Balancing Competition and Worker Protection: The EU’s Platform Work Package in a Social Market Economy

2.1 Moving away from liberal market ideology

In 1939, Hayek’s The Economic Conditions of Interstate Federalism argued for a unified European Union based on classical liberalism.⁸ Hayek expected that a multinational character of an interstate federation would in fact stimulate market liberalism.⁹ He predicted that such a federation would allow ‘the free movements of men and capital between the states of the federation’.¹⁰ In order not ‘to drive labour and capital elsewhere’, member states would shun all sorts of taxation.¹¹ In his famous article The Asymmetry of European Integration, Scharpf concludes, after analysing the evolution of European integration, that Hayek’s prediction came true.¹² However, it took longer than expected. Hayek’s assumption that political integration would precede the creation of a common market occurred in reverse. The European Economic Community (EEC), established by the Treaty of Rome in 1957, was primarily a common market rather than a political project.¹³ This allowed member states to retain control over when trade would be liberalised, ensuring no legislation could remove economic boundaries without their agreement. Scharpf notes that this enabled states to balance economic liberalisation with their welfare policies.¹⁴ In the EEC, the economic integration would not exceed the limits of what John Ruggie described as the ‘embedded liberalism’ of the post-war world economy.¹⁵ That is to say, a regime in which markets would be allowed to expand within politically defined limits that would not undermine the precondition of social cohesion and stability at the national level.¹⁶ As Hayek predicted, the balance between markets and stability at the national level changed as the process of integration took longer. Scharpf argued that the EU over time has been very successful in its programme of negative integration through case law (which has increased liberalisation of the internal market), but it is much less well equipped to adopt measures of positive integration such as common welfare policy.¹⁷ Especially since the early 1980s, economic integration has accelerated and intensified and the liberal transformation which Hayek had expected has indeed been taking place.¹⁸ This resulted in the emergence of a liberal market economy on a European scale.¹⁹ Scharpf attributes this shift to the structural dynamics of European integration itself. He highlights an asymmetry between the ECJ’s power to dismantle barriers, driving negative integration and the limited capacity for political decision-making, constrained by high-consensus requirements and the diversity of national interests. This makes it easier to remove trade barriers and improve market access than to implement harmonised social policies that uphold welfare state principles. Scharpf concludes that the EU’s history of integration illustrates why the EU cannot become a social market economy: it lacks the institutional capacity to balance economic liberalisation with social protection.

The evolution of European competition law mirrors these broader developments. Rooted partly in ordo-liberalism, European competition law was initially shaped by the belief that markets must be regulated to prevent the accumulation of market power that threatens both economic efficiency and societal fairness.²⁰ Ordo-liberals believed that, while markets can produce optimal outcomes, they do not form spontaneously. Instead, markets need proper regulation.²¹ Ordo-liberals argued that the state must structure and regulate the market, although such intervention must be guided by constitutional norms to prevent overreach. As a result, competition law was seen as not only an instrument to maintain market efficiency but also to ensure societal fairness. At the beginning, European competition law operated within the framework of embedded liberalism, supporting both market freedom and active state intervention to promote social goals like full employment.²² This led to a framework where competition law’s twin goals were to protect economic freedom and create an internal market, which became the cornerstone of European integration.²³ However, since the 1990s, during the liberal transformation which Hayek had predicted, European competition law underwent significant reforms, driven by the economisation and modernisation of competition policy.²⁴ This shift placed consumer welfare and economic efficiencies at the centre of competition law, aligning with the broader liberal market economy ideology. While this focus on economic efficiencies made competition law more predictable and rational, critics argue that it has led to tensions with important societal concerns such as fairness, solidarity or other non-economic issues.²⁵

As outlined in the Introduction, these tensions are particularly pronounced in the context of platform workers whose struggles exemplify the effects of competition law’s evolution. Driven by competition for work, platform workers often face a race to the bottom in terms of wages and conditions. Efforts to strengthen their collective position, whether through cooperation or trade unions, have been prohibited under competition law, as such actions are considered to be violations of the cartel prohibition under Article 101 TFEU. This reflects the broader liberal market economy ideology that has shaped competition law since the 1990s, with its prioritisation of market competition over worker solidarity. The emphasis on economic efficiency and consumer welfare has entrenched competition as a core value, in this case at the expense of social protection. The precarious situation of platform workers arises from a system that systematically favours market forces, highlighting the need to reassess the role of competition law. This prohibition on collective action exacerbates the power imbalance between workers and digital labour platforms, increasing the dependency of individual workers on these platforms. As the EU seeks to align itself more closely with the principles of a social market economy, it aims to address these tensions by rebalancing competition law to ensure that social objectives, such as worker protection, are given greater prominence.

2.2 Social market economy: clashing values?

In contrast to the liberal market economy that emerged in Europe the framework of a social market economy, which is formally enshrined in the Lisbon Treaty, represents the objective to strengthen the social dimension within the European Union. This ties in with the broader social objectives of the EU, which have been gradually developed since the 1980s, including the promotion of the social dimension, the European Social Model in the 1990s and, more recently (2017), the European Pillar of Social Rights.

The concept of the social market economy resonates strongly with the original roots of ordo-liberalism and embedded liberalism, as previously explored.²⁶ It seeks to ensure that economic growth goes hand in hand with social cohesion. The social market economy embodies the idea that markets require structure and regulation to avoid undermining social stability, aligning with the belief that economic freedom must be accompanied by measures that protect individuals from the adverse effects of market dynamics.

Of course, balancing these two, ‘the social’ and ‘the market’, is easier written down than actually done in practice. It is quite conceivable that both components may come into conflict with each other.²⁷ This friction becomes evident, for example, in the realm of the market and competition law, especially in light of the economic transformation discussed in Section 2.1. Coleman, in his description of a perfectly competitive market, highlights how competition is driven by individualism and self-interest, which fosters rivalry rather than cooperation.²⁸ The value of competition, as the objective of competition law, could stand in opposition to values like solidarity, as the objective of social policy. This paradigm resonates with the liberal market economy, where agents are motivated purely by maximising their own utility, a framework that has moral limits, as Hussain points out. He suggests that competition can sometimes pit people against each other, leading to detrimental consequences, especially for vulnerable workers, such as self-employed platform workers. A liberal democracy can create institutions which ‘define a framework in which people have to struggle against each other’.²⁹ Platform workers exemplify this dynamic. They compete with one another for work, driving conditions downward, while cooperation through trade unions has historically been restricted under competition law. In this clash of values – market competition versus worker solidarity – the former has consistently taken precedence due to the political realities that have shaped competition law. If the EU seeks to establish a different political model, this would imply that competition law might need to adapt to align with evolving priorities.

The EU’s recent efforts to establish a social market economy signal a shift in political priorities. In 2021, the European Commission launched the Platform Work Package under the European Pillar of Social Rights Action Plan to provide platform workers with better protections, including collective bargaining rights. Before examining the specifics of this package, the next section delves into the legal challenges which platform workers face and the historical treatment of collective bargaining under EU competition law. This analysis sets the stage for evaluating whether these reforms can effectively address the tensions identified in this Section.

3.1 Self-employed workers and workers under EU competition law

Competition law in general, and thus the cartel prohibition under Article 101 TFEU, applies to undertakings, which are broadly defined as ‘every entity engaged in an economic activity regardless of the legal status of the entity and the way in which it is financed’. ³⁰ This definition extends to individuals offering goods or services on the market while assuming the financial risks associated with their performance. ³¹ However, under EU law, a worker is not considered an undertaking, as they do not engage in autonomous economic activity. ³² Workers do not offer goods or services independently on the market, nor do they bear the financial risks associated with such activities.³³

The Becu case shed light on the distinction between worker and undertaking.³⁴ The ECJ was presented with a case involving the organisation of dockworkers in Belgium. The Belgian government had enacted a law prohibiting employers from hiring non-registered workers. However, some individuals, including Becu and others, chose to ignore this law in order to reduce their labour costs. As a result, they faced criminal charges. During the criminal proceedings, the defendants argued that the national law in question violated EU competition rules. The central issue at hand was whether workers could be considered ‘undertakings’ as individuals who sell their labour in the market, thereby competing with other workers. In its ruling, the ECJ directly addressed this question and concluded that workers did not fall under the definition of undertakings. Therefore, they could not be subject to EU competition rules.

In the same judgment, when deciding who qualifies as a ‘worker’, the ECJ emphasised the element of control which employers have over workers. The ECJ ruled that the dockwockers’ employment relationship with the organisations for which they work is defined by the fact that they perform work for and under the direction of each undertaking. Consequently, they are considered ‘workers’ according to Article 45 TFEU, as interpreted by the ECJ’s case law. During this relationship, dockworkers are integrated into the concerned undertakings and form an economic unit with each of them, rather than being separate undertakings themselves.³⁵ In doing so, the ECJ’s decision aligned the concept of ‘employee’ (which is not an EU concept) with the broader definition of ‘worker’ in the sense of Article 45 TFEU, the ‘free movement of workers’ Treaty provision.³⁶ The concept of employee is derived from national legal systems and varies across jurisdictions. The EU does not have a uniform definition of ‘employee’; instead, it uses the broader term ‘worker’ in its case law and legal provisions.

This case laid the groundwork for what would later become a more comprehensive labour exemption, first fully articulated in the Albany judgment.³⁷ While Becu clarified that workers acting within the structure of an employment relationship could not be considered to be undertakings, Albany further developed this by establishing a formal exemption from competition law for collective agreements. The Albany exception provides immunity from EU competition rules for agreements concluded between associations of workers (labour unions) and employers, provided two conditions are met: (1) the agreement is entered into within the framework of collective bargaining between employers and employees, and (2) it directly contributes to improving employment and working conditions.³⁸ The ECJ acknowledged in Albany that restrictions of competition are inherent in such agreements but stressed that social policy objectives would be undermined if collective bargaining were subject to competition law.³⁹ This highlights the discussed conflictual nature of the EU’s dual objectives of promoting social policy and safeguarding competition law. This balancing act is at the core of the broader challenges that arise when applying competition law to collective labour agreements.

Exercising a liberal profession (e.g. lawyer, accountant, doctor), on the contrary, has usually been considered as an undertaking and therefore falling under competition rules if there is no relationship of employment.⁴⁰ The first cases which dealt with the tension between self-employment and competition law date back to the 1970s. It has become increasingly clear that national employment status is taken into account by the ECJ in determining whether competition law rules apply.⁴¹ In this way, national labour law and European competition law were gradually turned into each other’s opposites. On the one hand, labour law distinguishes between ‘employees’, ‘employers’ and ‘self-employed’. Various factors, such as the level of subordination and integration within a company, are considered to determine the appropriate classification for a person. The central objective of labour law revolves around ensuring fairness and addressing power imbalances that may arise between employers and employees.⁴² In this regard, engaging in collective bargaining processes is justified on the grounds of promoting fairness and combating subordination. On the other hand, competition law applies to ‘undertakings’ and case law has formulated a labour exemption. The aim of competition law in this light is the prevention of collective negotiations between undertakings that can lead to distorting markets and harming business and consumers.⁴³

The ECJ considers the national employment status when determining whether EU competition law applies. An individual in an employment relationship is not considered an undertaking. If a person is not in an employment relationship, e.g. is self-employed, this person is, for the purpose of competition law rules, an undertaking. As a result, they would not have the right to engage in collective bargaining.

3.2 Workers in today’s economy

Section 3.1 analyses how the relationship between labour law and competition law prevents self-employed workers from being able to collectively bargain.⁴⁴ This relationship, described as ‘antagonistic’ by some authors, does not align any longer with the dynamic nature of the evolving labour market.⁴⁵ Labour law is increasingly recognising the importance of including self-employed workers in collective bargaining processes.⁴⁶ This is because, in recent years, more and more economic transactions in the labour market involve self-employed individuals working with clients or customers, rather than being in traditional employee-employer relationships. An example of this is of course the platform worker who operates as an independent contractor for a digital platform. The continued classification of these workers as undertakings perpetuates a system where market competition is prioritised over worker solidarity, illustrating the ongoing tension between these competing values.

The distinction between employee and employer/self-employed has long functioned as an indication of power distribution. The employee could be in a weaker position compared to the powerful employer and therefore needed protection through labour law. A self-employed worker has been often associated with the employer’s side: protection was not necessary.⁴⁷ However, the concept of a strong self-employed worker has become outdated in light of recent developments on the labour market. In recent decades, the number of self-employed individuals in the EU has increased.⁴⁸ These so-called ‘new self-employed’ are not as independent and financially stable as the traditional self-employed.⁴⁹ This demonstrates that the distinction between employees on the one side and employers/self-employed on the other side is no longer adequate for the protection that various parties seek. A group of ‘new self-employed workers’ has emerged who require some form of protection, such as a strong bargaining position.

In 2014, the ECJ addressed how to reconcile labour and competition law in the FNV/Kiem case.⁵⁰ The ECJ examined collective agreements by trade unions for subordinate workers that also included minimum labour costs for self-employed workers. While Becu clarified the distinction between workers and undertakings, and Albany confirmed that collective agreements between employers and employees fall outside the scope of competition law, FNV/Kiem tackled the grey area concerning agreements for self-employed individuals.⁵¹ This issue arose after the Dutch competition authority’s 2007 reflection document, which stated that agreements for self-employed workers were not exempt from competition law. This had a significant impact on collective bargaining negotiations in sectors like arts, information and media, where the trade unions sought to establish minimum tariffs for self-employed workers.⁵²

The ECJ’s ruling primarily focused on the concept of undertaking.⁵³ The ECJ held that ‘in so far as an organisation representing workers carries out negotiations acting in the name, and on behalf, of those self-employed persons (…), it does not act as a trade union association (…), but, in reality, acts as an association of undertakings’,⁵⁴ and is therefore also exposed to the full application of EU competition law rules. An exception to these rules, the ECJ said in FNV/Kiem, is only possible ‘if the service providers, in the name and on behalf of whom the trade union negotiated, are in fact “false self-employed”, that is to say, service providers in a situation comparable to that of employees’.⁵⁵

3.3 Protecting the ‘false’ self-employed workers

The ECJ made it clear that a service provider can lose his status of an undertaking, ‘if he does not determine independently his own conduct on the market, but is entirely dependent on his principal’.⁵⁶ The ECJ based this on the fact that the service provider does not bear any of the financial or commercial risks arising out of the principal’s activity and operates as an auxiliary within the principal’s undertaking. This assumption aligns with the approach taken in the application of Article 101 TFEU to agency agreements, where the absence of significant financial or commercial risks borne by the agent may exempt such agreements from the scope of competition law. The ECJ stated that it is settled case law ‘that the essential feature of that relationship is that for a certain period of time one person performs services for and under the direction of another person in return for which he receives remuneration’.⁵⁷ That means that the classification of a self-employed person under national law does not prevent this person from being classified as a worker within the meaning of EU law.⁵⁸

The ECJ seems not to acknowledge that a genuine self-employed person could also have the right to collective bargaining. It aims to assist the new group of self-employed individuals by squeezing them into the already existing dynamic between labour law and competition law, as outlined in Section 3.1. If a self-employed individual meets the objective criteria mentioned by the ECJ, EU law will regard them as a worker. Consequently, they will not be classified as an undertaking under EU provisions and will fall outside the scope of Article 101 TFEU. By doing this, the ECJ aims to safeguard false self-employed and ensure their protection under labour law, specifically the right to collective bargaining. In this way not only are traditional employees allowed to engage in collective bargaining, but now also the false self-employed individuals can do so. The genuine self-employed worker, who does not meet these objective criteria, is still hindered by competition law rules. The ECJ seems to refer to the concept of false self-employment when describing a situation in which a worker, from the perspective of EU law, is not legitimately engaged as a self-employed individual.⁵⁹ A false self-employed worker is, therefore, not so much an intermediate category as part of the worker category.⁶⁰

The notion that a false self-employed worker is equivalent to a worker under EU law is also reflected in the Yodel ruling from 2020.⁶¹ This case centred on whether an English courier with a service agreement could be considered as a worker under the Working Time Directive. The ECJ stated that it was up to the referring national court to determine if the Yodel courier should be classified as a worker, as it requires an assessment of all the circumstances. However, the ECJ does indicate that the classification of an independent contractor under national law does not prevent that person from being classified as a worker under EU law. To designate a worker in the EU context, the ECJ refers to the aspects that point to false self-employment as outlined in the FNV/Kiem case.⁶²

3.4 Beyond the binary

The ECJ attempted to address the issue but only provided a solution for false self-employed individuals. This approach has not provided a seamless continuum of legal protection for the diverse spectrum of labour arrangements.⁶³ You are either a worker (including the false self-employed) with the possibility for collective bargaining under Article 101 TFEU, or you are an undertaking (including the genuine self-employed) that does not meet the objective criteria and cannot invoke the collective agreement exception. To adequately address the problem of the precarious self-employed (platform) worker, it is evident that the legislative framework should be altered. Platform workers are in a precarious position because they are classified as self-employed contractors, which means they lack the protections and benefits afforded to traditional employees, such as insurance, paid leave and pensions. Relying solely on employment status as a gateway for collective bargaining is insufficient and does not take into account the vulnerability of certain workers outside regular employment.

Recognising this gap, the European Commission introduced in 2021 the Platform Work Package as part of the European Pillar of Social Rights Action Plan. This Package, aimed at addressing the precariousness of platform work, seeks to extend collective bargaining rights to some self-employed workers. The Package included a proposal for a Directive on improving the working conditions in platform work and Guidelines that should help to clarify the relationship between EU competition law and self-employed people.⁶⁴

In doing so, it marks a shift in the EU’s approach to labour law, as it moves away from the rigid distinction between worker and undertaking. Section 4 will explore the contents of the Platform Work Package in detail, analysing how the European Commission aims to reconcile the competing objectives of market competition and worker protection.

4.1 Directive addresses the false self-employed workers

In the proposal for a Directive, the European Commission followed the path that the ECJ had taken with the FNV/Kiem case. The Commission’s aim is to address false self-employment: platform workers should have the legal employment status that corresponds to their actual working arrangement. The proposal thus provides a framework that should reduce the uncertainty surrounding the status of platform work.⁶⁵

The Directive included measures to correctly determine the employment status of people working through digital labour platforms.⁶⁶ Nine out of ten platforms active in the EU currently are estimated to classify people working through them as self-employed.⁶⁷ The European Commission estimates that, among the 28 million people who work through digital labour platforms, most people are genuinely self-employed. However, there may be up to 5.5 million people who are false self-employed.⁶⁸ According to the settled case law (such as FNV/Kiem) of the ECJ, the essential feature of the employment relationship is that ‘for a certain period of time one person performs services for and under the direction of another person in return for which he receives remuneration’. The Commission indicated that it is indeed possible for a platform worker to prove through court cases that the actual contractual description of their status is false self-empoloyed. As research points out, in most cases, the rulings have confirmed that self-employed workers were misclassified and should actually be categorised as workers.⁶⁹ However, this can be considered as quite a struggle time- and money-wise, especially for workers in a weak position, such as low-wage earners, young workers or those with a migrant background. That is why the Commission proposed in December 2021 that the contractual relationship between a digital labour platform must be legally presumed to be an employment relationship when the platform ‘controls the performance of work’.⁷⁰ In relation to the employment status, the Directive put forward the possibility of rebutting the legal presumption.⁷¹ If the digital labour platform intends to do so, the burden of proof is on the platform. The platform controls the performance of work if at least two of five criteria are met, such as the platform determining the remuneration and imposing rules regarding work appearance.⁷²

The publication of the European Commission’s Directive proposal initiated a lengthy legislative process. The European Parliament established its position on the proposal in February 2023 after intense negotiations.⁷³ Subsequently, EU ministers in the European Council reached an agreement on the proposal in June 2023.⁷⁴ The European Parliament aimed to strengthen the position of platform workers as much as possible and has modified the proposed Directive to legally presume, in principle, an employment relationship between a digital labour platform and a platform worker.⁷⁵ The presumption is a given and does not depend on meeting specific criteria, unlike the Commission’s proposal. It allows platforms the opportunity to challenge the presumption before any reclassification decision. Conversely, the Council’s proposal offered more flexibility for digital platforms to categorise workers as self-employed. While it employs similar criteria to the Commission, a key distinction is the requirement to satisfy three criteria (rather than two, as proposed by the Commission) for an employment relationship, granting digital labour platforms greater leeway to challenge the presumption.⁷⁶ The final adopted Directive (in March 2024) is significantly diluted in comparison to the original proposal. Article 4 mandates that member states must implement procedures to correctly determine the employment status of platform workers. Article 5 states that an employment relationship should be legally presumed when evidence of control and direction by the platform is found. The Directive leaves it to individual member states to define what these evidences are according to their national laws.

The Directive aligns with the direction set by the ECJ in the FNV/Kiem case and aims to improve the working conditions of these platform workers by reclassifying them as employees. The issue of the restriction on collective bargaining for self-employed (platform) workers imposed by the cartel prohibition is no longer relevant for this group since they fall under labour law rights. The implementation of the Directive will lead to an expansion of the group of employees. The Directive will ensure that false self-employed workers have the right to collectively negotiate through the reclassification of a traditional employment relationship.

4.2 Guidelines: providing a broad spectrum of protection

The central issue facing the self-employed worker who seeks the opportunity for collective bargaining lies in the interaction between competition law and labour law, which does not accommodate this need. For bargaining purposes, an employment status is essential. The recently adopted Directive will provide more workers with this opportunity, but it also intensifies the distinction between employees and self-employed individuals. As a result, some self-employed workers will be reclassified, thereby gaining employment status. This adjustment resolves the issue for these individuals but does not tackle the deeper problem. Through the published Guidelines, the Commission aims to enhance the protections for genuine self-employed workers against the barriers of competition law. The Commission identifies two groups of genuine self-employed workers who need protection: (1) self-employed workers comparable to workers, and (2) self-employed workers in a weak bargaining position.

Before this section proceeds to consider these two groups and the contents of the Guidelines, it first discusses this instrument itself and a related concern regarding its effectiveness. The Commission aims to develop policy in which it outlines how EU law (regarding supervision and enforcement of competition rules) should be interpreted. These Guidelines represent a form of soft law issued by the Commission.⁷⁷ The reason for the Commission’s choice might have a political motive. It saves time, as there is no need for a lengthy legislative process (which was necessary for the Directive). As a result, this form of soft law enables the Commission to quickly and explicitly explain how it will act in certain situations. The Commission ‘clarifies’ what various workers can expect, aiming to reduce legal uncertainty.⁷⁸ A drawback of Guidelines is that they are non-binding, which means that member states can choose to deviate from the course set by the European Commission. Member states are free to pursue their own path, and the effectiveness of the Guidelines largely depends on the persuasive force of the arguments put forward.⁷⁹ There is no way to enforce these Guidelines with national competition authorities. Furthermore, it is important to note that the ECJ is not bound by this form of soft law.⁸⁰ What if a case involving self-employed platform workers’ negotiations were to reach the ECJ? These cases could reach the ECJ through various routes: direct appeals against Commission decisions, preliminary references, or legal actions brought by private entities challenging such agreements on competition law grounds. When such cases come before the ECJ, it is tasked with interpreting EU competition law as laid out in the TFEU. The soft law Guidelines of the Commission are explicitly not part of this. Furthermore, as described in this article, the ECJ has followed an approach to granting collective bargaining rights based on the binary relationship between a worker and an undertaking. The ECJ could choose an approach in line with the Commission, but there is currently no certainty for self-employed workers in this regard. This uncertainty, stemming from the soft law instrument, may lead stakeholders to be hesitant about entering into collective agreements.⁸¹

4.3 Self-employed persons comparable to workers fall outside the scope of Article 101 TFEU

The Guidelines address the permissibility of self-employed individuals to engage in collective bargaining. In contrast to the Directive, the Guidelines focus on the challenges faced by self-employed individuals in the broader context of employment, rather than exclusively addressing digital labour platforms. They aim to address the vulnerable position of self-employed persons in the overall labour market. If a genuine self-employed worker is ‘comparable’ to a worker, this self--employed person is free to engage in collective bargaining. In such cases, a collective agreement falls outside the scope of Article 101 TFEU.⁸² The Commission considers that the following categories of self-employed persons are in a situation comparable to that of workers and that collective agreements concluded by them therefore fall outside the scope of Article 101 TFEU:

1. Economically dependent self-employed persons who provide their services exclusively or predominantly to one counterparty or who are likely to be in a situation of economic dependence on that counterparty. The Guidelines consider that a self-employed worker is in a situation of economic dependence where the person earns at least 50% of her or his total annual work-related income from a single counterparty.

2. Self-employed persons working side-by-side with workers who perform the same or similar tasks for the same counterparty and do not bear the commercial risks of the counterparty’s activity or enjoy any independence regarding the performance of the economic activity concerned.

3. Self-employed persons working through digital labour platforms. This final category is of course most relevant to the situation of platform workers. The Guidelines ensure that even if a platform worker is not a false self-employed person (see the Directive), he will fall outside the scope of Article 101 TFEU. This category does raise the question of the definition of a digital labour platform. According to the Commission, a digital labour platform is any natural or legal person meeting the following requirements: (a) it is provided, at least in part, at a distance through electronic means, (b) it is provided at the request of a recipient of the service, (c) it involves, as a necessary and essential component, the organisation of work performed by individuals, irrespective of whether that work is performed online or in a certain location. Criterion (b) for example seems to exclude content marketplaces such as YouTube. This criterion implies a specific workflow with customer requests, worker responses, job completion and payment. However, content creators on these platforms often generate content without explicit requests, and any requests are informal, such as viewer suggestions in comments.⁸³ Criterion (c) can also raise some questions. A mere digital bulletin board function, like listing available workers for third-party contact, does not meet the criteria for a digital labour platform under the Guidelines. But what if a platform provides a digital bulletin board and an extra service for payment processing.⁸⁴ The Guidelines lack a precise definition of what qualifies as an ‘essential component’ and the legal interpretation in the future will likely provide further clarity on when a platform fulfils such a role in work organisation.

The fact that the Commission indicates that negotiations between workers in all these three categories do not, in fact, meet the conditions of Article 101 TFEU is significant. While there could be multiple underlying reasons for this decision, two possible reasons stand out. It is also important to recognise that political considerations likely play a role in shaping the Commission’s approach, which should not be overlooked.

The first reason could be that the nature of the negotiations does not aim to distort competition. This would build on the Albany case, which shows that some agreements between undertakings can, in fact, fall outside the scope of competition rules if they are derived from social dialogue and aim to directly improve working conditions. In this case, a collective agreement was made between organisations representing employers and organisations representing workers, the former being undertakings. In Albany, the ECJ clearly took the view that it was ‘beyond question that certain restrictions of competition are inherent in collective agreements between organisations representing employers and workers’.⁸⁵ However, it was also willing to concede that ‘the social policy objectives pursued by such agreements would be seriously undermined if management and labour were subject to [EU competition rules] when seeking jointly to adopt measures to improve conditions of work and employment’. This approach was partly based on the understanding that the ‘nature and purpose’⁸⁶ of the agreement was that of ‘improving (…) working conditions, namely (…) remuneration’.⁸⁷ Consequently, the agreement at issue did not ‘by reason of its nature and purpose’, fall within the scope of EU competition rules.⁸⁸ This raises the question: what about horizontal agreements between self-employed workers? ⁸⁹ Most cases have shown that, prior to the publication of the draft Guidelines, such agreements were indeed in conflict with competition law provisions. In the Pavlov case for example, the ECJ explicitly held that an exception cannot be applied to an agreement which is not concluded between employers and employees.⁹⁰ Therefore, the self-employed Dutch medical doctors in this case could not rely on the Albany exception, so their agreement did not ‘by reason of its nature or purpose’ fall outside the scope of Article 101. The ECJ emphasised 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…’.⁹¹ However, there are some examples in which self-employed persons are constituted as undertakings and competition law allows those persons to come up with a horizontal agreement. In the Wouters case, for example, the ECJ had to consider the legitimacy of a decision by the Dutch Bar Association to disallow the creation of joint ventures between lawyers and accountants. The ECJ agreed that this ban could be justified having regard to domestic policy considerations relating to the provision of legal services even if other member states did not prevent the operation of such joint ventures. Here the effective provision of legal services justified a restriction of competition. As held by the ECJ in Wouters, not all agreements that distort competition necessarily violate Article 101(1) TFEU.⁹² The cases Albany and Wouters demonstrate that it is possible for agreements between undertakings to fall outside the scope of Article 101 TFEU if distorting competition is an inherent result of pursuing relevant objectives, such as social or policy goals. It could be that this is also the reason why negotiations by self-employed workers, comparable to workers, fall outside the scope of Article 101 TFEU. The Guidelines clarify that agreements between certain self-employed workers can distort competition but also serve the social policy objectives of the European Union.⁹³

The second reason could be related to the undertaking criterion. This group consists of genuine self-employed individuals who are comparable to workers. The Commission points out that service providers can lose their status as independent traders (undertakings) if they do not independently determine their conduct in the market but are entirely dependent on their principals, as they do not bear financial or commercial risks and work as if they were employees integrated into the client’s company.⁹⁴ This would mean that the genuine self-employed person, from a competition law perspective, would no longer be considered an undertaking because he forms one economic entity with the client.⁹⁵ This does not mean that there is an immediate employment relationship. This was also evident in the Uber Spain case of the ECJ (2017).⁹⁶ This case dealt with the legal classification of Uber’s service. Uber argued that it merely offered a technical platform, while the ECJ ruled that Uber provided transportation services. The central issue was not whether the drivers were self-employed or should be classified as employees. In Advocate General Szpunar’s opinion, he noted that, in his view, Uber and its drivers qualify as an economic entity (from a competition law perspective).⁹⁷ However, he pointed out that forming an economic entity does not automatically mean that the drivers are employees.

It is difficult to determine if either of these two reasons influenced the Commission, but in any case, this can be interpreted as a cautious departure by the Commission from the binary distinction between workers and undertakings that determines whether an entity is eligible for collective bargaining. Even without reclassified employment status (labour law), a worker can fall outside the scope of Article 101 TFEU (competition law).

4.4 Self-employed workers in a weak bargaining position fall within the scope of Article 101 TFEU

The category ‘comparable to a worker’ ensures that certain types of genuine self-employed workers can collectively bargain. As mentioned, this represents a slight shift in the dichotomy of worker versus undertaking that determined collective bargaining rights. It is a minor change because the second category still mostly reflects the paradigm of the employment or subordinate relationship.⁹⁸ While there is room for genuine self-employed workers to negotiate, it is only for those who are comparable to a worker. That is why the category of self-employed workers in a weak bargaining position, identified by the Commission, is interesting. For this category, the Commission decided ‘not to enforce’ competition law.⁹⁹ In some situations, self-employed persons who are not in a situation comparable to that of workers may nevertheless be in a weak bargaining position vis-à-vis their counterparties and therefore may be unable to significantly influence their working conditions.¹⁰⁰ Self-employed persons who deal with counterparties that have a certain level of economic strength, and hence buyer power, may have insufficient bargaining power to influence their working conditions. In that case, collective agreements can be a legitimate means to correct the imbalance in bargaining power between the two sides. Therefore, the Commission proposes that to the extent that collective agreements aim to correct a clear imbalance in the bargaining power of self-employed persons relative to their counterparties and are intended to improve working conditions, the Commission will not intervene. Such an agreement does fall within the scope of Article 101 (1) TFEU. An imbalance in bargaining power will be considered to exist ‘at least’ where self-employed workers negotiate with a counterparty which represents the whole sector or industry or where self-employed persons negotiate with a counterparty whose annual aggregate turnover exceeds EUR 2 million or whose staff headcount is equal to or more than 10 persons.¹⁰¹

This means that the European Commission goes further than the approach previously developed by the ECJ.¹⁰² Even if both conditions of Article 101 TFEU are met, there is room for a genuine self-employed worker to negotiate. With this category and the second category of self-employed workers in a situation comparable to workers, the European Commission creates opportunities alongside the worker (employee) and undertaking dichotomy.

4.5 Decoupling the right to collective bargaining from a worker’s labour law status

The Platform Work Package introduced by the European Commission reshapes the relationship between labour law and competition law. Section 3 showed how the traditional binary distinction between an employee and an undertaking determined the potential for collective bargaining. A worker either fell under labour law, avoiding competition law concerns, or a self-employed worker was treated as an undertaking under competition law, making them ineligible for collective bargaining. Labour law status was the gateway to bargaining rights. Despite the European Court of Justice’s attempts to protect new self-employed individuals, it still clung to this dichotomy. That raised the question how genuine self-employed workers were protected by collective bargaining rights. Sections 4.3 and 4.4 show that the European Commission is breaking away from this binary approach. The position of genuine self-employed workers is decisive, not the presence of an employment relationship. Consequently, labour law and competition law are no longer in opposition. The classification of a self-employed individual as an undertaking no longer affects their right to engage in collective bargaining.

Figure 1 illustrates the protection of the bargaining position of the various groups discussed within the Platform Work Package. On the far left is the traditional worker in an employment relationship. The green colour indicates that this category falls outside the scope of Article 101 TFEU, allowing for collective bargaining. Next to the traditional worker is the false self-employed worker. The Directive reclassifies these workers as employees, also placing them outside the scope of Article 101 TFEU. Before the Platform Work Package, self-employed workers were subject to the cartel prohibition, preventing them from engaging in collective bargaining. However, the Commission acknowledges that in today’s labour market, the distinction between certain self-employed workers and employees has become blurred, warranting protection. Self-employed workers who are comparable to workers fall outside the scope of Article 101 TFEU and are also indicated in green. The next category, to the right, consists of self-employed workers in a weak bargaining position. While they are not comparable to employees, they still face a clear imbalance in bargaining power with their counterparties. Agreements between these workers fall within the scope of Article 101 TFEU, but the Commission has stated it will not intervene. This category is indicated in orange. Finally, other self-employed workers remain fully subject to Article 101 TFEU and are not granted the possibility of collective bargaining. This is indicated in red.

Figure 1

5.1 Rebalancing the market in the EU social market economy

The Platform Work Package includes measures to improve worker protections, particularly through the legal presumption of employment for some platform workers and the provision for collective bargaining for some self-employed workers. These initiatives aim to rebalance the market in favour of workers.

Yet, despite these advancements, the process leading up to the adoption of the Directive and the final form it took (discussed in Section 4.1) exposes the EU’s limitations in implementing a truly harmonised social policy for platform work across member states. As outlined in Section 2.1, Scharpf’s analysis highlights the EU’s proficiency in negative integration – removing trade barriers and facilitating market liberalisation – but its struggle to implement positive integration, such as a coordinated welfare policy, remains.¹⁰³ The Directive follows this pattern: while it mandates action, it leaves much of the responsibility for implementing labour protections to individual member states.¹⁰⁴

The Directive that was eventually adopted in March 2024 is a weaker version of the European Commission’s original proposal, as well as the more ambitious position brought forward by the European Parliament. Several member states, which opposed the Commission’s proposals, were able to delay the legislative process and ultimately weaken the Directive by eliminating the originally proposed criteria for determining worker classification.¹⁰⁵ Instead, the final version of the Directive merely requires each member state to establish its own framework for classifying platform workers, without specifying any common criteria. This shift toward a minimum harmonisation approach means that each country is free to set its own standards, leading to potentially wide variations in worker protections across the EU. This creates a decoupling of social policy (which is shaped at the national level) and the market (which operates at the EU level).

This lack of a unified approach risks undermining the goals of the Platform Work Package. It is unlikely that member states that were resistant during the legislative process will now enact national laws that reflect the original, stronger proposals of the Commission. This creates the possibility of uneven implementation, with some countries providing more robust protections for platform workers than others. For example, countries that were opposed to the Directive may introduce frameworks that still leave platform workers in precarious positions, offering them little protection despite the EU’s intention to strengthen their rights. Consequently, the Directive, while promoting social protections, does not fully achieve the social market economy’s goal of balancing economic liberalisation with social justice. This is problematic for platform workers as there is a risk that a national framework will offer insufficient protection, and the different criteria may vary widely. It is also problematic for the EU itself, which strives for a robust social dimension within the EU-wide project but only partially succeeds in enforcing this dimension.

5.2 The limits of employment classification

Even if member states were to implement a translation of the Directive that follows the European Commission’s original structure, it remains uncertain to what extent this would provide effective protection.¹⁰⁶ Recent research shows that platform companies have introduced new forms of false self-employment, resulting in formally classified employees still lacking basic labour rights.¹⁰⁷ This study of Berlin, Lisbon and Paris, found that ride-hailing companies like Uber employ subcontracting and creative compliance tactics to avoid reclassifying drivers as employees. These methods exploit regulatory loopholes, allowing companies to formally comply while bypassing real worker protections.¹⁰⁸ As a result, drivers often earn a below minimum wage, work unpaid overtime, and miss out on paid leave and social security benefits, even when formally classified as employees.¹⁰⁹ Such deceptive models rely on subcontracting, trapping drivers in low-hour contracts or informal arrangements that sidestep social security contributions. This demonstrates the limitations of regulating platform work through employment classification alone.¹¹⁰

Van Doorn, Ferrari, and Graham similarly conclude that reclassification alone is insufficient to protect low-wage platform workers, particularly migrants and minorities.¹¹¹ They argue that platform-mediated employment has largely failed to uphold workers’ rights and dignity, advocating instead for broader protections, redistributive policies, and immigration reforms aimed at social justice.¹¹² Recognising the unique challenges of platform work, the authors call for transnational agreements beyond labour law, such as encompassing consumer protection and competition legislation.¹¹³

5.3 Dependency relationship as the starting point for counterpower

The Guidelines attempt to mitigate the power imbalance between digital platforms and workers by focusing on the dependency relationship that ties platform workers to these platforms. Historically, collective bargaining rights were reserved for employees, whose economic dependence on their employer justified additional legal protections. However, platform work challenges this traditional employer-employee model. The EU’s Platform Work Package attempts to address this dependency by allowing workers to collectively negotiate in cases where they are economically dependent on a platform. The idea is that platform workers, despite being classified as independent contractors and therefore undertakings, may find themselves in a situation similar to that of employees – highly reliant on a single platform for their income. In such cases, the size and influence of the platform over the worker become critical indicators of a dependency relationship, warranting protection (see Sections 4.3 and 4.4). This shifts the balance by granting these workers the ability to form a collective counterforce, aimed at reducing the platform’s control over them.

However, this approach of allowing counterpower is essentially responsive. It strengthens the worker’s position by granting exemptions to competition law, allowing for collective bargaining, but it does little to challenge the underlying market power of platforms. The question remains to what extent this legal space will actually be utilised in practice through collective action. Many factors contribute to the fact that gig workers do not easily form a unified group: for example, the fragmentation of the labour market, the heterogeneous nature of the group and the simple fact that there is no physical meeting place.¹¹⁴ The approach is therefore only partial, as it mitigates the symptoms of the power imbalance rather than addressing the structural sources of platform dominance. Differently phrased, in the two routes to addressing a power imbalance – strengthening the dependent party or weakening the powerful party – only the first is addressed.¹¹⁵

5.4 Addressing digital labour platform power in a social market economy: a new role for competition law

A more balanced approach to platform power would involve exploring ways to address both the strengthening of workers and the weakening of platform dominance. In the broader context of the social market economy, competition law could take a more proactive role in addressing the dominance of platforms. The dependency relationship that ties platform workers to their labour platforms remains a central issue. The broader structural power held by platforms – manifested through algorithmic management, dynamic pricing and the control over work conditions – remains largely unchecked. This sheds light on several directions for further research on this topic. Competition law could take a more interventionist role by addressing the practices that bind workers to platforms. By scrutinising these practices and ensuring they are transparent and non-exploitative, competition law could reduce the platforms’ control over key aspects of workers’ conditions. Furthermore, the notion of relative market power could be raised.¹¹⁶ Recent developments at the EU level and in member states like Germany and Belgium indicate a trend toward recognising relative market power – the power of one undertaking over another, even without traditional dominance.¹¹⁷ This framework could be extended to platform work, where the dependency of workers on a single platform creates significant relational power imbalances, even if the platform does not dominate the market in a traditional sense.

In the broader context of the social market economy, where the aim is to balance competition and social protections, competition law should be viewed as an active tool for rebalancing the power dynamics in platform work. While the Platform Work Package provides important protections for workers, these measures only treat certain symptoms of the power imbalance. To achieve a more just market for the workers in it, competition law must also address the root cause of dependency: the concentration of power within digital labour platforms.