The effective application of the right to collective bargaining for self-employed (platform) workers: ‘Not such an easy task’

Introduction ¹

Gus Tyler, an American socialist activist, labour union official, and author, once said:

‘‘[w]hen all is going well, there is little need to look around or to look within. But when the present gets shaky and the future threatening, it becomes necessary to ask what is wrong. Easy answers give way to hard questions.’’ ²

Unsurprisingly, the developments in the labour market have also affected industrial relations systems. Self-employment has grown significantly in recent years and is expected to continue to do so, especially with the rise of platform work. The situation, therefore, makes it even more urgent than in the past to ask the right questions by ‘look[ing] around or [and] [by] looking within’ the current system(s) of industrial relations. When someone attempts to look around, one of the main challenges is the conflict between competition law and collective bargaining. ⁴ As has been pointed out elsewhere, existing competition law approaches may not be broad enough to address the changing realities and inequalities of contemporary labour markets. ⁵ Indeed, as most national legislations do not exclude collective bargaining agreements from the scope of competition law, antitrust authorities have been using their influence to target certain categories of workers – platform workers amongst them ⁶ –who have concluded collective bargaining agreements on the grounds that these are price-fixing agreements between undertakings, and hence, they hamper competition. To address such issues, some suggest that competition law should be viewed through the lens of recognising the right to bargain collectively as a fundamental human right. ⁷ Despite this, antitrust laws still hinder the ability of self-employed workers to engage in collective bargaining by considering such agreements as price-fixing agreements that could stifle competition.

That being said, what if competition laws were not a significant obstacle to the exercise of collective bargaining as a human right? In other words, what if collective agreements for self-employed workers were entirely legitimate under competition law, or had only limited restrictions in cases of dominant market power? ⁸ If this were the case, the situation would not appear as challenging as it currently is when one tries to ‘look around’. But then, the question that arises as to how the situation appears when we wish to ‘look within’ the current framework of industrial relations. As discussed in the next sections, the legal apparatus for collective bargaining is often out of date and does not reflect the realities of the new world of work and the increasing number of self-employed workers who are dependent on their principals. This observation, in turn, raises the question of whether it is necessary to reconsider how industrial systems should function to address these new realities and comply with international human and labour rights instruments and their supervisory bodies concerning the right to collective bargaining.

The next sections of this article aim to make some important observations and initiate a discussion on the future of industrial relations in light of collective bargaining as a human right, rather than offering universally applicable solutions, given the diverse landscape within national industrial systems. Section one, in particular, briefly discusses the nature of collective bargaining as a fundamental human right and the ramifications of such recognition for its scope of application. In turn, section two, examines the conclusion of collective agreements for self-employed workers, by particularly focusing on developments in the context of platform work. Section three discusses some of the reasons behind the inability of current industrial relations systems to allow self-employed workers to fully exercise their right to collective bargaining in practice. Finally, I conclude with some recommendations and a discussion on the future of industrial relations.

1. The human right to collective bargaining and its broad scope of application

It is well-established that an essential element in the exercise of the right to freedom of association, ⁹ is the right to collective bargaining. ¹⁰ It is enshrined as a fundamental human right in several human rights instruments, ¹¹ as well as in the ILO Declaration on Fundamental Principles and Rights at Work. ¹² The latter, albeit not a human rights instrument per se, stipulates that the rights included in the Declaration are universal and should be applicable to all workers in all Member States, by the very fact of their membership and irrespective of ratification of the relevant conventions. ¹³ The nature of collective bargaining as a human right has also been emphatically confirmed by the jurisprudence and pronouncements of bodies supervising the implementation of human rights treaties. ¹⁴

Therefore, while under international law, there is a strong legal basis and a broad consensus that both the rights to freedom of association and collective bargaining are human rights with universal application, there is no justification that they should only be enjoyed within the realm of an employment relationship. Indeed, ILO Convention No. 87 is clear on the scope of the application of such rights. In its Article 2 it is enshrined that: ‘[w]orkers and employers, without distinction whatsoever, shall have the right to establish and, subject only to the rules of the organisation concerned, to join organisations of their own choosing without previous authorisation.’ ¹⁵ As Dorssemont argues, the phrase ‘without distinction whatsoever’, does not solely constitute a warning against the recognition of the right to freedom of association in a discriminatory manner, which would, for example, exclude some workers on the basis of their sex or religious belief, but it also seems to accord to the word ‘worker’ the broadest possible denotation. ¹⁶ Moreover, both the Committee on Freedom of Association (CFA) and the Committee of Experts on the Application of Conventions and Recommendations (CEACR) have repeatedly pointed out that Conventions No. 87 and 98 should apply to all workers, regardless of the existence of an employment relationship that is often non-existent. ¹⁷ Only limited categories of workers can fall outside the scope of application of the principles and rights enshrined in these Conventions. ¹⁸

In a similar vein, within the context of the Council of Europe, the European Committee of Social Rights (ECSR) has also explicitly confirmed the wide scope of both the rights to freely associate and bargain collectively. ¹⁹ Concomitantly, things are less straightforward with respect to the European Court of Human Rights (ECtHR). ²⁰ The Strasbourg Court, however, has never explicitly opinioned that only employees should enjoy the right in question. Not only this, but the ECtHR has not hesitated to find violations of Article 11 of the European Convention on Human Rights (ECHR) in cases brought by workers without an employment relationship. ²¹

Going back to the ECSR's approach, the latter, based on Article 6§2 of the European Social Charter (ESC), has clarified that all workers should have the right to bargain collectively, including self-employed workers. ²² It did so in a case that concerned the pre-2017 ban on collective bargaining for freelance journalists, voice-over actors, and session musicians in Ireland, where the domestic law has been recently amended so as to include some specific categories of workers, beyond employees, that may be allowed to conclude a collective agreement. ²³ The case was brought in front of the ECSR by the Irish Congress of Trade Unions (ICTU) back in 2016. It was claimed that the blanket ban on all agreements between self-employed workers and organisations of self-employed workers ²⁴ was at odds with the country's obligations to protect the rights of workers enshrined under Article 6 ESC. Despite the 2017 amendments, which exempted certain categories of self-employed workers from antitrust laws, the ICTU insisted that restrictions on the right to bargain collectively still remained.

The ECSR held that the blanket ban in Ireland prior to the 2017 amendments was indeed in breach of Article 6 of the Charter. It seized the opportunity to point out that the purpose of competition law is not to restrict the rights to freedom of association and collective bargaining for those workers who may need it the most - in this particular case, self-employed workers, such as voice-over actors, freelance journalists and musicians. It went on to mention that ‘an outright ban on collective bargaining of all self-employed workers would be excessive as it would run counter to the object and purpose of this provision’ and, therefore, it was ‘not necessary in a democratic society in that the categories of persons included in the notion of “undertaking” were over-inclusive.’ ²⁵

Throughout its ruling, the ECSR made extensive reference to international labour standards and particularly to the universal applicability of the ILO Conventions 87 and 98. In relation to the ICTU's position on the 2017 statutory amendments, the majority of ECSR's members found no violation of Article 6(2). The claim was considered to be ‘speculative’, as there was still not enough evidence as to how the new law would be applied in practice by the relevant authorities. ²⁶ Albeit this conclusion is undeniably open to some criticism, ²⁷ for the purposes of this section let us instead focus on another important point made by the ESC. This can be found in paragraph 38 of the ruling, where it was pointed out that:

[i]n establishing the type of collective bargaining that is protected by the Charter, it is not sufficient to rely on distinctions between worker and self-employed, the decisive criterion is rather whether there is an imbalance of power between the providers and engagers of labour. Where providers of labour have no substantial influence on the content of contractual conditions, they must be given the possibility of improving the power imbalance through collective bargaining. ²⁸

Overall, the language used in several international labour and human rights instruments, as well as the findings and wording of their respective supervisory bodies, illustrate that the right to collective bargaining should enjoy a broad personal scope of application, irrespective of workers’ employment status. However, as the next sections discuss, things are much less straightforward when it comes to the exercise of collective bargaining in practice.

2. Collective bargaining for platform workers and other self-employed workers

In most European countries, trade unions can legally recruit and organise self-employed workers, including those in the platform economy. At the same time, though, as Fulton observes, it is much harder for unions to engage in collective bargaining for these workers. ³⁰ It is true that some national models make it easier for self-employed workers to access collective bargaining. For example, in some countries it is common to permit self-employed persons in specific occupations and industries, such as journalists, actors, artists and farmers, to conclude collective agreements. ³¹ Moreover, in some legal systems, mainly in Europe, there are provisions in mainstream collective bargaining legislation that authorise workers falling within an intermediate category between employment and self-employment to undertake collective bargaining. ³² Some countries have, furthermore, amended the provisions of other legislation in order to permit certain self-employed persons to have their terms and conditions set by collective agreement, an action that would otherwise possibly be unlawful. This is, for instance, the case in Ireland, where the Competition Act was recently amended so as to exempt self-employed workers in certain sectors from the ambit of antitrust laws. All these types of national models are not mutually exclusive.

Significantly, at EU level, the European Commission recently published its long-awaited Guidelines on the application of EU competition law to collective agreements concerning the working conditions of solo self-employed persons, including platform workers. ³³ The Guidelines stipulate that they cover three categories of solo self-employed persons. These categories include: a) those who are economically dependent, meaning that they have to receive at least 50% of their total annual income from a single principal; b) those who work ‘side-by-side’ ³⁴ with workers working for the same principal; and c) those who provide their services via the use of digital labour platforms. ³⁵ Although it is too early to assess the impact that the EU Guidelines will have for self-employed persons in Europe, they are expected to resolve some of the issues between labour law and competition law. Still, as has been argued elsewhere, by no means should the Guidelines be seen as a panacea for all the issues vulnerable self-employed workers are facing, especially considering their soft law nature. ³⁶

As noted, within the realm of the platform economy there are also a few instances where platform workers enjoying self-employed status have managed to negotiate collective agreements or similar types of settlements. For example, in the UK in 2019, the GMB union signed a collective agreement with Hermes, a courier platform company. Hermes’ self-employed couriers now have the option to acquire a ‘self-employed plus’ status that provides them with important benefits, such as holiday pay and guaranteed earnings. ³⁷ Innovative in its own way, the agreement further attempts to challenge the isolation of platform workers by permitting the GMB union, with the company's assistance, to hold recruitment days eight times a year. Such a development was celebrated by both the trade union and the management team of the company as a groundbreaking agreement that can pave the way for more initiatives of this kind in the UK. It is the first and the sole instance of a digital platform permitting its alleged self-employed workers to be represented by a trade union. ³⁸ As Katsabian observes, nonetheless, it is highly contested whether the new agreement really allows those with ‘self -employed plus’ status to enjoy their collective labour rights, as it solely refers to ‘the individual right of each courier to negotiate her pay separately from the other workers’. Hence, the right to collective representation appears to be purposely missing from the agreement at issue. ³⁹

Another example exists in the territorial collective agreement concluded in Bologna, through the establishment of the ‘Charter of fundamental digital workers’ rights within an urban setting’. ⁴⁰ The latter was signed by institutional trade unions, autonomous workers collectives, the city council and the local food delivery platforms, Sgnam and MyMenu. ⁴¹ The Charter, though applicable on a voluntary basis, contains minimum standards on aspects of work such as remuneration and working time. It also has a broad scope of application as it is applicable to all platform workers operating in the city of Bologna, irrespective of their employment status. ⁴²

Another example can be found in the efforts of the Transport Union Workers (TWU) in Australia to collectively represent platform workers and, more specifically, a group of Deliveroo workers. Following the death of four delivery riders in the course of their work, Deliveroo workers, supported by the TWU, requested that the company initiate the formation of work groups and the election of health and safety representatives. ⁴³ Under the relevant national law(s), such representatives can be elected upon the request of ‘workers’ in one or more ‘work groups’. ⁴⁴ For the purposes of the law at issue, the definition of who is a worker makes no distinction between an independent contractor and an employee. ⁴⁵ Deliveroo, which had previously appointed its own ‘rider panel’, comprised of workers selected by the company itself, opposed the idea brought forward by the union. Its denial led the TWU to bring a claim before the work health and safety regulatory agency. Importantly, the latter provided Deliveroo with a notice obliging it to comply with the national legislation and permit the election of workers’ representatives. ⁴⁶ Although it is not a typical scenario of collective bargaining, the agreement between Deliveroo and TWU represents an alternative method by which unions have attempted to safeguard workers’ rights via specialised legislation with a broad scope of application.

All in all, and despite the aforementioned examples, the fact that the majority of platform workers continue to be classified as self-employed automatically excludes them from the coverage of existing collective agreements. This situation occurs although they often share many of the characteristics found in a typical employment relationship. ⁴⁷ At the same time, even when collective agreements for self-employed platform workers exist, as discussed, they have not necessarily followed the typical scenario of collective bargaining. That is to say, they have adopted a more ‘diluted’ version of collective bargaining, as in the case of the Hermes collective agreement in the UK. The latter, albeit an important step in the right direction, has failed to secure the collective representation of the couriers involved in the agreement.

3. The current structure of industrial relations systems: ‘the enemy within’

As discussed, traditional unions representing solo self-employed workers, including platform workers, are often able to negotiate working conditions for their members despite potential obstacles posed by competition authorities. However, the question remains whether simply allowing solo self-employed persons to engage in collective bargaining is enough to enable them to fully exercise their right to collective bargaining. In responding to this question, authors, like Christina Heißl, do not to seem that convinced. By basing her view on a comparative analysis between several European countries, she points out that ‘simply allowing collective bargaining for the self-employed may not be sufficient to enable significant bargaining activities in reality’. ⁴⁸ This, according to Bogg, can result in a situation where there is only a ‘liberty’ to bargain collectively rather than a ‘right’ to bargain collectively; ⁴⁹ a situation that, arguably, can contradict countries’ obligations under international labour and human rights law to facilitate and support social dialogue and collective bargaining for self-employed workers.

Despite their distinctions, it can be argued that in the majority of cases, current national collective bargaining systems are not equipped enough to cater to the needs of self-employed workers, even if the latter are permitted to negotiate collectively. This is due to the fact that most collective bargaining systems are structured around formal employees and particularly those working full-time in fixed workplaces. Indeed, by looking at the international encyclopaedia of labour law and industrial relations, which conducts a comparative study of many countries worldwide, it can be observed that the term ‘self-employed’ is either entirely missing or infrequently showing up in the respective sections of national research pertaining to collective labour relations. ⁵⁰ Arguably, such an absence implies that national legislators did not consider the self-employed when drafting these provisions. High representation thresholds for union recognition, a lack of multi-employer collective bargaining, the absence of statutory rights to force employers to recognise collective agreements for self-employed workers, and other factors, are currently insurmountable obstacles to the effective exercise of collective bargaining rights for such workers. Moreover, practical difficulties, such as low unionisation rates, a limited attachment to the workplace and individualised work arrangements, cannot be overlooked, and in certain national contexts these issues have been found to be more significant than legal impediments, like the intervention of competition authorities. ⁵¹ Needless to say, depending on the national system in question, impediments can take different forms.

Let us take, for instance, the example of Belgium. Although the issued EU Guidelines suggest that collective bargaining agreements for solo self-employed workers, including platform workers, will not violate competition law, the practical implementation of such agreements, as Pecinovsky explains, is not straightforward. ⁵² In Belgium, the relevant legislation with respect to the right to bargain collectively is the Collective Labour Agreement Act of 1968. ⁵³ The Act prescribes that collective agreements can be ‘concluded between one or more employees’ organisations on the one hand, and one or more employers’ organisations or one or more employers’ on the other hand […].’ ⁵⁴ Against this background, collective agreements, governed by the Act of 1968, are considered binding in nature, whereas those which fall outside the ambit of the Act are regarded as second-rate collective bargaining agreements. Although they come with rights and obligations, they are not legally enforceable against many people and do not offer collective protection. ⁵⁵ That said, as Pecinovsky informs us, for solo self-employed persons, the major obstacle to access to the Act for the purposes of collective bargaining is the latter's limited personal scope of application. This is because Article 2 of the Act is only applicable to workers in furtherance of an employment contract and to employers. ⁵⁶ According to Belgian labour law, an employment contract ‘is defined as the agreement whereby an employee undertakes, in return for remuneration, to perform work under the authority of an employer’. ⁵⁷ At first glance, the Act indeed appears to de facto exclude solo self-employed from its scope. Nevertheless, as Franquet and Maisin thoroughly explain, it is now settled law that the Act’ s provisions are also applicable to ‘all persons who provide work benefits under the authority of another person, whether or not under a contract of employment’. ⁵⁸ The key, therefore, is to determine what the concept of authority entails, as the Act remains silent on this issue. Courts use various criteria to determine the presence of authority or subordination under national law, such as an employer's ability to supervise and control the work, the existence of a work schedule, etc. ⁵⁹ Frank Hendrickx argues that the concept of authority does not necessarily require a person to relinquish all autonomy over her or his work, and a subordinate relationship can exist even if a person ‘retains full independence of the work and responsibility for the work.’ ⁶⁰ As the same author explains, though, it is indeed crucial that the authority remains present in the organisation of the independent work performance; otherwise, a person will likely be considered self-employed. ⁶¹

Franquet and Maisin suggest that a more flexible interpretation of the concept of authority could potentially allow the application of the Collective Labour Agreement Act to self-employed individuals. They seem aware, nonetheless, that this alone will not solve the problems faced by such workers in exercising their right to collective bargaining. ⁶² For example, it is difficult to imagine how freelance crowdworkers, who typically organise and manage their own work, could be covered by the Act. After all, as Pecinovsky contends, a debate around authority, even if the latter can be interpreted in rather a broad manner, brings us back to the conversation about the employer-employee relationship and its well-known complexities. ⁶³ That said, as the same author highlights, the challenges in respect of self-employed persons’ collective interests do not end here. More specifically, the 1968 Act only allows employers’ organisations to represent self-employed workers in collective bargaining, which further complicates matters. ⁶⁴ While in practice, organisations representing solely self-employed workers, like Unizo and the Union des Classes Moyennes, have negotiated working conditions for their members, such organisations are also considered to be part of employers’ organisations. This is because ‘they both represent their members in the National Labour Council among the quota allocated to employers’. ⁶⁵ All in all, the Collective Labour Agreement Act of 1968, drafted a long time ago, does not seem to be able to accommodate the complexities of modern work arrangements, with the increase of self-employed workers in need of collective representation. As noted, collective agreements concluded for the self-employed and falling outside the Act's scope do not have a binding force. On top of that, unlike collective agreements protected by the Act (based on its Articles 19 and 26), they cannot have universal applicability, meaning that they do not possess the legal force to cover all self-employed workers engaged, for instance, in a specific sector or industry. ⁶⁶

Moving forward, the complicated legal situation regarding the right of self-employed workers to bargain collectively can also be observed in other countries with less institutionalised systems of collective bargaining compared to Belgium. This is the case, for example, in the UK, where self-employed platform workers, including those classified as independent contractors, have managed to conclude a collective agreement with a digital labour platform, Hermes. The existence of this agreement does not necessarily mean that we can easily draw conclusions from this single example. Several authors have demonstrated that platform workers and other self-employed persons have difficulties meeting the requirements set out in the Trade Union and Labour Relations (Consolidation) Act 1992 (TULRCA), which provides the statutory mechanism for trade union recognition. ⁶⁷ Only recognised unions are protected by the Act and have the legal authority to negotiate with employers and engage in collective bargaining or action. However, as in the case of Belgium, the problem with the TULRCA lies in its limited personal scope of application. The Act defines a trade union as an organisation that is comprised ‘wholly or mainly of workers of one or more descriptions and whose principal purposes include the regulation of relations between workers of that description or those descriptions and employers or employers’ associations’. ⁶⁸ Conversely, where an association is not comprised ‘wholly or mainly’ of ‘workers’, it cannot constitute a trade union for the purposes of the Act, a status that, as explained elsewhere, is essential in order to apply for a certificate of ‘independence’. ⁶⁹ As Bogg discusses, without the formal recognition and certificate, a workers’ association is barred from accessing the legal apparatus (schedule 1 of the Act), which offers legal support for collective bargaining. ⁷⁰

The broad definition of ‘worker’ in the Trade Union and Labour Relations (Consolidation) Act 1992 (TULRCA) includes ‘limb-b workers’, but it does not apply to solo self-employed workers. This was confirmed by the Central Arbitration Committee (CAC) in a recent case taken against Deliveroo. ⁷¹ However, this does not prevent independent contractors from being represented by trade unions. Still, if such workers wanted to form their own union tailored to their specific needs, they would not be able to obtain a certificate of independence under the current Act, a situation that would limit their access to legal support for collective bargaining. Nonetheless, self-employed workers are not legally prohibited from engaging in collective bargaining with employers. Particularly, as Bogg explains, considering the ‘voluntarist’ roots of industrial relations in the UK, ⁷² there is nothing to prevent employers ‘from negotiating with any labor association, regardless of whether it is a “trade union” or whether the represented individuals are “self-employed”.’ ⁷³ The Hermes collective agreement is illustrative of what the author describes. Still, as in the case of Belgium, such agreements can be characterised as ‘second-rate collective bargaining’. Among other reasons, this is because employers are not legally obligated to bargain collectively with self-employed unions, and voluntary recognition is unlikely to occur unless the union has adequate power to push the employer to accept its claim. ⁷⁴ Solo self-employed workers, therefore, ‘have no statutory right to compel an employer to recognise them for collective bargaining purposes’. ⁷⁵

The examples of the UK and Belgium demonstrate that merely granting self-employed workers the right to collective bargaining does not necessarily guarantee success in practice. Outdated collective bargaining systems and practical obstacles can hinder the effective exercise of this right. Another example - and in my view a rather straightforward one - is the recent labour law amendment that was introduced in Poland in 2018 and entered into force a year later. In order to comply with international standards and the Polish Constitution, national law was amended by extending collective bargaining rights to ‘all persons working for money’. ⁷⁶ Consequently, as Mirtus explains, the right to collective bargaining is now vested in self-employed workers who are economically dependent upon a single client and do not employ other people. ⁷⁷ As the same author as well as other national experts argue, though, it is difficult to imagine that collective agreements will start covering non-employees in practice, including platform workers, due to this change. ⁷⁸ Negotiations for non-employees are lacking. ⁷⁹ Due to trends towards decentralisation, multi-employer collective bargaining, which could possibly benefit solo self-employed workers, is almost non-existent in the country, despite being permitted in theory. ⁸⁰ At the same time, although the number of company-level agreements is larger, these have been mostly negotiated for specific types of establishments. In the private sector, these types of agreements are very few and far between, and in small and medium enterprises, almost non-existent. ⁸¹ On top of that, as Pisarczyk highlights, the factual challenges are currently magnified by inadequate legislative solutions that further complicate the work of trade unions and the facilitation of collective bargaining. ⁸² In a similar vein, it has been also contended that the Polish state has done little for the facilitation of social dialogue at a national level. ⁸³ Thus, once again, a mere permission to bargain collectively for self-employed workers does look enough in bring about the full realisation of the right at issue.

To sum up, systems of collective employment relations, established on a reality of a different era (e.g., a fixed workplace and a homogeneous workforce, working full-time), often do not appear suitable for accommodating the needs of solo self-employed workers when it comes to their collective protection and right to collective bargaining. The obstacles discussed previously constitute only some of the possible legal impediments which hinder the effective enjoyment of the right in question by self-employed persons, including platform workers. Anthony Forsyth, for example, dedicates considerable space in his recent book to addressing the challenges stemming from high representation thresholds for union recognition and the validity of collective agreements – a phenomenon that can indeed obstruct the efforts of self-employed workers to form trade unions and bargain collectively. ⁸⁴ Obstacles of this nature go against recognised international standards, according to which ‘social partners or governments shall install objective procedures for recognising the most representative organisations (in so far as only these may bargain collectively) with reasonable quotas if representative quotas are imposed.’ ⁸⁵ Moreover, many point out that industrial systems, such as the one in the US, ⁸⁶ currently structured around collective bargaining procedures in a single worksite, and at most with a single employer, are unable to deal with the challenges that derive from fissured workplaces and the increase in the number of self-employed. ⁸⁷ As a consequence, fewer and fewer atypical workers are covered by collective agreements. That said, it is beyond the scope of this section to fully address the totality of legal obstacles embedded within national industrial systems, which render the application of a human rights- based approach towards collective labour rights an extremely challenging task. Still, this small sample of legal complexities illustrates that even if we take for granted that all workers, irrespective of their employment status, can enjoy their right to collective bargaining, then the implementation of such an assumption seems practically impossible for many solo-self-employed workers.

4. Looking ahead: there are no easy answers

The previous section's findings suggest that existing national industrial systems appear inadequate in providing collective representation and protection for solo self-employed workers, including those who work for platform operators. Although exceptions exist, in many cases the freedom to collectively bargain has proven ineffective, and self-employed workers are facing serious limitations in the enjoyment of their fundamental human rights, like the right to collective bargaining. This situation raises difficult questions for policymakers, trade unions, and scholars, which can be divided into two categories. The first group of questions relates to how national industrial systems can adequately ensure the right to collective bargaining for self-employed workers. Should new collective bargaining laws be created, or should existing ones be modified? Such questions have already been raised by some authors. ⁸⁸ After all, it is rather ironic to discuss how competition law can be changed to align with the fundamental human right of collective bargaining for all workers without having similar discussions regarding collective labour law.

That said, the second set of questions pertains to what will happen if no significant changes are made, or if progress is slow. Will we witness the emergence of weakened forms of collective bargaining, which are not as robust or binding as traditional models? In the context of the platform economy, these, for instance, could take the form of agreements on a voluntary or non-binding basis, such as the 2020 Bologna Charter, or agreements with a narrow material scope, such as the transport workers’ agreement in Australia for the election of health and safety representatives.

It may be necessary to conduct new research or publish multiple works to address the complex questions at hand. The diversity and complexity of national industrial relations systems make this task particularly challenging. Moreover, as Otto Kahn-Freund mentioned a long time ago, a one-size-fits-all solution can accommodate several hazards. ⁸⁹ Nonetheless, such an observation, along with the diversity of collective labour laws around the world, should not discourage us from seeking to identify some common requirements that can facilitate in practice the exercise of collective bargaining for self-employed, including many platform workers. Therefore, it is important to identify common requirements that can facilitate such an exercise.

One such requirement may drive from the adoption of an autonomous definition of ‘worker’, applicable solely for the purposes of unionisation and collective labour laws. Such a definition should be in line with international obligations and insulated from definitions used for other labour-related rights. As observed, both the ILO and the ESC, through the language of their supervisory bodies, have embraced an all-inclusive notion of ‘worker’ that is different and/or autonomous from the notion of ‘worker’ that employment and labour legislation currently uses; or similarly, a notion of ‘worker’ to which labour courts are typically referring when ruling on whether a person can be considered to be in an employment relationship sensu stricto. Similarly, therefore, national labour legislation setting out the conditions for union recognition and the exercise of collective bargaining should insulate itself from definitions found in other legislation regulating employment and labour relations more generally, as well as the decisions of labour courts on the existence of an employment relationship (‘insulation approach’). For instance, in the UK, section 296(1) of the TULRCA, which defines a worker for the purposes of the Act in question, is identical to section 230(3)(b) of the Employment Rights Act 1996. ⁹⁰ This cross-reference results in the failure to deliver the right to collective bargaining to all the self-employed in need of it, and introduces complications with respect to official union recognition. A similar observation can also be made with respect to the Collective Bargaining Act in Belgium, which, as mentioned, also covers persons without an employment contract who work under the authority of another person. However, given that the notion of authority is not defined within the Act, reference needs to be made to other pieces of legislation, such as the Labour Relations Act 2006. This reference can open the Pandora's box of the classic discussion of the employee-employer relationship. ⁹¹ Having such complexities in mind, collective labour laws should include their own autonomous definition of ‘worker’, away from the influence of definitions used for the exercise of other labour-related rights. In this way, an insulation approach can prevent considerations around authority and subordination from considerably restricting the personal scope of application of unionisation and collective bargaining rights. Moreover, the exercise of such rights will not be dependent upon the eagerness and capacity of individual workers to bring claims before the courts.

In accordance with the ESC's language, among other possible elements, a decisive criterion for the establishment of the insulation approach can be found in the ‘imbalance of power between the providers and engagers of labour’. ⁹² When the former ‘have no substantial influence on the content of contractual conditions, they must be given the possibility of improving the power imbalance through collective bargaining.’ ⁹³ Proposals for an autonomous definition of who should be considered a worker for the purposes of unionisation and collective bargaining, solely in the case of platform workers, have already been advanced in some countries. ⁹⁴ For instance, a Sharing Economy Committee appointed by the Norwegian Government expressed the view that self-employed platform workers who do not influence the setting of prices should have the right to collective bargaining. ⁹⁵ Tammy Katsabian on her side, contends that for the purposes of platform workers’ unionisation rights only, the term ‘worker’ in the UK can be redefined and expanded under the TULRCA based on the criterion of economic dependency. ⁹⁶ She also argues that the extension in question may possibly be used for other self-employed workers with similar characteristics.

Adopting an autonomous definition and separating collective labour laws from definitions of the term ‘worker’ found in other legal sources are both crucial first steps, but they may not be sufficient on their own. This is to say that platform workers and other self-employed individuals may still face significant obstacles to organising collectively due to the complex requirements for the creation and official recognition of trade unions. Formal processes developed on the existence of a workforce working full-time, based on a stable and continuous employment relationship, need to be (re)adjusted or redefined in order to capture the realities of self-employed persons. Relaxing the requirements for unionisation enshrined within national collective labour laws is an idea that has already been put forward by some scholars. ⁹⁷ Such proposals will mostly depend on the unique features of the relevant national legal system, which is why I choose to focus on another requirement that I believe is of paramount importance to right of self-employed workers to collective bargaining, namely, the enabling and facilitation of multi-employer collective bargaining settings.

The lack of constraints of such procedures for collective bargaining can be harmful to the self-employed, even in countries that technically allow these workers to bargain collectively, such as Poland. Research shows that countries with centralised bargaining systems that allow multi-employer negotiations have higher rates of unionisation and coverage under collective agreements. ⁹⁸ This may be due to the fact that centralised systems limit employer incentives to eliminate unions and reduce transaction costs for both unions and employers. ⁹⁹ This can benefit grassroots unions in the platform economy, which often have limited resources and negotiating power. Centralised systems also mitigate ‘the need for unions to gain recognition and recruit members in small firms, [a situation that in turn] helps to solve the latent conflict between capital and labor and can bring macroeconomic benefits so that governments may have an interest in maintaining strong unions’. ¹⁰⁰ Furthermore, it has been contented that multi-employer bargaining settings are associated with more inclusive labour protection, as they establish minimum standards for payment and other work-related aspects in a specific sector or industry. ¹⁰¹ To this extent, the coverage of collective bargaining agreements tends to extend beyond trade union membership. ¹⁰² All in all, without multi-employer collective bargaining settings, the effective exercise of the right to collective bargaining for self-employed workers, including platform workers, is difficult. Proposals for legal reforms of this kind already exist, ¹⁰³ but their implementation also requires political will, which often seems to be lacking in a neoliberal environment that goes hand in hand with trends towards the erosion of multi-employer collective bargaining systems. ¹⁰⁴ While the question of whether the enabling of multi-employer collective bargaining systems is feasible within this political environment is an important one, it is outside the scope of this article. Undoubtedly, though, it should be kept in mind for future interdisciplinary discussions.

A final remark regarding the requirements necessary in facilitating and promoting the exercise of collective bargaining for self-employed in practice relates to the difficulties that many self-employed workers - and especially platform workers - experience in communicating with each other. In this regard, it should be acknowledged that the dispersed nature of their work acts as a practical barrier to unionisation and collective negotiations. In certain cases, such obstacles can be viewed as equally significant as the legal complexities encountered by these workers. ¹⁰⁵

Nonetheless, the practical nature of this issue should not prevent national authorities from ensuring that in the context of industrial relations, employers undertake commitments for the creation of mechanisms through which workers can communicate with each other, and via which the organisation of workers by unions can indeed be facilitated. Such obligations on the employer side are extremely important, especially as far as platform work is concerned. The lack of communication channels is addressed by the draft EU Directive on platform work, ¹⁰⁶ whose Article 15 calls on Member States to guarantee that digital labour platforms will enable the creation of such mechanisms and will abstain from interfering with worker communication. The facilitation of communication channels and supervision of their correct implementation can be considered as part of the positive international obligations of states in respect of collective labour rights. ¹⁰⁷

The requirements discussed above are only a few examples of the potential changes that will enable the practical exercise of collective bargaining for self-employed individuals. Still, these conditions are just the initial step in a complex journey, and further research is necessary. For instance, it is important to explore how national industrial systems can accommodate the extension of collective agreements, which are currently established for employees, to cover self-employed individuals. The needs of the latter can significantly differ from those of employees, and vice versa. Should these agreements be revised, or should specific provisions for self-employed individuals be added to the existing ones? Furthermore, considering the diversity among self-employed workers, this process may become even more intricate.

Undoubtedly, these types of questions will occupy policymakers and workers’ representatives in future discussions concerning the adaptation or redesigning of national systems of collective bargaining. The answers to such questions are not only crucial for the protection of self-employed workers, including those in the platform economy, but they can also have broader implications for the foundation of labour law. As Bogg aptly argues, in order to maintain the normative coherence and integrity of labour law, the allocation of its fundamental rights should be sensitive to social and economic disadvantages, rather than tied solely to specific contractual forms like a ‘contract of service’. ¹⁰⁸ Consequently, many self-employed workers should fall within the scope of fundamental collective labour rights, which are also recognised as human rights, such as the right to collective bargaining.

That said, even if future changes occur in accordance with international obligations, implementing these changes will require time and depend on various factors for their success. This raises the question of what this potential delay means for the right to collective bargaining in the present, which leads us to the second set of issues identified at the beginning of this final section.

It can be proposed that recent developments in the realm of platform work, as highlighted in this article, can offer valuable insights on this matter. It has been demonstrated that certain digital labour platforms and trade unions representing self-employed platform workers have endeavoured to establish collective agreements. However, some of these agreements have faced significant challenges or have even been suspended due to hostile antitrust laws. Nevertheless, the recently issued EU Guidelines on collective bargaining for solo self-employed individuals may potentially address these issues at EU level.

That said, given the legal complexities within national industrial systems, self-employed platform workers have been compelled to engage in collective negotiations that may not adhere to traditional collective bargaining procedures. As observed, the Hermes collective agreement in the UK, for example, lacks a provision addressing the right of ‘self-employed plus’ workers to collective representation. Instead, it only grants couriers the individual right to negotiate their pay separately from other workers. Therefore, while important for certain protections, the Hermes agreement limits the collective power of self-employed platform workers, which contradicts the objectives of collective bargaining. In Italy, as noted, a noteworthy development is the conclusion of a territorial agreement known as the Bologna Charter, which sets out various labour safeguards for all platform workers, including the self-employed. Once again, this agreement does not follow the traditional framework of collective bargaining procedures, as it is voluntary in nature. This article has also discussed the collective negotiations of self-employed transport workers in Australia, which resulted in the election of health and safety representatives. In this context, the TWU union took advantage of the broad definition of ‘worker’ under the country's health and safety laws. Consequently, these instances demonstrate the alternative approaches that unions have employed to protect the interests of their workers.

It can be argued that these developments indicate a proliferation of either soft or diluted versions of collective bargaining, which, as mentioned, may not resemble robust or binding models of traditional regulatory forms of collective bargaining. Undoubtedly, in many cases these forms of collective bargaining offer temporary solutions or a sense of relief to platform workers, often due to a lack of alternative methods. They can also inform the work of trade unions representing self-employed workers beyond the platform economy. ¹⁰⁹ The Australian example demonstrates how self-employed workers, in general, can utilise other pieces of legislation to enhance their protection in specific areas of their occupation. One could also contend that such initiatives expand the notion of collective bargaining and the ways in which collective negotiations can be carried out. However, in my opinion, these soft and diluted models of collective bargaining should not be viewed as a cure-all. In other words, it is doubtful whether they strengthen the right to collective bargaining overall, as their limited material scope or legal enforceability constitute problematic aspects that should not be overlooked. This observation brings us back to the initial set of concerns.

As long as national collective bargaining systems fail to effectively ensure the exercise of the right to collective bargaining for self-employed individuals, including platform workers, the aforementioned models of collective bargaining will continue to proliferate. I am not suggesting, of course, that unions and other associations representing these types of workers should refrain from utilising all available means at their disposal to seek better working conditions. It is well-documented that by exercising their human rights to freedom of association, collective bargaining and strike, albeit with limitations, they have already achieved significant victories against powerful opponents such as platform giants. ¹¹⁰ In order for those wins to be multiplied and achieved by other workers, the effective facilitation and application of the right to collective bargaining for self-employed persons is an important step in the right direction and is also in line with the requirements of international labour and human rights instruments.

Conclusions

This article has demonstrated that as a result of the way existing systems of industrial relations are structured, the effective application of the right to collective bargaining for self-employed (platform) workers can be particularly challenging, even putting restrictions stemming from competition law aside. As noted, merely permitting collective bargaining for such workers, as a number of countries do currently, may not be enough to enable substantial bargaining activities in practice. Industrial relations systems such as those in the UK and Belgium indeed demonstrate that national legislators did not have the self-employed in mind when drafting provisions on collective labour relations. High representation thresholds for union recognition, the lack or decline of multi-employer collective bargaining, and the absence of statutory rights to compel employers to recognise collective agreements for self-employed workers, constitute only some of the impediments self-employed workers have to deal with when they attempt to exercise their right to bargain collectively. In practice, therefore, the latter takes the form of a mere liberty rather than the form of an actual right often enjoyed by employees. As explained, the challenges self-employed persons face in the conclusion of collective agreements can also be observed in the platform economy. Legal complexities have forced self-employed platform workers to proceed with collective negotiations that do not necessarily represent collective bargaining procedures of a traditional nature; they can be rather perceived as soft or diluted versions of collective bargaining. The Hermes agreement in the UK and territorial agreement via the establishment of the Bologna Charter in Italy are illustrative in this respect. Such agreements are expected to proliferate as long as impediments to collective bargaining for the self-employed persist within national industrial systems. While important in many respects, these types of agreements cannot be expected to have the same outcomes as agreements generated through collective bargaining processes of a traditional nature. Consequently, it is questionable whether such agreements point towards an effective application of a the right to collective bargaining, in line with the latter's international recognition as a human right with a broad personal scope of application. It remains to be seen how national systems of industrial relations will attempt to address such challenges and if their responses will be in line with the requirements of international labour and human rights instruments, as illustrated above.