Monday 29 January 2018

Does EU law protect gig economy workers? Tensions in the CJEU’s case law

Dr Maria Tzanakopoulou, Teaching Fellow, King's College London and University College London


The gig economy is on the rise precipitating much discussion about working conditions –from working time to remuneration and from maternity and paternity protection to the all important classification of individuals working in the gig economy pool (see, eg, here and here). The case of King marks the debut of CJEU judgments related to the regulation of business conduct and worker’s rights in the gig economy. Here, the CJEU upholds the right of a self-employed worker to indeterminately carry over entitlements deriving from unexercised paid leave, while it protects Mr King’s right to an effective remedy before the courts. The story continues with the Court’s much debated Uber decision (discussed here), which reportedly blows in the face of businesses and becomes a stepping stone to more comprehensive protection of the gig economy worker (here). Perhaps somewhat surprisingly for those familiar with the Court’s often dismissive approach to labour rights (see, eg, here), the CJEU now seems, at least at first sight, reluctant to leave the gig economy unregulated and to turn its back on gig economy labour forces.

Very briefly, the Court in Uber classifies the company as a provider that ‘offers urban transport services’ [para 38] rather than as a mere intermediary between drivers and clients, as the company itself maintains. The tangible effect of this decision is the subordination of Uber to national regulatory measures. A less visible corollary of the case will be Uber’s increased responsibilities towards its drivers. The case of King, touching as it does upon the crucial matter of paid leave, has a more visible effect on workers. So much so, that the press was quick to present the decision as a breakthrough for gig economy workers at large (here). However, the extent to which the case makes headway in bringing gig economy workers as a whole under the protective ambit of employment rights is a matter that is not entirely straightforward.


Mr King worked as a salesman for a company installing doors and windows (SWWL) from 1999 until his dismissal, brought into effect on the day of his 65th birthday, in 2012. According to his contract, a self-employed commission-only contract, Mr King was paid on the basis of the sales he concluded. His right to paid leave was unclear, as the contract was silent on that matter. In his Opinion, AG Tanchev reports that Mr King was in the meantime offered an employee contract, which would bring him into the sphere of full-blown worker protection, but he opted for carrying on his work on a self-employed basis. (Mr King later objected that the AG misunderstood the particular incident, but we lack further clarification, as the Court did not consider it necessary to reopen the oral procedure for what it apparently saw as a matter of secondary importance.)

Upon his dismissal in 2012, Mr King brought his case to the Employment Tribunal. Mr King succeeded in his claims that the dismissal was grounded upon discrimination on the basis of age and that he satisfied the definition of ‘worker’ for the purposes of the UK Working Time Regulations, implementing Directive 2003/88 (the ‘working time Directive’. The Employment Tribunal further found that Mr King was entitled to recover the sum of his untaken paid leave for his final year with SWWL, as well as the sum amounting to holiday he had taken from 1999 until 2012 and which had remained unpaid throughout his thirteen years of work.

So much remained undisputed by SWWL. However, the Employment Tribunal’s final finding, namely that Mr King was further entitled to recover the sum for any leave not actually taken during his work with SWWL, was appealed to the Employment Appeal Tribunal (‘EAT’). The latter accepted the appeal and passed the case back on the Employment Tribunal for reconsideration. According to the EAT’s line of argument, King would have to first take (unpaid) leave and subsequently raise a claim related to payment of that leave. This finding was primarily based on Regulation 13(9) of the domestic implementing legislation, which establishes that leave can only be taken in the relevant leave year and, if not, it cannot be replaced by payment in lieu, unless employment has been terminated. Regulation 30 further establishes that ‘[a] worker may present a complaint to an employment tribunal that his employer’ has either not allowed him annual leave or ‘has failed to pay him the whole or any part of any amount due to him’ in respect of annual leave. Logically then, failure to pay shall precede any complaint presented. The time limit for the complaint, laid down in Regulation 30(2)), is set at three months after the claim arises or at whatever period the tribunal considers appropriate where ‘it was not reasonably practicable for the complaint to be presented’. When that complaint is not presented in time then any entitlement, and in this case Mr King’s entitlement, shall be lost.

Mr King was of a different opinion. As failure to take annual leave was a direct result of his employer’s refusal to pay, the relevant rights were carried over from year to year until his termination of employment. His claim was therefore brought in time.

The EAT’s decision was, thus, appealed to the Court of Appeal, which referred five questions, of both a procedural and a substantive nature, to the CJEU. In a nutshell, the Court of Appeal asked whether the ‘use it or lose it’ approach of the Regulations is compatible with the right to an effective remedy. The question targeted the EAT’s interpretation that a worker can bring a complaint only upon taking leave, which the employer refuses to pay. The Court further asked if the right to paid leave carries over beyond the relevant leave year, in cases where non-exercise of that right is caused by the employer’s refusal to pay. If so, how long does that right carry over for?


The CJEU commenced on its assessment by emphasising the social significance of the right to paid leave. The Court was quick to bring the EU Charter of Fundamental Rights into its reasoning, referring in particular to article 31(2), which lays down the right to paid annual leave. The purpose of Directive 2003/88 read in the light of the Charter is to allow the worker to enjoy annual leave under conditions of remuneration comparable to those of working periods: the right to paid leave cannot be ‘subject to any preconditions whatsoever’ [para 33]. If leave itself or remuneration become uncertain as a result of the employer’s conduct, then the right is in jeopardy.

On the basis of the above, the CJEU rejected the EAT’s interpretation of the Regulations as incompatible with the Directive: ‘in the case of a worker in a situation such as that of Mr King, if the national remedies are interpreted as indicated [by the EAT], it is impossible for that worker to invoke, after termination of the employment relationship, a breach of Article 7 of Directive 2003/88 in respect of paid leave due but not taken, in order to receive the allowance referred to in paragraph 2 of that article. A worker such as Mr King would thus be deprived of an effective remedy’ [para 46]. As such, a complaint cannot be exclusively available after the employer has refused to pay for a leave already taken. This is so, despite the fact that the Directive itself is silent on remedies, not least because the Charter enshrines the right to an effective remedy in article 47 [para 41].

With respect to a worker’s ability to carry over paid annual leave rights, the Court first recognised that Mr King did not exercise his rights for reasons beyond his control. Whether or not Mr King was in the meantime offered employee status was irrelevant for the CJEU, which looked at the worker’s status as it ‘existed and persisted’ until retirement, whatever the reason for that status may be [para 50]. According to the Court’s settled case law on absence due to sickness, allowance in lieu should be available to those unable to exercise paid leave rights for reasons beyond their control. However, a carry over limit of fifteen months should be equally acceptable, given that the Court also has regard to ‘the protection of employers faced with the risk that a worker will accumulate periods of absence of too great a length and the difficulties in the organisation of work which such periods might entail’ [para 55]. The question thus came down to whether Mr King’s situation was comparable to absence due to sickness.

The Court here noted that ‘protection of [Mr King’s] employer’s interests does not seem strictly necessary’ [para 59], especially in light of the need for the right to paid leave to be interpreted broadly:

‘It must be noted that the assessment of the right of a worker, such as Mr King, to paid annual leave is not connected to a situation in which his employer was faced with periods of his absence which, as with long-term sickness absence, would have led to difficulties in the organisation of work. On the contrary, the employer was able to benefit, until Mr King retired, from the fact that he did not interrupt his professional activity in its service in order to take paid annual leave’ [para 60].

The Court concluded that ‘an employer that does not allow a worker to exercise his right to paid annual leave must bear the consequences’ [para 63], which in this case amounted to the sum due for all leaves untaken by Mr King over the years.


At the outset, King is a case that gives recognition to a significant employment right and should be welcome on that ground alone. The Court did not shy way from stretching the outer limits of the right to paid leave and fired straight at the employer. In symbolic terms, the case seems equally interesting, with the CJEU’s diction hinting at a robust defence of what the Court calls ‘EU social law’ against the interests of the employer.

Reliance on the Charter as a leg-up in the Court’s broad interpretation of the right is also interesting, albeit, I think, not necessarily decisive for this case. The point of a purposive interpretation, like the one pursued by the Court here, is precisely that it needs not rely on text (see here). The Court could have decided much the same without the Charter’s assistance. That said, the decision is a harsh message to businesses engaging freelancers or generally workers that lack employee status.

Nevertheless, the significance of King for gig economy worker rights at large should not be overestimated. The grand scheme of things suggests that, whereas workers in the gig economy can win the small individual battles before the Court, the CJEU is unwilling to open up the way to wide-ranging protection.

To begin with, Mr King’s worker status was not under dispute, thus disburdening the CJEU of examination of the personal scope of the right to paid leave. In other words, the Court did not discuss applicability of that right to gig economy workers as such. As the AG notes in his Opinion, the case was effectively concerned with the essence, rather than the existence, of the right [para 30]. In light of this, it could be argued that the Court’s primary concern was to interpret the right and only at a secondary level to protect the gig economy worker. Certainly, the specificities of the case, and of Mr King’s situation, render this a consequential decision for many workers in the gig economy. Nevertheless, the extent to which the relevant legislation will cover a gig economy worker remains contingent upon each individual worker’s exact employment status.

What is perhaps more important is that, once King is compared to the Court’s approach to collective labour rights in the gig economy, the picture becomes even less promising.

Indeed, the social dimension of employment rights emphasised in King appears to be neglected in cases dealing with collective labour rights of the gig economy worker. It is now established case law of the Court that collective bodies representing self-employed workers pursue an economic activity and are therefore caught by the restrictive framework of competition law rules (see Pavlov). This is so, the Court has aegued, despite these bodies’ pursuit of a social objective [Pavlov, para 118]. The importance of Pavlov is put down not to the specific facts of the case but to the Court’s refusal to invest a body representing self-employed workers with equal protection as, eg, collective bargaining agreements between employers and employees (see, eg, Albany).

On the contrary, the Court has noted that ‘the Treaty contains no provisions (…) encouraging the members of the liberal professions to conclude collective agreements with a view to improving their terms of employment and working conditions (…)’ [para 69]. Worse even, in the more recent FNW Kunsten, the Court declared in unequivocal terms that ‘in so far as an organisation representing workers carries out negotiations acting in the name, and on behalf, of those self-employed persons who are its members, it does not act as a trade union association and therefore as a social partner, but, in reality, acts as an association of undertakings’ [para 28] and therefore falls within the scope of competition law rules.

All things considered, the Court appears to be adopting double standards. It is ready to recognise the social significance of the right to paid leave of an individual freelancer, yet it stops short of shielding the right of the self-employed worker to collective bargaining.

One way forward, then, would be for the Court to recognise individual rights piecemeal, as in King. Another way would be for individuals to rely on the courts, European or domestic, for an inclusive approach to self-employed workers, as in Aslam. Here, the worker status of Uber drivers, recognised by the Employment Tribunal and upheld by the EAT, placed these individuals within the protective ambit of the National Minimum Wage Act and the Working Time Regulations. Like Mr King, the status of Uber drivers as workers for the purposes of these pieces of legislation is now beyond dispute. It is, however, doubtful whether any of the above would be able to negotiate their salary or pension through collective bargaining free from the burden of EU competition law rules. A third way forward would perhaps be for the CJEU to recognise the social significance of bargaining rights and to guarantee them for gig economy workers. Only then, I think, will it be justified to say that the war was won.

King wins a battle but the war is ongoing: when an employment right can be defended individually before the court, but not collectively in the field –even where that right is upheld– I shall remain hesitant to ring victory bell for gig economy workers.

Barnard & Peers: chapter 20

Photo credit:


  1. This comment has been removed by a blog administrator.

  2. This comment has been removed by a blog administrator.

  3. This comment has been removed by a blog administrator.