Abstract This article offers a critical analysis of the concept of ‘worker’ in European Labour Law as predominantly shaped by the CJEU on the basis of its jurisprudence in the area of free movement of workers. It suggests that, while nominally broad, this concept suffers from many of the strictures arising from the traditional binary divide between employment and self-employment on which it is essentially premised. Having analysed the residual fragmentation and legitimacy problems arising from a rather incomplete development of this area of EU law, the article advocates a broader concept of ‘worker’ shaped by reference to the fundamental rights language embraced by various regional and international labour and human rights mechanisms, including the Social Charter, the European Convention on Human Rights and the International Labour Organisation. The article concludes by identifying a number of possible reform options that may serve the purpose of establishing a suitably broad and autonomous concept of ‘worker’ for the purposes of EU labour law, and by offering a cursory analysis of some recent EU Pillar of Social Rights initiatives. 1. INTRODUCTION Traditionally, the letter of EU Labour Law has either shied away from defining the personal scope of application of its provisions or expressly reserved that task to domestic legal systems. The Treaty of Rome, for instance, referred to the term ‘worker’ in a number of provisions, most famously in those introducing the principle of ‘equal pay for equal work’1 and those guaranteeing ‘Freedom of movement for workers’,2 but without offering any explicit indication in respect of its meaning. Other provisions, for instance those contained in the core body of secondary instruments regulating working conditions,3 appear to reserve the role of defining terms such as ‘employee’ or ‘contract of employment or employment relationship’ exclusively to national law. An example of this second approach can still be found in Article 2(2) of the Transfer of Undertakings Directive 2001/23, providing that ‘This Directive shall be without prejudice to national law as regards the definition of contract of employment or employment relationship’.4 However, as further explored in section 2 of this article, there are also a number of instruments that do not expressly defer this key definitional task to ‘national law’, some of which5 have been interpreted by the CJEU as relying on a concept of ‘worker’ that is, by and large, equivalent to that eventually developed by Court’s jurisprudence in the free movement of workers context. As noted by the Commission, this fragmented approach ‘leaves a margin of appreciation to Member States and to their courts’.6 But fragmentation and flexibility come at a cost too. For instance, in the past, the Commission has recognised the risks engendered by this approach, acknowledging that the ‘consistent application of EU labour law can be put in question, particularly in the context of the transnational operation of businesses and services, through the variations in the definitions of worker used in different directives’ and that often ‘[c]ontinued reference to national rather than Community law could […] affect worker protection’.7 Indeed, in the past it went as far as exploring, and consulting publicly, as to whether there was ‘a need for more convergent definitions of ‘worker’ in EU Directives in the interests of ensuring that these workers can exercise their employment rights, regardless of the Member State where they work’.8 The Union’s concern with the personal scope of application of its ‘social acquis’ arguably reached a climax with the official launch of its ‘European Pillar of Social Rights’ initiative in 2016.9 The earlier documents underpinning this initiative demonstrated an acute awareness that ‘there are “grey zones”, such as “dependent” and “bogus” self-employment, leading to unclear legal situations and barriers to access social protection’10 with the Commission becoming increasingly vocal in suggesting that ‘[e]qual treatment shall be ensured, regardless of employment contract’.11 These strong assertions eventually lead, in Spring 2017, to a number of proposals for various legislative initiatives,12 at least two of which could have both direct and indirect consequences on the categories of workers to which EU social rights apply.13 Issues of consistent application and worker protection aside, the precise definition of the personal scope of EU employment legislation is also fundamentally important in terms of the nature of EU labour law and its relationship with other areas of EU law and with domestic labour law systems. As incisively noted by Bercusson, the ‘[c]ontention over the precise legal nature of the employment relationship is not new to European labour law. Frequently, it was the issue which, in some countries, enabled labour law to break free of civil law and become an autonomous discipline’.14 It is arguable that the emergence of an autonomous EU definition of ‘worker’ at the EU level could similarly assist EU labour law in its process of emancipation from other areas of EU law that exert some influence over its scope and nature, in particular the area of ‘free movement of workers’,15 and from the many and largely diverse national concepts of employment contract or employment relationship that are grafted upon EU directives when the latter are implemented domestically. This, of course, would not be a simple task. As sections 2 and 3 of this article explore, in recent years the ‘worker’ concept developed by the CJEU in its case law under Article 45 TFEU,16 has been applied to a growing number of EU labour law directives in a way that would render any significant deviation from this increasingly consolidated jurisprudential course extremely difficult and uncomfortable for the Court. Also, the sheer diversity of approaches that national labour law systems take when defining their personal scope of application inevitably suggests that any attempts to converge them towards a unified supranational concept would be met with some resistance, especially—though not exclusively—by employers’ groups.17 But the emergence of an autonomous EU definition of ‘worker’ uniformly applicable to all or most EU labour law instruments would arguably also entail fundamental constitutional implications in terms of the vertical relationship between the EU legal system and national legal orders. Ictu oculi, the Treaty suggests that EU intervention in the labour law sphere should be limited to the introduction of ‘minimum requirements for gradual implementation’.18 The Court has sometimes interpreted this provision as constraining EU action to ‘partial harmonization’ and as ‘not […] intended to establish a uniform level of protection throughout the Community on the basis of common criteria’,19 thus discouraging, if not preventing, the introduction, at least by means of judicial interpretation, of an autonomous or more convergent definition of worker in those directives expressly deferring to national concepts. According to Davies, ‘[t]his reflects the EU’s limited role in “harmonising” labour law rather than creating a uniform system throughout the Member States’.20 Claiming exclusive competence in this domain may run contrary to the spirit, if not the letter, of the Treaties and could be confronted with fundamental, competence based, objections. More recently, and in particular in the context of its Communication on the ‘collaborative economy’, the Commission has affirmed that ‘while EU Member States are responsible for deciding who is to be considered a worker in their national legal order, at EU level the Court of Justice (CJEU) has defined the concept of worker for the purpose of applying EU law [and has] confirmed that this definition shall also be used to determine who is to be considered as a worker when applying certain EU Directives in the social field’.21 So while there seems to be no shortage of compelling reasons for the EU to redress its current overly fragmented approach to defining the personal scope of application of its labour law directives, and of its social acquis as a whole, there are serious risks that this exercise could run against the grain of increasingly established jurisprudential practices, largely divergent national approaches to the conceptualisation of the employment relationship, and fundamental constitutional and competence based objections. Having clarified this background, the following sections of this article explore the extent to which the development of an autonomous concept of ‘worker’ at the EU level would be legitimate, practical and desirable. The following sections 2 and 3 begin by exploring the status quo in terms of the concept(s) of worker currently deployed in various areas of EU social law. These parts suggest that, for the time being, the Court has taken a leading role in construing the scope of application of EU labour law directives and its action relies on a two pronged strategy. For a number of directives that do not expressly reserve the scoping exercise to implementing Member States, the Court has progressively deployed and consolidated the ‘worker’ concept originally developed by its jurisprudence in the area of free movement of workers. However, in respect of other instruments that explicitly confine their scope of application to whatever notion of worker or employment contract prevails at a national level, the Court has taken a more cautious approach and has not systematically interfered or claimed the competence to introduce an autonomous and EU level ‘worker’ definition as such. Still, section 3B below notes, even in respect of this second group of instruments, the Court is increasingly steering their scoping provisions in ways that are pushing towards a convergence with the ‘worker’ concept mentioned above and does so, in particular, whenever Member States ‘apply rules which are liable to jeopardise the achievement of the objectives pursued by a directive and, therefore, deprive it of its effectiveness’.22 Section 3 also acknowledges that not all directives have been streamlined on the basis of this two pronged and increasingly converging approach and that there is a persisting fragmentation problem. On the basis of this analysis, section 4 proceeds to identify some of the protective gaps existing in the current scope of application of EU labour law ratione personae. This section posits that some gaps emerge as a consequence of the fragmentation problem outlined in the opening paragraphs of this chapter, but others are to a certain extent a consequence of the emergence of the Article 45 TFEU concept of ‘worker’ and its spread to various areas of EU labour law. That concept, it is suggested, lends itself to the criticism of being unduly premised on a crude binary divide between subordinate employment and autonomous self-employment in a way that does not assist much with the establishment of adequate working conditions amongst and increasingly diverse and segmented European workforce. Section 5 explores alternative options and approaches to redress some of these gaps, and discusses alternative framing and scoping concepts that are, or ought to be, much more in tune with the largely universalist and fundamental rights aspirations of some EU labour law provisions, as well as with some important developments taking place at the ECtHR and ILO level. Section 6 discusses some regulatory trajectories and reform options currently being explored by the European Commission, and concludes. 2. THE EMERGENCE OF THE WORKER NOTION(S) IN THE ‘FREE MOVEMENT OF WORKERS’ CONTEXT As noted in the opening paragraph of this article, EU primary and secondary law had initially shied away from defining the concept of ‘worker’ in any substantial and meaningful way. One would of course struggle to find a national legal system where the concepts of ‘work’, ‘worker’ or ‘contract of employment’ are defined in great detail in primary legislation, but it is fair to say that the silence of the EU founding treaties, and of the first few directives adopted in the 1960s and 1970s to confer rights to workers,23 is particularly surprising given that the regulation of the ‘four freedoms’, including of course free movement of workers, was one of the main preoccupation of the Treaty of Rome. By contrast, one could appreciate that the drafters of the Treaty did at least attempt to give some indications in respect of their understanding of the concept of ‘services’ and of natural persons entitled to establish themselves in other members states.24 But as early as 1963, in the Case 75/63, Hoekstra,25 the Court of Justice decided to claim for itself the power to establish a ‘Community meaning’ for the term ‘worker’, at least in respect of the EC/EU provisions shaping the law on ‘Free movement of workers’, under what is now Article 45 TFEU. The Court has since refined its jurisprudence on the concept of ‘worker’ relying on three main criteria to identify who is a worker for the purposes of Article 45 TFEU. The first criterion is arguably one that many labour lawyers from continental Europe would recognise as a ‘subordination’ requirement. As noted in Lawrie-Blum, ‘The essential feature of an employment relationship…is that for a certain period of time a person performs services for and under the direction of another’.26 The second criterion is the remuneration element, with the Court requiring that work provided for and under the direction of another be work in return for which remuneration is received. The Court has interpreted this requirement broadly in the free movement context, accepting for instance that ‘the sole fact that a person is paid a “share” and that his remuneration may be calculated on a collective basis is not of such a nature as to deprive that person of his status of worker’,27 and even that services and other benefits in kind provided in lieu of a regular salary, ‘may be regarded as being an indirect quid pro quo for their work’.28 And it may also accept as remuneration payments that do not derive exclusively from the employer but that are ‘largely provided by subsidies from public funds’.29 The third requirement is that the worker is engaged in ‘effective and genuine activities, to the exclusion of activities on such a small scale as to be regarded as purely marginal and ancillary’.30 In this context, the Court has stated that the services provided should not be ‘merely a means of rehabilitation’ but instead form ‘part of the normal labour market’,31 although it has made considerable allowances for ‘the fact that the productivity of persons employed…is low’ and that a person may be employed under public supported schemes.32 Arguably, this third criterion is the one that the Court has sought to expand as broadly as possible in order to construe a ‘worker’ definition that would not discourage persons engaging in low-intensity or low-productivity forms of employment from exercising their free movement rights. By and large the ‘worker’ concept that has emerged from the Court’s jurisprudence in the free movement context reproduces the traditional binary divide between subordinate employment and autonomous self-employment embedded in the labour law systems of the original founding member states.33 While, eventually, some of those systems—as well as the systems of some of the new Member States—evolved to recognise new and, broadly speaking, intermediate categories of quasi-subordinate or economically dependent self-employment,34 the Court has been consistent in asserting that ‘any activity which a person performs outside a relationship of subordination must be classified as an activity pursued in a self-employed capacity’.35 However, and this goes almost without saying, in the free movement context, the strictures of this Euro-binary divide are substantially mitigated by the fact that self-employed and own-account workers enjoy separate and autonomous free movement rights from both the Treaties and secondary legislation.36 This is a very important consideration, as in practice it means that the stakes arising from the ‘binary divide’ and the dependent worker v self-employed professional classification in the area of free movement are lower than those arising in the labour law context. In the former case, the classification of a work relationship as autonomous will not deprive the person of her free movement rights. What will deprive the person of free movement rights is the extent to which she or he is genuinely engaged in an economic activity, whether under the control of an employer or not. By contrast, in the labour law context, being classified as a dependent worker is an essential requirement to qualify for the protective panoply offered by EU or domestic employment protection systems. The following section explores the extent to which, in spite of these shortcomings, the ‘free movement’ concept of ‘worker’ has progressively been applied to a growing range of instruments, mainly but not exclusively directives, regulating EU level labour rights and the extent to which the Court has sought, expressly or impliedly, to engage with some of the fragmentation, legitimacy and appropriateness concerns outlined in the introduction of this paper. 3. THE UBIQUITOUS NOTION OF ‘WORKER’: BETWEEN THE ROCK OF LEGITIMACY AND THE HARD PLACE OF COHERENCE In parallel with the early and bold developments in the ‘free movement’ context outlined above, the Court was also developing a separate and much more cautious jurisprudence in respect of the concept of ‘worker’ or ‘contract of employment or employment relationship’ notions referred to in a number of EU social and labour law directives. As noted in the introduction to this article, this more cautious approach can be partly explained by the rather strong textual argument offered by some, though not all, labour law directives that seemed to reserve the definition of these terms to the domestic legal systems of Member States. A good example of this more cautious approach can be found in Case 105/84, Danmols Inventar, where the Court established that, in contrast with the area of free movement of workers,37 ‘directive no 77/187 is intended to achieve only partial harmonization’ and thus that ‘it follows that directive no 77/187 may be relied upon only by persons who are, in one way or another, protected as employees under the law of the member state concerned’.38 The ‘Danmols orthodoxy’, premised on the idea that the instruments it refers to are instruments of ‘partial harmonization’, endured the test of time rather successfully, certainly until the early years of the twenty-first century, with one or two minor qualifications. More recent case such as C-343/98, Collino and Chiappero,39 for instance, fundamentally relied on Danmols to exclude some workers from the scope of the Transfer of Undertakings Directive. Other cases have relied on the same jurisprudential line, but to confirm the inclusion of workers already recognised as such at the national level.40 But while there are several instruments that, just like the Acquired Rights Directive analysed in Danmols, seemingly reserve their personal scope definition to national legal systems,41 some other EU labour law provisions do not do so, or certainly do not do so explicitly.42 And in respect of these latter instruments, the Court has taken a different approach, slowly but surely departing from Danmols. A. Departing from the Danmols Orthodoxy In Case C-78/98, Preston, the Court decided that intermittent contracts could constitute a ‘stable employment relationship’ for the purposes of equal pay for work of equal value, but without claiming or referring to an autonomous concept of worker. However, soon after this decision, in Case C-256/01 Allonby, the Court took a further step in the direction of an autonomous concept and, having noted that ‘there is no single definition of worker in Community law: it varies according to the area in which the definition is to be applied’, it effectively imported in the area of equal pay the ‘free movement’ definition of worker outlined in section 2 above.43 The Court boldly added that ‘The formal classification of a self-employed person under national law does not exclude the possibility that a person must be classified as a worker within the meaning of Article 141(1) EC if his independence is merely notional, thereby disguising an employment relationship within the meaning of that article’.44 With this statement the Court went as close as possible to the suggestion that some national classifications could be tantamount of ‘shams’ at least for the purposes of the application of EU ‘worker’ rights. For independence to be more than just notional, the Court would typically expect the absence of ‘any relationship of subordination concerning the choice of that activity, working conditions and conditions of remuneration’, and for the activity to be performed ‘under that person’s own responsibility [and] in return for remuneration paid to that person directly and in full’.45 Following Allonby, the Court has progressively sought to reclaim an autonomous ‘worker’ concept for a number of other EU labour law instruments as well. For instance, the Court has been willing to provide an autonomous definition for the personal scope of application of two Directives that share the common trait of being adopted on a health and safety legal basis and policy rationale,46 the Pregnant Workers Directive and the Working Time Directive. Again, in doing so, it expressly referred to its jurisprudence on the free movement ‘worker’ concept. So, in Case C-116/06, Kiiski, it held that ‘the sui generis nature of the employment relationship under national law cannot have any consequence in regard to whether or not the person is a worker for the purposes of Community law’, and that ‘the Community legislature, with a view to the implementation of Directive 92/85, intended to give the concept of “pregnant worker” a Community meaning’,47 eventually applying the free movement ‘worker’ concept to the directive.48 No less importantly, in C-232/09, Danosa the Court applied this jurisprudence to include within the scope of Directive 92/85 a pregnant member of the board of directors of a capital company.49 In this case, the Court expressly noted that even though, because of her managerial role, Ms Danosa ‘enjoyed a margin of discretion in the performance of her duties’,50 she had to be treated as a ‘worker’ covered by the directive because, inter alia, ‘she had to report on her management to the supervisory board and to cooperate with that board’.51 So it is clear that the Court adopts a fairly generous and nuanced notion of subordination that does not require an employer to be constantly watching over the shoulders of a worker, and can effectively amount to a power of ‘control’,52 ‘direction or supervision’53 or ‘to cooperate’,54 especially when such workers are ‘an integral part of’55 the company they provide services to. Subsequently, the Court applied this approach to the interpretation of the term ‘worker’ referred to in the Working Time Directive, noting in C-428/09 Union syndicale Solidaires Isère that ‘for the purposes of applying Directive 2003/88, that concept may not be interpreted differently according to the law of Member States but has an autonomous meaning specific to European Union law [and] must be defined in accordance’ with the definition developed ‘for the purposes of Article 39 EC, Case 66/85 Lawrie-Blum  ECR 2121, paragraphs 16 and 17’.56 But it is arguably in the context of Directive 98/59 that the Court has departed most markedly from the Danmols orthodoxy. It is worth noting that the collective redundancies directive, unlike the Transfer of Undertakings and the Insolvency Directives, does not expressly defer to national definitions of workers, employment contract or employment relationship. In Case C-32/02, Commission v Italy, the Court held that excluding from the scope of Directive 98/59 redundant employees solely because of the not-for profit nature of their employer amounted to an incomplete transposition of the Directive. Perhaps most remarkably, in Case C-596/12, Commission v Italy, the Court—in dealing with the Italian exclusion of ‘managers’ from the protection offered by the Directive—noted that ‘la notion de «travailleur» […] a une portée communautaire’,57 and continued by confirming the application, by analogy, of its ruling in Danosa.58 More recently, in Case C-229/14, Balkaya v Kiesel Abbruch- und Recycling Technik GmbH, it went on to include within the scope of this instrument a member of the board of directors of a capital company and a trainee performing services ‘with financial support from, and the recognition of, the public authority responsible for the promotion of employment—in order to acquire or improve skills or complete vocational training’.59 So, de facto, even in respect of the collective redundancy directive, the CJEU has moved away from the ‘partial harmonization’ justification to one premised on the approximation of national rules and embraced the broad Danosa concept of ‘worker’, which includes workers in senior managerial positions but also trainees. B. Recasting the Danmols Orthodoxy: Effectiveness and Legitimacy Abandoning the ‘Danmols orthodoxy’ in respect of those directives that do not expressly confine to national legal orders the detailed definition of their personal scope could have been seen as a relatively unproblematic exercise, at least from the Court’s point of view. Sure, doubts could have been raised in terms of subsidiarity and appropriate choice, but a suggestion that the Court/EU had no competence in spelling out the scope of application of these instruments could not have been supported by a clear textual argument. But the Court must have surely realised that departing from Danmols in respect of instruments that reserve more explicitly this scope defining task to the Member States would have caused major resistances and questioned the legitimacy of the Court’s activity. For these instruments, the Court has arguably adopted a different approach to the one deployed in and since Allonby. But, as this subsection explores, this approach is constantly evolving and it remains questionable whether it can protect the Court from the allegation of judicial activism. A first erosion to Danmols was produced by a series of judgments prescribing the inclusion of specific categories of workers within the scope of some of these directives. For instance, in respect of the application of the fixed-term work Directive to public sector employees, at some point the Court sought to include such workers within the Directive’s scope, even though the instrument suggested that the scope definition was, in essence, a matter reserved to national systems.60 In Del Cerro Alonso, it went as far as including within the equal treatment provisions of Directive 99/70 a temporary member of staff on the grounds that ‘Directive 1999/70 and the framework agreement are applicable to all workers providing remunerated services in the context of a fixed-term employment relationship linking them to their employer’.61 But these were tentative steps and the Court, for instance in cases such as C-313/02, Wippel, would generally be inclined to re-affirm and defer to the importance of national definitions.62 It is worth noting however, that in Wippel itself the orthodoxy of Danmols was questioned by AG Kokott noting that ‘It could therefore constitute a breach of the duty of cooperation (Article 10 EC) if a Member State were to define the term “worker” so narrowly under its national law that the Framework Agreement on part-time work were deprived of any validity in practice and achievement of its purpose…were greatly obstructed’.63 But a more marked departure64 from the Danmols orthodoxy occurred with Case C-393/10, O’Brien and the more recent judgment of the Court in Case C-216/15, Betriebsrat der Ruhrlandklinik,65 two cases that partly rely on AG Kokott’s insightful comments. In O’Brien, a UK part-time judge paid on a fee basis was effectively being excluded from an occupational retirement scheme because under UK law he was not an employee but, rather, a ‘judicial office holder’. The Court noted that ‘the discretion granted to the Member States…in order to define the concepts used in the Framework Agreement on part-time work is not unlimited…[they] may not apply rules which are liable to jeopardise the achievement of the objectives pursued by a directive and, therefore, deprive it of its effectiveness’.66 It also went on to offer concrete ‘principles and criteria’ to the referring court on how to assess whether a part-time worker was a worker for the purposes of the Directive, in particular by reference to the ‘differentiation between that category and self-employed persons’ and by reference to ‘the rules for appointing and removing [workers], and also the way in which their work is organised’.67 Although the Court maintained that ‘it is for the Member States to define the concept of “workers who have an employment contract or an employment relationship” in Clause 2.1 of the Framework Agreement on part-time work’, in a way that arguably sought to distinguish this judgment from the Allonby and Kiiski/Danosa line of cases discussed above, it also qualified this by demanding that ‘that does not lead to the arbitrary exclusion of that category of persons from the protection offered by Directive 97/81’, with such exclusions being ‘permitted only if the relationship between [part-time worker and employer] is, by its nature, substantially different from that between employers and their employees falling, according to national law, within the category of workers’.68 This line of reasoning is relied upon, and arguably further entrenched, by the more recent Betriebsrat der Ruhrlandklinik judgment. In this case, the employer was arguing that nurses supplied by the German Red Cross to its clinic did not have the status of employees of a temporary-work agency under German law, in order to circumvent the opposition by the local works council that, in line with domestic legislation, objected to the secondment of agency staff on a non-temporary basis. In its decision that these nurses were ‘workers’ for the purposes of Directive 2008/104, the Court acknowledged that the ‘concept covers “any person who, in the Member State concerned, is protected as a worker under national employment law”’.69 But having done that, it recited the Danosa definition of ‘worker’ and, having pointed out that the ‘directive applies not only to workers who have concluded a contract of employment with a temporary-work agency, but also to those who have an “employment relationship” with such an undertaking’,70 went on to suggest that a Red Cross nurse ‘cannot be excluded from the concept of “worker” […], on the sole ground that she does not have a contract of employment with the temporary-work agency and that she therefore does not have the status of worker under German law’.71 Upon reflection there is a strong sense that the Court reached this conclusion by relying, in effect, on its Allonby/Danosa case law, and that the principle of effectiveness, while acknowledged,72 is arguably less central to the Court’s ratio decidendi than in the O’Brien judgment. The ‘national definition’ proviso contained in the Directive, and acknowledge ob iter by the Court, emerges as severely diminished by the Court’s emphasis that it ‘cannot be interpreted as a waiver on the part of the EU legislature of its power itself to determine the scope of that concept for the purposes of Directive 2008/104’,73 a fortiori in respect of a directive that, the Court says, in Article 3(1)(a) provides a basic definition of ‘temporary agency worker’. The Court suggests instead that ‘that provision means only that the EU legislature intended to preserve the power of the Member States to determine the persons falling within the scope of the concept of “worker” for the purposes of national law and who must be protected under their domestic legislation’,74 an aspect that Directive 2008/104 does not aim to harmonise. This is arguably a very flimsy distinction, especially considering that, according to the Court, a worker is ‘any person who has an employment relationship in the sense set out in [Danosa] and who is protected, in the Member State concerned, by virtue of the work that person carries out’,75 as opposed to a particular national status or sui generis classification. With Betriebsrat der Ruhrlandklinik, the Court appears to have gone full circle in its claim that the personal scope of application of EU labour law instruments ought to be, ultimately, a matter for EU law to define. This is a strong claim to autonomy that goes well beyond the more cautious O’Brien approach and arguably departs from the ‘partial harmonization’ approach of Danmols. The Court is clearly increasingly aware that the equal treatment protective objectives pursued by the Directive can be jeopardised by some rather peculiar and idiosyncratic national classifications of non-standard workers as non-employees, and is obviously willing to intervene and substitute any national classification with its own concept of ‘worker’ at least for the purposes of the application of the rights and principles contained in EU labour law directives. C. A Residual Fragmented Approach The Betriebsrat der Ruhrlandklinik line of reasoning has yet to be deployed consistently across all directives that reserve the scope definition exercise to Member States. So the fragmented approach discussed in the opening paragraphs of this article, and the problems that it entails, remains very much a concern. Examples of the persistence of this fragmented approach arise from cases such as C-306/07, Ruben Andersen, where the CJEU asked national courts deciding on the concept of ‘temporary contract or employment relationship’ for the purposes of Article 8 of Directive 91/533, to ensure that ‘duration must, however, be fixed so as to provide effective protection of the rights conferred on workers by the directive’.76 Not an explicit departure from Danmols, but not exactly a hands-off approach either. Similarly, in case C-385/05, CGT, the Court noted that although ‘the second subparagraph of Article 3(1) of [Directive 2002/14] provides that Member States are to determine the method for calculating the thresholds of employees employed. […] that provision concerns determination of the method of calculation of the thresholds of employees employed and not the actual definition of the concept of an employee’.77 A similar conclusion was reached in Case C-176/12, AMS, with the result that national provisions excluding employees on ‘assisted contracts’ from the calculus for the information and consultation threshold was deemed incompatible with Directive 2002/14. It is important to note however that in both CGT and AMS, the Court found that the MSs itself, France, already considered these relationships on a par with standard employment contracts, which made it harder to justify their exclusion from the threshold calculus. In Case C-242/09, Albrons, the Court held that a worker could be held to be employed through an ‘employment relationship’ by a ‘non-contractual employer’, even in the presence of a ‘contractual employer’ because ‘it is not apparent from Directive 2001/23 that the relationship between the employment contract and the employment relationship is one of subsidiarity and that, therefore, where there is a plurality of employers, the contractual employer must systematically be given greater weight’ (paragraph 25).78 The Court is yet to pronounce itself on the personal scope of application of Directive 2010/18/EU on Parental Leave, though we know from the Kiiski judgment that although ‘according to paragraph 7 of Clause 2 of the framework agreement, the Member States and/or management and labour are to define the status of the employment contract or of the employment relationship for the period of parental leave provided for under that agreement’, the CJEU understands this provision as suggesting, at the very least, ‘that the working relationship between the worker and his employer [is] maintained during the period of child-care leave. As a result, the beneficiary of such leave remains, during that period, a worker for the purposes of Community law’.79 It is difficult to anticipate whether this fragmented landscape is likely to be restructured and brought in line with the Betriebsrat der Ruhrlandklinik jurisprudence so that, ultimately, the scope of EU labour law provisions will be framed, in essence, by reference to the FMW/Allonby/Danosa ‘worker’ notion. Directive 2008/104 is admittedly rather more insistent that other directives in defining at least ‘the contours of the definition of “temporary agency worker”’.80 The Court may feel less inclined to act as robustly with other instruments that do not define such contours, and may decide to assess their scope by reference to the more cautious O’Brien approach. What is certain, however, is that the contention that ‘at EU level the Court of Justice (CJEU) has defined the concept of worker for the purpose of applying EU law [and has] confirmed that this definition shall also be used to determine who is to be considered as a worker when applying certain EU Directives in the social field’81 is becoming increasingly tenable, even though a number of core of EU labour law instruments still do not benefit from a single, EU-level, ‘worker’ definition. 4. MIND THE GAP: UNCHARTERED WATERS AND PROTECTIVE GAPS While legitimacy and coherence are important questions, this section is more explicitly preoccupied with a third set of concerns that pertain to the worker protective potential, or lack thereof, of the Court’s approach outlined above, in sections 2 and 3, above. While the Court has undoubtedly developed its jurisprudence on the personal scope of application of EU Social law instruments in a worker protective direction, a number of jurisprudential and protective lacunae remain. Some of these protective lacunae, it is suggested, directly arise from the choice by the CJEU to embrace the ‘worker’ definition developed in the free movement context and extend it to the labour law context. While the Court should not be over-criticised for taking this step—the free movement ‘worker’ concept being both broad and jurisprudentially consolidated—it should be noted that in the free movement context, disputes typically arise at the boundary between economic activity and inactivity. As long as a person is economically active, she will be allowed to enjoy EU free movement rights regardless of her classification as employee or self-employed, the latter group also enjoying EU free movement rights. Unsurprisingly, the part of the EU ‘worker’ definition that the Court of Justice has sought to expand as broadly as possible is precisely the ‘effective and genuine economic activity’ part, since the non-economically active are entitled to a reduced and heavily qualified set of free movement and residence rights.82 By contrast, disputes over labour rights mainly arise at the boundary between subordination and independence. But this is precisely the part of the ‘worker’ definition that the Court has felt less compelled to expand in its free movement jurisprudence, as by and large EU citizens can enjoy free movement rights either as subordinate workers or as independent professionals and services providers. The ‘worker’ definition relied upon by the Court can possibly disenfranchise a number of workers that are not, strictly speaking, dependent but that may well be in need of protection. A. Rights of Quasi-Dependent Workers? The CJEU’s approach, both in Allonby, Betriebsrat der Ruhrlandklinik, and in O’Brien, de facto embraces a binary notion of employment relationship: a person is either a subordinate worker/employee under the broad Danosa definition or if they perform an activity ‘outside a relationship of subordination must be classified as an activity pursued in a self-employed capacity’.83 Unlike a number of EU Member States,84 the Court does not contemplate any intermediate category of quasi-subordination or economic dependence.85 So if a quasi-subordinate worker were to raise a question of coverage and application of a particular EU social law instrument and that question were to be referred to the CJEU, the Court would probably want to pigeon-hole that relationship by suggesting it be categorised by the referring court as either a relationship of employment or one of self-employment. As noted above, in C-256/01, Allonby, the Court expressed no hesitation in re-classifying a relationship of self-employment into a ‘worker’ relationship, and in O’Brien, it was not concerned by the UK’s definition of part-time judges as ‘office holders’. Unlike some domestic conceptions of ‘employee’ or ‘worker’, such as the UK one for instance,86 the Court is not overly preoccupied with the form of the relationship and with whether it meets particular contractual requirements and is adamant that ‘neither the legal characterisation, under national law, of the relationship between the person in question and the [employer], nor the nature of their legal relationships, nor the form of that relationship, is decisive for the purposes of characterising that person as a “worker”’.87 The EU ‘worker’ concept is undoubtedly one that endorses a notion of subordination that includes ‘control’,88 but it also acknowledges more nuanced and loose concepts such as ‘direction or supervision’,89 that may include within its scope some domestic notions of quasi-subordinate work relations. However the extent to which the Court may be willing to bring under the EU concept of ‘worker’ national self-employed workers that are economically dependent from one main ‘client’90 or user remains an open question.91 On a more positive note, it must be acknowledged that the Court has equipped itself with a jurisprudence that ought to allow it to reclassify national employment statuses that arbitrarily deprive workers of their EU rights. Both Allonby and O’Brien are good examples, but the Court went further in case C-413/13, FNV Kunsten, offering instructions to referring courts on how to deal with ‘false self-employed’ workers.92 There is little doubt that this instruction could apply to false para-subordinate workers, if the effect of this national label were to deprive them of rights derived from EU law. Also some legal instruments now expressly refer to the need of ‘combating bogus self-employment’,93 and some go as far as suggesting the introduction of ‘legal presumptions’ of status and duration.94 And it is certainly encouraging that some AGs would have categorised workers with rather atypical or sui generis relationships as EU ‘workers’.95 But the Court has hitherto resisted the notion of ‘quasi-subordination’ as a genuine third category. Its broad concept of ‘worker’ may, to a certain extent, mitigate the consequences of a strict binary divide. But this notion, as broad as it may be, effectively pushes the economically dependent self-employed fully within the realm of genuine self-employment, a category that, the next subsection suggests, is possibly not attracting a sufficient level of protection under EU law. B. Rights of Genuinely Self-employed Persons? If working persons are labelled as ‘self-employed’, a number of rights that EU Social Law provisions reserve to ‘workers’ or persons with a ‘contract or employment relationship’ may not be applicable to them. While some areas of EU social law apply expressly to the self-employed,96 the majority of EU labour law provisions do not cover autonomous workers. A particularly problematic manifestation of this issue appears in respect of collective bargaining (and possibly of collective labour rights at large), where organisations representing self-employed persons in processes effectively amounting to the collective negotiations of rates of pay, may be seen by the CJEU as operating outside the ‘Albany exclusion’ and possibly acting in violation of EU Competition Law.97 This is so because, according to the Court, the Treaties do not contain any provisions ‘encouraging the members of the liberal professions to conclude collective agreements with a view to improving their terms of employment and working conditions’.98 In FNV Kunsten, the Court offered the referring court detailed suggestions on how to identify ‘“false self-employed” that is to say, service providers in a situation comparable to that of employees’.99 In doing so, national authorities must assess the status of workers ‘during the contractual relationship’ and compare them with ‘employees who perform the same activity’.100 The tests and indicators suggested include the familiar concepts of control, in the sense of acting ‘under the direction of his employer as regards, in particular, his freedom to choose the time, place and content of his work’, business risk, and integration, in the sense of ‘forming an economic unit with that undertaking’.101 But once more, these test may prove to be too strict for some quasi-dependent workers that are genuinely allowed a wide degree of autonomy in choosing the time, place and content of their work, and whose activities are mainly coordinated by the ‘client/employer’, and will most certainly expose to the harsh realities of EU competition law a number of workers that provide personal work or services with the support of some capital assets owned by them, and thus be presumed to be a separate economic unit. This approach is recognised as being particularly problematic in the context of posting practices under Dir 96/71 where the host Member State protections recognised by the instrument may not apply to the genuinely self-employed.102 Some EU labour rights have been designed as specifically applying to self-employed workers. For instance EU equal treatment provisions have been mostly articulated so as to apply very broadly ‘in relation to […] conditions for access to employment, to self-employment and to occupation, including selection criteria and recruitment conditions’.103 EU equal pay instruments, however, do not expressly refer to the self-employed and have been designed and interpreted by the CJEU, for instance in Allonby, as applying to ‘workers’.104 In some legal systems, the Court’s insistence that EU equal pay legislation ought to apply to (subordinate) ‘workers’ has been understood (arguably wrongly) by national courts as implying or requiring that the whole EU equal-treatment legal apparatus be exclusively applicable to subordinate employees.105 In other systems, equal-pay legislation is applied, in principle, beyond employment and may even extend to volunteering.106 The Court of Justice has yet to pronounce itself on this issue, but it is clear that, for the time being, self-employed workers in the EU may not be enjoying a set of fundamental labour rights including the right to bargain collectively and possibly even aspects of the right not to be discriminated against. 5. BEYOND THE EURO-BINARY APPROACH: EXPLORING A BROADER CONCEPT OF WORKER Most of the provisions contained in the Charter of Fundamental Rights of the EU are very broadly phrased as applying to either ‘everyone’ or to ‘workers’ at large. Some rights are designed as applying to ‘all areas, including employment, work and pay’.107 A number of provisions, especially the ones in the Solidarity chapter, are qualified by reference to the according to ‘national laws and practices’ formula. In Case C-316/13, Fenoll, the Court clarified that a disabled person in a non-profit rehabilitation centre for disabled people (whose status was classified domestically as that of a ‘user’) was indeed a ‘worker’ for the purposes of the application of both Directive 2003/88 and Article 31(2) of the Charter, by reference to leave entitlements. While it is arguable that a similar decision would have been taken even in the absence of the Charter provision,108 the Court explicitly noted ‘that the concept of a “worker” within the meaning of Article 7 of Directive 2003/88 and Article 31(2) of the Charter must be interpreted’ (emphasis added) as applying even to work relations, such as the one in Fenoll, regardless of ‘the level of productivity of the individual concerned, nor the origin of the funds from which the remuneration is paid, nor even the limited amount of that remuneration’109 as long as they ‘are not created for the sole purpose of providing an occupation, derivative if necessary, for the persons concerned [and] although adapted to the capabilities of the persons concerned, have a certain economic value too’.110 While consistent with its earlier jurisprudence, it is fair to say that this judgment pushes the boundaries of the ‘worker’ concept further, partly on the back of the fundamental nature of the right in question. Following Fenoll it remains unclear whether the Court would have taken a similar and equally broad approach in respect of a Charter right that is not also restated or contained in a Directive, for instance in respect of the right collective bargaining, or the right to protection against unjustified dismissal, covered by Articles 28 and 30, respectively. As noted by the Opinion of AG Mengozzi in Fenoll, the established jurisprudence of the CJEU on the horizontal application of Charter provisions, especially Case C-282/10, Dominguez and the Advocate General’s opinion in that case, suggests that this may not be possible.111 But, strictly speaking, in Fenoll, the Court only explicitly excluded that the Charter could have ‘retroactive’ horizontal effect and did not exclude (or accept) that the provision may be capable of horizontal effect. Fenoll may thus be an important step in the direction of asserting, eventually, an autonomous notion, that is, one that is independent from the existence of secondary legislation, of the term ‘worker’ for the purposes of the Charter’s provisions. A further point to be taken into account is that, according to Article 52(3) of the Charter, for ‘rights which correspond to rights guaranteed by the Convention for the Protection of Human Rights and Fundamental Freedoms, the meaning and scope of those rights shall be the same as those laid down by the said Convention’. Convention rights are very broadly phrased usually by reference to the term ‘everyone’. The Strasbourg Court has interpreted and applied this term generously. In the case of Vörđur Ólafsson v. Iceland, for instance, it did not hesitate to recognise a self-employed artisan as a beneficiary of the protection to freedom of association granted by Article 11 of the Charter.112 In Sindicatul Păstorul cel Bun v Romania the Court included members of the Romanian clergy within the scope of Article 11(1) of the Convention and, importantly, did so by reference to ILO R-198, on the grounds that ‘the duties performed by the members of the trade union in question entail many of the characteristic features of an employment relationship’,113 noting that ILO Convention 87 provides, in Article 2, that ‘workers and employers, without distinction whatsoever’ have the right to establish organisations of their own choosing.114 Given the growing recognition granted by the ECtHR to ‘elements of international law other than the Convention, [and] the interpretation of such elements by competent organs’,115 including ILO instruments and their interpretation by ILO mechanisms, it is worth recalling that in construing the scope of application of Conventions 87 and 98, the ILO Committee on Freedom of Association has long established that ‘The criterion for determining the persons covered by that right…is not based on the existence of an employment relationship, which is often non-existent, for example, in the case of agricultural workers, self-employed workers in general or those who practice liberal professions, who should nevertheless enjoy the right to organize’.116 More recently the Committee requested the Korean Government to take the necessary measures to: (i) ensure that ‘self-employed’ workers, such as heavy goods vehicle drivers, fully enjoy freedom of association rights, in particular the right to join the organizations of their own choosing; (ii) to hold consultations to this end with all the parties involved with the aim of finding a mutually acceptable solution so as to ensure that workers who are self-employed could fully enjoy trade union rights under Conventions Nos 87 and 98 for the purpose of furthering and defending their interest, including by the means of collective bargaining; and (iii) in consultation with the social partners concerned, to identify the particularities of self-employed workers that have a bearing on collective bargaining so as to develop specific collective bargaining mechanisms relevant to self-employed workers, if appropriate117 The Preamble to the EU Charter also notes the link between its provisions and those contained in the Social Charter, a point further reaffirmed by the Presidium to the Convention, whose ‘explanations’118 are considered as authoritative by the Charter itself.119 These explanations systematically refer Charter rights to the equivalent Social Charter provisions, that are (often) cast and interpreted as applying to a rather broad range of employment and work relations, including to some self-employed workers.120 The recent complaint filed by ICTU in respect of the Irish competition authorities declaring unlawful a collective agreement between Equity/SITP and the Institute of Advertising Practitioners, on the ground ‘that self-employed actors are undertakings and that Equity is an association of undertakings when it acts on behalf of self-employed actors’121 may well provide an opportunity for the Social Charter’s European Committee for Social Rights to define its jurisprudence on the collective labour rights of self-employed workers.122 What is clear is that, in the not so distant future, the fundamental rights nature of a number of labour rights protected at a regional and international level could, and arguably should, act as a catalyst for the future expansion of the personal scope of application of important areas of the European Social Pillar,123 including an expansion ‘au-delà de l’emploi’. 6. CURRENT AND FUTURE TRAJECTORIES: INTEGRATION, SUBORDINATION AND UNIVERSALISM The previous sections have identified three main trajectories in the, arguably still developing, evolution of the personal scope of application of EU labour law. The first trajectory is a ‘segmented integration’ trajectory, whereby an autonomous and rather broad and relational EU concept of ‘worker’, mainly derived from the ECJ/CJEU’s jurisprudence on free movement of workers, is emerging as the central defining paradigm for a large number of EU labour and social law directives. Even EU directives expressly conferring their personal scope to national systems, for instance the Part-time Workers or the Temporary Agency Workers Directives are no longer immune from this pervasive concept of EU worker. The Court’s willingness to offer to national systems clear and often detailed guidance, replicating the essential features of its ‘worker’ notion, when it thinks that the effectiveness of the protection afforded by these directives is endangered by dubious and narrow domestic personal scope definitions, certainly reinforces this integration trajectory. The ‘segmentation’ aspects of this integration process manifest themselves in two ways. First, some areas of EU labour law have yet to be visited or revisited by the CJEU and pigeonholed into the Allonby or the O’Brien approaches. This is likely to occur in the future, although it is possible that not all areas could be affected in the short or medium term. Second, some national systems are giving a very minimalistic and indeed segmented application to the Court’s interpretation of the ‘worker’ concept. For instance, in the UK, the same domestic definition of ‘worker’ under section 230(3)(b) ERA 1996 is likely to be framed according to the CJEU’s instructions in disputes affecting rights and matters protected by EU directives, and be shaped in much narrower and contractualistic terms in disputes about rights understood by national judges as defined by national law exclusively, for instance rights protecting whistleblowers.124 This element of fragmentation is possibly inevitable but it is also undesirable in terms of clarity and coherence of national labour law systems and in terms of their less than integrated relationship with EU law. The second trajectory manifests itself in the form of a ‘modified subordination approach’. Essentially, the personal scope of application of most EU labour law directives remains anchored to an idea of work as the provision of labour ‘for and under the direction’ of an employer. Autonomous work will normally not fall under the protective scope of EU labour law. This approach is a ‘modified’ one in that EU law seeks to make two adjustments to the strictures arising from a narrow binary divide. First, in a number of cases the Court has embraced slightly more nuanced forms of subordination, contenting itself with work performed under the ‘direction or supervision’ of the employer,125 or deeming it sufficient for workers ‘to cooperate’126 with the employing entity in order to fall within the scope of EU labour law directives. These nuances should lead to the inclusion of some national intermediate or quasi-subordinate employment relationship within the EU concept of worker, as long as their status is not one of genuine self-employment. Second, the Court is increasingly willing to look beyond national definitions and classifications of workers as independent contractors, stating that the ‘formal classification of a self-employed person under national law’ does not exclude the possibility that a person must be classified as an EU worker if independence is merely notional.127 The Court’s approach in that sense is neither contractualistic nor formalistic, and it is clear that the Court understands the formula ‘contract or employment relationship’ as essentially requiring the application of EU labour rights to workers whose employment patterns may not meet national contractual requirements or formalities.128 The third trajectory is one that could be defined as an ‘incrementally universalistic’ trajectory. EU law in the social sphere is not confined to subordinate workers but, on occasion, expressly confers rights to the self-employed, with a number of EU institutions and bodies recently advocating for a further gradual extension of important parts of the EU social acquis to self-employed workers.129 The Court of Justice is arguably beginning to play a more active role in developing this trajectory. Judgments such as Fenoll arguably do more than just embrace the human rights parlance. In effect they extend the status of ‘workers’ to persons that provide personal work or services to another party even in a context of, strictly speaking, unpaid work, as long as the work or services are recognised as having a certain economic value. These case-law developments have led some academic authors to argue ‘that remuneration is not an essential part of the employment relationship if all the other conditions are satisfied…in particular the condition of subordination’.130 If this assessment is correct, and there is no reason to suggest the contrary, the CJEU concept of worker could soon be expanded to include a number of marginal labour market arrangements, such as volunteering, that in some Member States have been hitherto understood as falling outside the scope of EU social law.131 However, the much more universalistic approaches embraced by other regional and international courts and interpretative bodies, such as the European Court of Human Rights or the ILO Committees, suggest that the CJEU and EU law as a whole have a considerable way to go before the rights of self-employed and of persons engaging in various forms of unpaid work are truly protected. It is reasonable to expect these three trajectories to consolidate in the years to come, with the Court continuing to play a leading role in the process. But the process is also likely to crystallise some of the shortcomings associated with the emergence and spread of the ubiquitous concept of EU worker in EU labour law. It is also increasingly apparent that the EU law making institutions and, within their prerogatives, the European social partners, may have to interject in order to consolidate the acquis and, possibly, to redirect its development on the basis of more solid constitutional and policy foundations. The ‘European Pillar of Social Rights’ initiative suggests that there is a certain awareness of the pitfalls, actual and potential, arising from the presence of ‘“grey zones”, such as “dependent” and “bogus” self-employment, leading to unclear legal situations and barriers to access social protection’.132 It also reflects an awareness that ‘Regardless of the type and duration of the employment relationship, workers have the right to fair and equal treatment regarding working conditions, access to social protection and training’.133 Other initiatives, such as the one ‘establishing a European Platform to enhance cooperation in tackling undeclared work’,134 also identify ‘The abuse of the status of self-employed persons, as defined in national law, either at national level or in cross-border situations’135 as a regulatory challenge to be addressed by ‘enhancing cooperation between Member States’ relevant authorities and other actors involved in order to tackle…bogus self-employment’.136 Some of these initiatives are increasingly framed in regulatory terms, as best exemplified by the recent first phase consultation of the European Social Partners by the Commission in respect of a possible revision137 of Directive 91/533 and in respect of a new instrument ‘addressing the challenges of access to social protection for people in all forms of employment’.138 The document introducing the proposed revision of the Written Statement Directive correctly identifies ‘the “grey” area between self-employment and subordinate employer-employee arrangements, especially in the case of bogus self-employment’,139 as the ‘main issue’ affecting compliance with the existing instrument, and suggests that ‘a common definition of worker or employee for the purpose of the application of the Directive could be explored to align it with the trend of Court’s case-law’.140 A similar development would in many ways assist in addressing some of the shortcomings identified in the present article. But, arguably, any further advances in terms of guaranteeing an adequate coverage for this instrument would require a further extension of its scope to include a broader range of employment relationship,141 including those currently potentially excluded under its Articles 1(2) and 3(1),142 and a sounder legal basis for the ‘sham self-employment’ doctrine developed by the Court. This could be achieved by consolidating the Court’s jurisprudence and reviving the prior elaborations developed in the context of sham self-employed workers in the road transport sector,143 with the view of including within the scope of the Directive both ‘workers’ and ‘any person who is not tied to an employer by an employment contract or by any other type of working hierarchical relationship, but: a) who does not have the freedom to organise the relevant working activities; b) whose income does not depend directly on the profits made’.144 Obviously, the concept of sham self-employment would have to be understood in a context where ‘the worker must be regarded as the weaker party to the employment contract and it is therefore necessary to prevent the employer being in a position to disregard the intentions of the other party to the contract or to impose on that party a restriction of his rights’.145 Finally, and from a worker protective angle, it is arguable that the emerging universalistic trajectory discussed in this article ought to be consolidated and supported by the EU, with a number of fundamental labour rights, from freedom of association, to collective bargaining, to equal treatment and access to social security, being applicable to all, including the genuinely self-employed providing personal work and services. Article 2(a) of Directive 2010/41 could offer a rudimentary definitional base for the purpose of clarifying the concept of self-employment and own-account work (or, under the Eurostat definition, the concept of ‘self-employed persons without employees’146) and distinguishing it from other concepts that are more intrinsically linked to ideas of entrepreneurship and the coordination, let alone accumulation, of capital assets, that may not sit comfortably within the protective remit of labour law,147 and in fact may suggest that an individual or an entity is actually an employer.148 It is worth noting that some of the more progressive national reform discourses are increasingly advocating broader constructions of the personal scope of domestic labour rights, by referring, for instance, to any ‘person…engaged by another to provide labour and is not genuinely operating a business on his or her own account’,149 or ‘all workers’ with a ‘contract of subordinate and autonomous employment’,150 or to both ‘dependent’ employees and ‘autonomous’ or ‘externalised’ salaried workers.151 Some of these re-regulatory initiatives could be complemented by means of one or more soft law instruments offering guidance to Member States on how to define, domestically, the concept of worker, in particular, but not exclusively, for the purposes of EU law. ILO Recommendation 198 of 2006 could offer a basis for a similar exercise, with aspects of this guidance being incorporated in the Europe 2020 processes. It is certainly true that ‘in today’s economy, the distinction between the traditional categories of worker and self-employed person is at times somewhat blurred’ and that in ‘a number of cases […] the working relationship between two persons (or one person and one entity) [may] not—because of its peculiar features—fall neatly into one or other category, displaying features characteristic of both’.152 But many of these definitional challenges can be managed simply by systematising and consolidating a range of existing doctrines and juridical concepts, while others admittedly require a slightly more ambitious commitment to the understanding of personal work relations153 in twenty-first-century Europe, and the development of a ‘worker’ notion that is genuinely protective and autonomous and capable of breaking free of both national and EU strictures. Footnotes 1 Art 119 of the original EEC Treaty. 2 Art 48 of the original EEC Treaty. 3 Cf. below n.41. It should be noted that all three Directives incorporating Framework Agreements concluded by the European social partners embrace this model. 4 Although the instrument seeks to constrain Member states’ discretion, for instance were they to exclude some forms of short-term, fixed-term or temporary work, see Art 2(2)a–c of Directive 2001/23. 5 For instance the Working Time Directive or the Collective Redundancies Directive. Cf. below n.42. 6 EU Commission, ‘Commission Staff Working Document – The EU social acquis – Accompanying the document Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions Launching a consultation on an European Pillar of Social Rights’, (COM(2016) 127 final) (SWD(2016) 51 final), para 3.1. 7 EU Commission, ‘Green Paper Modernising Labour Law to Meet the Challenges of the 21st Century’ COM(2006) 708 final, p 14. 8 Ibid. 9 See ‘Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions – Launching a Consultation on a European Pillar of Social Rights’, COM(2016) 127 final. 10 EU Commission, ‘ANNEX – First Preliminary Outline of a European Pillar of Social Rights Accompanying to the Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions – Launching a Consultation on a European Pillar of Social Rights’, COM(2016) 127 final ANNEX I, p 5. Emphasis added. 11 Ibid. In the first preliminary outline of the Social Pillar, above n.10, these words constituted Principle 5(a) entitled ‘Flexible and Secure Labour Contracts’. 12 See ‘Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions – Establishing a European Pillar of Social Rights’, COM(2017) 250 final. See in particular the accompanying ‘Commission Staff Working Document SWD’ (2017) 201 final, especially at p 21. This Communication should ultimately lead to the adoption of a joint Proclamation adopted by the Commission, Parliament and Member States, and outlined, in a draft form, in ‘Proposal for a Interinstitutional Proclamation on the European Pillar of Social Rights’, COM(2017) 251 final. 13 See EU Commission, ‘Consultation Document of 26.4.2017 – First phase consultation of Social Partners under Article 154 TFEU on a possible action addressing the challenges of access to social protection for people in all forms of employment in the framework of the European Pillar of Social Rights’, COM(2017) 2610 final; and EU Commission, ‘Consultation Document of 26.4.2017 – First phase consultation of Social Partners under Article 154 TFEU on a possible revision of the Written Statement Directive (Directive 91/533/EEC) in the framework of the European Pillar of Social Rights,’ COM(2017) 2611 final. 14 B. Bercusson, European Labour Law, 2nd edn (Cambridge: CUP, 2009), 370. 15 On the adequacy, or lack thereof, of the free movement of workers ‘worker’ definition beyond the limited scope of A 45 TFEU see L. Nogler, ‘Rethinking the Lawrie-Blum Doctrine of Subordination’ (2010) IJCLLIR 83–101; on its limitations even within the free movement of persons framework, see T. van Peijpe, ‘EU Limits for the Personal Scope of Employment Law’ (2012) ELLJ 35–53. 16 On the emergence of the concept see N. Countouris, The Changing Law of the Employment Relationship (Aldershot: Ashgate, 2007), esp. Ch 5, and Van Peijpe, above. 17 A tangible example of such resistances can be found in some of the submissions made by various organisation appearing in the 2007 House of Lords EU Committee Report ‘Modernising European Union labour law: has the UK anything to gain? Report with Evidence’ (22nd Report of Session 2006–07) esp. at pp 30–1. See in particular the submission by Business Europe that there is ‘no need for a more convergent definition of worker in EU directives and would strongly oppose moves seeking to indirectly harmonise existing national definitions’, pp 131 of the report, and by the EEF that ‘A standard EU definition of “worker” and “employee” is unworkable, largely due to the differences in Member States’ tax and social security regimes’, pp 148 of the report. 18 Art 153(2)(b). 19 Case 105/84, Foreningen af Arbejdsledere i Danmark v A/S Danmols Inventar  ECR 2639, para 26. 20 A. C. L. Davies, EU Labour Law (Cheltenham: Edward Elgar, 2012), at 174. 21 European Commission, ‘Communication From the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions – A European Agenda for the Collaborative Economy’ COM(2016) 356 final, page 12. 22 Para 35 of Case C-393/10, O’Brien. 23 See for instance Regulation (EEC) No 1612/68 of the Council of 15 October 1968 on freedom of movement for workers within the Community. 24 Cf. Art 60 of the Treaty of Rome, specifying that ‘“Services” shall in particular include: (a) activities of an industrial character; (b) activities of a commercial character; (c) activities of craftsmen; (d) activities of the professions’. 25  ECR 177. 26 Case 66/85, Lawrie-Blum, para 17. 27 Case C-3/87, Agegate, para 36. 28 Case 196/87, Steymann, para 12. 29 Case C-344/87, Bettray v Staatssecretaris van Justitie, para 15. 30 Case 53/81, Levin, , Case C-337/97, Meeusen, . 31 Case 344/87 Bettray; C-456/02 Trojani. 32 Case 344/87 Bettray, para 15; Case Case C-1/97, Birden, 23–32. 33 See N. Countouris, The Changing Law of the Employment Relationship, esp. Chs 1 and 2. 34 A. Perulli, ‘Economically Dependent/Quasi-subordinate (Parasubordinate) Employment: Legal, Social and Economic Aspects’ (Report for the European Commission, 2003); EESC, Opinion of the European Economic and Social Committee on ‘New trends in self-employed work: the specific case of economically dependent self-employed work’ (own-initiative opinion)  OJ C 18/08. Also, G. Davidov, ‘Setting Labour Law’s Coverage: Between Universalism and Selectivity’ (2014) 34 OJLS 543. 35 Joined Cases C-151/04 and C-152/04, Durré, para 31. See also Case C-268/99, Jany and Others  ECR I-8615, para 34 and the decisions cited there. 36 For instance under Arts 49 and 56 TFEU and the Citizenship Directive 2004/38. 37 See the analysis at para 24. 38 Paras 26–27. 39 Esp paras 36–41. 40 See Case C-108/10, Scattolon, para 39 in particular. 41 To name a few examples, cf. Art 2(2) of Directive 2008/94/EC of the European Parliament and of the Council of 22 October 2008 on the protection of employees in the event of the insolvency of their employer; Clause 2(1) of Directive 97/81 regulating part-time work and Directive 99/70 on fixed-term work, and Art 3(1)(a) of Directive 2008/104 on temporary agency work; Clause 1(2) of Council Directive 2010/18/EU of 8 March 2010 implementing the revised Framework Agreement on parental leave concluded by BUSINESSEUROPE, UEAPME, CEEP and ETUC and repealing Directive 96/34/EC; Art 1(1) of Directive 91/533/EEC of 14 October 1991 on an employer’s obligation to inform employees of the conditions applicable to the contract or employment relationship; Art 2(d) of Directive 2002/14 of the European Parliament and of the Council of 11 March 2002 establishing a general framework for informing and consulting employees in the European Community. 42 For instance cf. the Directive 2003/88/EC of the European Parliament and of the Council of 4 November 2003 concerning certain aspects of the organisation of working time; Council Directive 98/59/EC of 20 July 1998 on the approximation of the laws of the Member States relating to collective redundancies; Council Directive 92/85/EEC of 19 October 1992 on the introduction of measures to encourage improvements in the safety and health at work of pregnant workers and workers who have recently given birth or are breastfeeding. 43 Allonby, para 67. 44 Allonby, para 71. 45 Case C-268/99, Jany and Others v Staatssecretaris van Justitie, para 70. 46 Although it is not suggested here that the Court would extend this by analogy to other H&S instruments, and the obiter dicta in para 27 of Case C-116/06, Kiiski, suggest that the Court is aware that, for instance, the letter of Directive 89/391 suggests ‘the definition of a “worker” to be derived from national legislation and/or practices’. 47 Para 24 of Kiiski. 48 Paras 25–26. 49 paras 38–42 of the Danosa judgment. 50 Para 49. 51 Ibid. 52 51. 53 56. 54 49. 55 56. 56 Para 28. 57 Para 16. 58 Para 17. 59 Case C-229/14, Balkaya v Kiesel Abbruch- und Recycling Technik GmbH, para 49. 60 Cf. Clause 2(1) of the Framework agreement on Fixed-term Work. See cases such as Case C-212/04 Adeneler and Others, paras 54–57; Case C-53/04 Marrosu and Sardino, paras 40–43; and Case C-180/04 Vassallo, paras 32–35. 61 See para 28, see also 27–29 and the opinion of AG Maduro paras 11–12 in particular. 62 See para 40 of Wippel in particular. 63 See para 45 of her Opinion. 64 This departure is all the more relevant and visible because in an earlier case on part-time workers, C-313/02, Wippel, the Court had appeared to be re-affirming the importance of national definitions, cf. para 40 of Wippel. 65 Of 17 November 2017 (yet to be reported). 66 O’Brien, para 34. 67 O’Brien, paras 43–45. 68 Para 51 of O’Brien. 69 Paras 25 and 26, Betriebsrat. 70 Ibid., para 28. 71 Ibid., para 29. 72 In a rather cursory way at para 36. 73 Para 32. 74 Para 31. Emphasis added. 75 Case C-216/15, Betriebsrat, para 33. 76 C-306/07, Ruben Andersen, para 54. 77 C-385/05, CGT, para 33. 78 On the concept of ‘employer’ in EU law see J. Prassl, The Concept of the Employer (Oxford: OUP, 2015), esp. ch 3. 79 Case C-116/06, Kiiski, para 32. 80 Betriebsrat, para 32. 81 EU Commission, ‘Communication From the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions – A European Agenda for the Collaborative Economy’ COM(2016) 356 final, page 12; see also the even bolder assertion that ‘C-216/15, Ruhrlandklinik ... indicates that, whatever the references made to national definitions in the text of the Directive 2008/104/EC (Temporary Agency Work) and even in the absence of explicit reference to an EU definition, the concept of worker developed in the field of free movement of workers is applicable’, in European Commission ‘First phase consultation of Social Partners under Article 154 TFEU on a possible revision of the Written Statement Directive (Directive 91/533/EEC) in the framework of the European Pillar of Social Rights’ C(2017) 2611 final, page 8, n.18. 82 See for instance Art 7(b) of Directive 2004/38. 83 Joined Cases C-151/04 and C-152/04, Durré, para 31. See also Case C-268/99, Jany and Others  ECR I-8615, para 34 and the decisions cited there. 84 For an extensive comparative analysis cf. European Parliament, ‘Social Protection Rights of Economically Dependent Self-employed Workers’, IP/A/EMPL/ST/2012-02 (April 2013). 85 This is something that can be implied by the dicta in O’Brien, esp. at . 86 For a recent examples, see Gilham v Ministry of Justice, Appeal No UKEAT0087/16/LA, where a District Judge was held to be an ‘office holder’ and therefore not a ‘worker’ for the purposes of whistleblowers’ protection legislation. 87 Case C-216/15, 29. 88 Danosa, 51. 89 Danosa, 56. 90 Such as the Spanish Trabajadores autónomos económicamente dependientes. Cf. E. Sánchez Torres, ‘The Spanish Law on Dependent Self-Employed Workers: A New Evolution in Labor Law’ (2010) 31 CLLPJ 231; S. Sorge, ‘German Law on Dependent Self-Employed Workers: A Comparison to the Current Situation Under Spanish Law’ (2010) 31 CLLPJ, 249; J. Fudge, ‘A Canadian Perspective on the Scope of Employment Standards, Labor Rights, and Social Protection: The Good, the Bad, and the Ugly’ (2010) 31 CLLPJ, 253; and J-P Landa Zapirain, ‘Regulation of Dependent Self-employed Workers in Spain: A Regulatory Framework for Informal Work?’ in J. Fudge, S. McCrystal and K. Sankaran (eds), Challenging the Legal Boundaries of Work Regulation (Oxford: Hart, 2012), 157. 91 Cf V. De Stefano, ‘Non-Standard Work and Limits on Freedom of Association: A Human Rights-Based Approach’ (2016) 45 ILJ 22–3. 92 Case C-413/13, paras 33–38. 93 Para 10 of the Preamble of the Posted Workers Enforcement Directive—which however reiterates that ‘According to Directive 96/71/EC, the relevant definition of a worker is that which applies in the law of the Member State to whose territory a worker is posted’ 94 See Art 6(3) of Directive 2009/52/EC of the European Parliament and of the Council of 18 June 2009 providing for minimum standards on sanctions and measures against employers of illegally staying third-country nationals. It is fair to say that its track record in respect of national legal presumptions of status is all but encouraging, see Case C-255/04, Commission of the European Communities v French Republic and Case C-577/10, European Commission v Belgium. 95 See AG Kokott Opinion in Wippel at paras 53–54. 96 For instance, free movement and freedom of establishment under Arts 49 and 56 TFEU and the Citizenship Directive 2004/38; a number of Health and Safety rights under Directives 86/613/EEC and 2002/15; arguably specific aspects of equal-treatment law under Directive 2010/41, but also under some specific provisions such as Art 14(1)(a) of the Recast Directive 2006/54, and Arts 3(1)(a) of the Race Directive 2000/43 and the Framework Directive 2000/78. 97 See the decision in Case C-413/13, FNV Kunsten. 98 Joined Cases C-180/98 to C-184/98, Pavlov and Others v Stichting Pensioenfonds Medische Specialisten,  ECR I-6451, para 69. 99 FNV Kunsten, para 31. 100 FNV Kunsten, para 37. 101 FNV Kunsten, para 36. 102 See the critical analysis developed in N. Countouris and S. Engblom, “‘Protection or protectionism?’: A Legal Deconstruction of the Emerging False Dilemma in European Integration” (2015) 6 European Labour Law Journal 20. 103 For example, Art 3(1)(a) of Directive 2000/43. 104 Cf. for instance Art 4 of Directive 2006/54. 105 See in particular the UK Supreme Court case Jivraj v Hashwani  UKSC 40. This set of issues is further explored in the report N. Countouris and M. Freedland, ‘The Personal Scope of the EU Sex Equality Directives’ (2012), report produced for the European Commission by the European Network of Legal Experts in the Field of Gender Equality. Available at http://ec.europa.eu/justice/gender-equality/files/your_rights/personal_scope_eu_sex_equality_directive_final_en.pdf (date last accessed 27 August 2017). 106 N. Countouris and M. Freedland, ‘The Personal Scope of the EU Sex Equality Directives’ (2012), at p 4 refer to at least four EU Member States applying equal pay legislation to the self-employed. 107 Art 23 ‘Equality between women and men’, but see also Art 15, Freedom to choose an occupation and right to engage in work. 108 The judgment refers copiously to precedents such as Bettray and Kiiski, discussed above. 109 Fenoll, para 34. 110 Ibid., para 40. 111 See in particular para 60 of the Opinion of the AG in Fenoll. 112 Application no 20161/06. 113 Application no 2330/09, see para 143. 114 Ibid. 142. It is worthwhile contrasting this with the narrower approach expressed in the concurring but partly dissenting opinion by Judge Wojtyczek, that Art 11 ECHR ‘applies to all those who carry on a gainful occupation involving a relationship of subordination vis-à-vis the person they are working for’ (para 3). 115 See most famously para 85 of the Court’s decision in Demir and Baykara v Turkey  ECHR 1345. See K. D. Ewing and J. Hendy, ‘The Dramatic Implications of Demir and Baykara’ (2010) 39 ILJ 2–51. 116 ILO Committee on Freedom of Association (2001) Report No 326, Case No 2013, para 416. See also C. Rubiano, Collective Bargaining and Competition Law: A Comparative Study on the Media, Arts and Entertainment Sectors (IFM, 2013). See also the report of the discussion held within the Conference Committee on the Application of Standards (CAS) on the application of C-98 to Irish freelance journalists, held in the 2016 International Labour Conference, available at http://www.ilo.org/dyn/normlex/en/f?p=NORMLEXPUB:13101:0::NO::P13101_COMMENT_ID:3082151 (date last accessed 27 August 2017). 117 ILO Committee on Freedom of Association (2012) Report No 363, Case 2602, para 461. See further in the same report the recommendations in paras 508 and 1085–1087. 118 http://www.europarl.europa.eu/charter/pdf/04473_en.pdf (date last accessed 27 August 2017). 119 See the Preamble to the Charter. 120 For instance, according to the Committee, ‘for the purposes of Article 3§1 of the Charter, all workers, including non-employees, must be covered by health and safety at work regulations . . . [The European Committee of Social Rights] has consistently maintained this interpretation, on the grounds that employed and non-employed workers are normally exposed to the same risks in this area’ (European Committee of Social Rights, Conclusions XVI-2 [Austria], Strasbourg, 2005), 11. 121 Decision of the Irish Competition Authority No. E/04/002 (2004). 122 Irish Congress of Trade Unions v Ireland, Complaint No 123/2016, 10 October 2016, written by John Hendy QC. 123 It is notable that according to the Commission ‘The Pillar takes direct inspiration from the existing wealth of good practices across Europe, and builds on the strong body of law which exists at EU and international level’, and that it ‘reaffirms the rights already present in the EU and in the international legal acquis and complements them to take account of new realities’. See COM(2017) 250 final page 6, emphasis added. 124 Cf. Gilham v Ministry of Justice, Appeal No UKEAT0087/16/LA, see para 6 in particular. 125 Danosa 56. 126 Danosa 49. 127 Allonby para 71. FNV Kunsten para 35. 128 This point was already suggested and explored in B. Bercusson, European Labour Law (London: Butterworths, 1996), at 429. 129 Cf. European Parliament, ‘European Parliament resolution of 14 January 2014 on social protection for all, including self-employed workers’, T7-0014/2014, and European Parliament resolution of 19 January 2017 on a European Pillar of Social Rights; also EESC, ‘Opinion: New Trends in Self-employed Work: The Specific Case of Economically Dependent Self-employed Work’, SOC/344/2010. European Commission, Consultation Document of 26.4.2017—First phase consultation of Social Partners under Art 154 TFEU on a possible action addressing the challenges of access to social protection for people in all forms of employment in the framework of the European Pillar of Social Rights, COM(2017) 2610 final. 130 C. O’Brien, E. Spaventa and J. De Coninck, ‘Comparative Report 2015 The Concept of Worker under Article 45 TFEU and Certain Non-standard Forms of Employment’ (Directorate-General for Employment, Social Affairs and Inclusion, 2016), p 20. 131 Cf. for instance X v Mid Sussex Citizens Advice Bureau and another  UKSC 59. And in the same vein it may emerge as a hurdle to the exclusion of certain workers in the gig economy through the introduction of so called ‘safe harbors’, see M. A. Cherry and A. Aloisi, ‘Dependent Contractors’ in The Gig Economy: A Comparative Approach (Saint Louis U. Legal Studies Research Paper No. 2016-15). Available at SSRN: https://ssrn.com/abstract=2847869 or http://dx.doi.org/10.2139/ssrn.2847869 (date last accessed 27 August 2017). 132 EU Commission, ‘First Preliminary Outline of a European Pillar of Social Rights’ COM(2016) 127 final, at p 5. 133 EU Commission, ‘Proposal for an Interinstitutional Proclamation on the European Pillar of Social Rights’ COM(2017) 251 final, Principle 5(a). 134 Decision (EU) 2016/344 of the European Parliament and of the Council of 9 March 2016 on establishing a European Platform to enhance cooperation in tackling undeclared work,  OJ L65/12. 135 Ibid., para 8 of the Preamble. 136 Ibid., Art 4(a). 137 COM(2017) 2611 final. 138 COM(2017) 2610 final. It is notable that the material scope of the proposal is limited to social security, as encouraged by Principle 12 of the Social Pillar, and to access to employment services and training, but does not include ‘working conditions’, as seemingly envisaged by Principle 5. 139 COM(2017) 2611 final, page 6. 140 Ibid., p 8. 141 The European Parliament expressly called for the Directive ‘to cover all forms of employment and employment relationships’, including casual work and dependent self-employment. Cf. European Parliament resolution of 19 January 2017 on a European Pillar of Social Rights. 142 It is noted that equivalent exclusions are contemplated in a number of other instruments, most notably under Clause 2(2) of Directive 97/81 or Art 5(4) of Directive 2008/104. 143 EU Commission, ‘Proposal for a Directive of the European Parliament and of the Council amending Directive 2002/15/EC on the organisation of the working time of persons performing mobile road transport activities’ COM(2008) 650 final, Art 1; cf. also Opinion of the European Economic and Social Committee on the Proposal for a Directive of the European Parliament and of the Council amending Directive 2002/15/EC on the organisation of the working time of persons performing mobile road transport activities COM(2008) 650 final  OJ C228/78. 144 Ibid. The original proposals also referred to a third criterion, ‘c) who does not have the freedom, individually or through a cooperation between self-employed drivers, to have relations with several customers’, but it is arguable that this criterion ought to be superseded or qualified in light of the practices by online platforms, such as Uber, that are construed to give the impression that individual workers are establishing relations with several customers, whereas it is now increasingly accepted that this element requires close scrutiny c.f. Aslam and Farrar v Uber (Case 2202550/2015) of 28 October 2016. 145 Case C-397/01, Pfeiffer, para 82. 146 Cf. Eurostat lfsa_esgan2 definitions. 147 The Court of Justice is quite adept at identifying labour intensive services and distinguishing them from capital intensive ones, as best exemplified by its case law on the Acquired Rights Directive, see for instance Case C-173/96, Sánchez Hidalgo and Others, paras 31–32 in particular. 148 On the concept of the employer cf. J. Prassl, The Concept of the Employer (Oxford: OUP, 2015). 149 K. D. Ewing, J. Hendy and C. Jones (eds), A Manifesto for Labour Law: Towards a Comprehensive Revision of Workers’ Rights (Liverpool: IER, 2016), at p 35. 150 CGIL, Carta dei diritti universali del lavoro – Nuovo statuto di tutte le lavoratrici e di tutti i lavoratori (2016), Art 1. 151 E. Dockés (ed), Proposition de Code du Travail (Paris: Dalloz, 2017), Arts L. 11-1 to L. 11–18. 152 Opinion of AG Wahl in Case C-413/13, FNV Kunsten, para 51. 153 On the idea of personal work relations cf. M. Freedland and N. Kountouris, The Legal Construction of Personal Work Relations (OUP, 2011). © Industrial Law Society; all rights reserved. For permissions, please e-mail: firstname.lastname@example.org.
Industrial Law Journal – Oxford University Press
Published: Aug 30, 2017
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera