo-Herron, so that it no longer entirely precludes dynamic clauses in the context of a transfer of an undertaking, provided that the transferee has freedom to make changes to the transferred employees’ terms and conditions of employment. ⁶⁰ What is more concerning is the possibility that invoking Article 16 might become a routine feature of employers’ arguments in labour law cases, so that rather than simply interpreting EU labour law, the Court’s task becomes one of resolving a supposed ‘clash of rights’ in every case.

The ruling in AGET Iraklis is of particular relevance because it draws together the two strands of discussion about employers’ interests: the employer claimed that Greek law on collective redundancies infringed both its freedom of movement and its Article 16 rights. ⁶¹ Although the case has some unusual features, it may be a worrying sign of things to come.

AGET Iraklis concerned Greek legislation which in effect prohibited firms from carrying out collective redundancies without the agreement of worker representatives or the approval of the relevant administrative authority. The Court began its consideration of the case by interpreting the Collective Redundancies Directive itself. ⁶² It identified the purpose of the Directive as being to protect workers in the event of collective redundancies, but not to preclude them completely. ⁶³ As a partial harmonisation measure, the Directive did not seek to regulate all aspects of redundancy decisions and, in particular, it did not address the substantive conditions under which employers could make workers redundant. Thus, it was open to the Greek government to regulate these issues in legislation. However, national legislation could not operate so as to preclude the practical possibility of making workers redundant which was envisaged by the Directive. ⁶⁴ It was for the referring court to determine whether this was the case.

The Court then examined the situation under Article 49, on freedom of establishment. This was found to be relevant because the Greek firm in the case had a French company as its majority shareholder. The Court held that the freedom to reduce the number of workers employed or even to cease operations altogether in a particular state was a key part of freedom of establishment. Thus, the Greek legislation constituted a restriction on freedom of establishment:

National legislation such as that at issue in the main proceedings is thus such as to render access to the Greek market less attractive and, following access to that market, to reduce considerably, or even eliminate, the ability of economic operators from other Member States who have chosen to set up in a new market to adjust subsequently their activity in that market or to give it up, by parting, to that end, with the workers previously taken on. ⁶⁵

In a confusing passage of the ruling, the Court then appeared to reinforce this aspect of the claim by noting that the employer’s Article 16 rights were also at stake. ⁶⁶ From this perspective, the state’s attempts to justify the legislation in the public interest would only be acceptable if they were also compatible with the employer’s Article 16 rights.

Following on from earlier decisions, the Court accepted that protecting workers and protecting employment levels were legitimate aims for the state to pursue. ⁶⁷ It then stated that it was necessary to ‘reconcile and to strike a fair balance’ between these aims and the employer’s freedom of establishment and Article 16 rights. ⁶⁸ Interestingly, the Court found that the regime put in place by Greek law was both ‘appropriate’ and ‘necessary’ to protect workers’ rights, and capable in principle of being proportionate, but that it was disproportionate in practice (as a restriction on freedom of establishment and the employer’s Article 16 rights) because the criteria to be applied by the public authority (and any reviewing court) were either inadmissible (in the case of the criterion relating to the national economy) or too vague to provide any guidance. ⁶⁹ Moreover, it could not be justified by the special circumstances of the economic crisis in Greece. ⁷⁰

The ruling in AGET Iraklis may be an illustration of the old adage that hard cases make bad law. The Greek legislation on collective redundancies was particularly restrictive of employers’ freedom to make what many people would regard as the core management decision of how many workers to employ. Perhaps it is not a good example of how the Article 16 case law might develop in the future. But what it does demonstrate is that almost any national labour law is potentially open to the charge that it infringes the rights of employers. The trigger is the relatively easily satisfied requirement that the law in question might infringe an employer’s freedom of contract or freedom to do business, and make a particular member state’s market less attractive. Nor does there seem to be any acknowledgement of the underlying purpose of labour law and labour rights, to redress the inherent inequality of bargaining power between workers and their employers. Workers’ rights are presented as something to be balanced with employers’ rights, as if their starting points were equal. Increasing reliance on Article 16 seems likely to enhance the process of ‘real’ displacement of Social Europe set in motion by Viking and Laval.

Conclusion

In this article, I have sought to explore the displacement of Social Europe within the jurisprudence of the Court. I have shown that it has become more difficult for the Court to interpret directives in the field of labour law, either because they no longer pursue straightforwardly worker-protective objectives (as in the case of the directives on non-standard forms of work) or because the Court has been presented with conflicting claims of different groups of workers (as in the example of retirement ages). This has led to an ‘apparent’ displacement in which Europe’s social agenda has become more politically contested. The Court’s interpretations of internal market law and the EU Charter have also elevated employers’ interests to a fundamental status, ignoring the inherent imbalance of power between workers and employers and raising the spectre of ‘real’ displacement: that virtually any aspect of national labour law might be treated as an infringement of employers’ rights. Taken together, these various developments leave Social Europe occupying a rather precarious position in the Court’s jurisprudence.

On one level, perhaps we should not worry about this too much. It is arguable that the Court itself is being ‘displaced’, in the sense of becoming less important as a driver of EU labour law broadly defined. Nowadays, the EU’s influence over national labour laws does not just take the form of the traditional process of enacting and implementing directives. Indeed, as it has become more difficult to secure member states’ agreement to new directives, it is arguable that this is no longer the main way in which the EU influences national labour laws. Instead, the focus has shifted first to the bailouts during the financial crisis, and then to the Open Method of Co-ordination (OMC). ⁷¹ While this method has been employed for many years, the difference since the financial crisis is that the employment element of the Open Method of Co-ordination is now wrapped up in the EU’s attempts to scrutinise the member states’ economies. ⁷² Some member states have come under sustained pressure to deregulate certain aspects of their labour laws, particularly in relation to job security. ⁷³ For present purposes, what is interesting – and challenging – about this development is that it has become much more difficult to identify EU legal acts that might be the subject of review or interpretation by the Court. Of course, it has been argued – including, persuasively, by Kilpatrick – that the Court’s jurisdiction extends widely, but the uncertainty surrounding this issue seems likely to be a factor in dissuading potential claimants from bringing claims. ⁷⁴

But what can be done about the displacement, apparent or real, of Social Europe within the Court’s jurisdiction? I want to propose two, admittedly somewhat contradictory, alternatives. First, it is arguable that the division of powers between the Court and the political actors in the EU is not ideal. The precise balance between the EU’s economic and social aims is, in large part, the proper province of governments and electorates, but it is clear that the difficulty of securing agreement on the part of the member states and the social partners to new directives in the field of labour law leads to a tendency to leave sensitive matters to the Court to decide with little in the way of guidance (either because no directive can be agreed at all or because what is enacted is vague on key points). This leaves the Court with no alternative but to decide contentious political questions as best it can. One challenge for labour lawyers, therefore, is to find ways to encourage greater political engagement with labour law matters, for example, through lobbying or advising their national governments and social partners. Second, where polycentric or politically-sensitive matters do fall to the Court to decide, it is important to ensure that the Court is presented with the best possible information about the social and legal background to the cases it must decide. For example, the Court needs good information about the industrial relations context and the labour laws in the member states: what they say, why they were enacted, and what empirical data there is about their effects. It is important for scholars and for advocates before the Court to research these matters thoroughly, going beyond traditional legal frames of reference and engaging more fully in interdisciplinary work. It is not easy for lawyers to admit that we do not have all the answers, but it may be essential to do so if we are to restore the pursuit of Social Europe to its proper place as one of the EU’s core purposes as defined in the Treaties. ⁷⁵