ETUC reply to the First phase consultation of social partners under Article 154 TFEU on possible action addressing the challenges related to working conditions in platform work

new elected team 2019

ETUC reply to the First phase consultation of social partners under Article 154 TFEU on possible action addressing the challenges related to working conditions in platform work

Resolution adopted at the virtual Executive Committee meeting on 22-23 March 2021

Main messages

The ETUC welcomes the first phase consultation of social partners under Article 154 TFEU on possible action addressing the challenges related to working conditions in platform work. Recent Court cases and administrative decisions have shown that the platforms are still in breach of the laws regarding the respect of workers' rights and recognise again and again the misclassification of workers as (bogus-) self-employed while the platform behaves, with the help of its algorithmic management tool, as an employer.

The ETUC has two objectives: 1. to win rights for non-standard workers whether they work online or offline (including those in platform companies) and 2. to make the digitalisation of the economy compatible with the employment relationship and the respect for fundamental workers' rights.

The ETUC’s answer is articulated around the following seven political priorities:

1. From the current situation where the most vulnerable in the relationship (workers) in platforms work often are treated as self-employed without benefiting from the autonomy of this status, we must move towards a presumption of employment status in general, complemented by 

2. Reversal of the burden of proof by platforms, which will have to provide sound evidence that no employment relationship exists between them and workers in their respective platforms. Armies of digital platform lawyers will undoubtedly work hard to prove that a worker is truly self-employed if this is the case. 

3. A digital platform company is (just as a regular business) an employer, a (temporary work) agency or an intermediary. Platforms are not mere digital intermediaries, but they are in fact ‘companies’ endowed with a vast range of managerial prerogatives and powers and should therefore shoulder all the obligations that this status entails, including the function of employer when it applies. It will therefore be necessary to link these digital platform companies to their sector of activity and to the various provisions and regulations that exist there and have been negotiated in collective bargaining by the social partners.

4. The ETUC strongly opposes to the creation of a third status between workers and self-employed. Workers do not need a specific (and more limited) labour legislation which is different to the one which applies to workers.

5. A European initiative should cover all non-standard workers[1] and workers in platform companies (including the self-employed) in their access to collective and individual rights. The ETUC considers that workers in platform companies are not a new category of workers per se because a musician, a rider, a journalist or a cleaner are in the same situation in matters of the lack of social protection; the difficulties to organise themselves and bargain collectively. If an initiative targets only workers in platform companies, what would allow to grant them more rights than on one hand a domestic worker (non-standard worker) or on the other hand a solo self-employed in the offline economy? This would be a de facto creation of a third status.

6. The scope of an initiative on platform work should cover both on-location and online labour platforms. There is no clear distinction in the operation of these platforms which can justify avoiding to regulating them. Digitalisation of the economy and the development of telework reinforce the need to frame a future on work where digital labour platform comply with the labour and social rights. If we fail in this objective, companies in a range of industries could use the opportunity to undermine employment protections.[2]

7. There is a need for a joint and coherent European action in full respect to national industrial relation systems as most platforms are multinational companies. ETUC proposals (presumption of employment relationship & reversal of burden of the proof, obligations of platforms as companies and employers) do not need changes in the definition of a worker set by Member States and thus respects the principle of subsidiarity and the autonomy of social partners.

The ETUC has high expectations for the EU initiative on platform work. Lack of action leaves millions of workers without access to their basic rights. Inappropriate legislative action, allowing platforms to continue to violate these rights, would have disastrous effects on workers and the world of work. If our redlines are not respected and the latter option appears to be the one chosen, the trade union movement will take its responsibility to prevent this scenario.

I. Do you consider that the European Commission has correctly and sufficiently identified the issues and the possible areas for EU action?

In answering this question, the ETUC seeks to engage with the issues identified by the European Commission while also highlighting some of the deficiencies of its analysis. We argue that the jurisprudence of the Court of Justice of the European Union has developed beyond the case law and concepts referred to in the Commission’s ‘Consultation Document’, and we identify some of omitted judgments.

Furthermore, the ETUC presents and disputes some of the misleading arguments used by platforms companies in their lobbying strategy, to prevent any European legislation from developing.

The ETUC agrees that the Commission’s identified areas to be tackled. The ETUC acknowledges that the glossary of the document (p. 5) identifies some of the key concepts shaping these forms of work and economic activities. However, there are reservations concerning the expression ‘de facto’ in the definition of the false self-employment, as it could make the presumption of employment relationship more difficult. If, as the Document suggests, ‘false self-employed’ are ‘employees of their contracting entity’, then the problem is not a factual but a legal one and the ‘contracting entity’ is in fact the employer. We will continue in our response to use the wording we find most appropriate talking about workers in platform companies.

There is a major issue that the Commission has not tackled: A digital platform company is (just as a regular business) an employer, a (temporary work) agency or an intermediary. Platforms are not mere digital intermediaries but they are in fact ‘companies’ endowed with a vast range of managerial prerogatives and powers and should therefore shoulder all the obligations that this status entails. The only novelty is the way they use digital tools, such as apps and algorithms to bring together workers and clients, while platform companies still can exercise labour control via the app. We venture to suggest that it is impossible to address many of the ‘challenges’ identified in the Consultation Document unless one accepts the fundamental premise that the corporate entities that own and control the algorithmic tools defining the fundamental terms and conditions for the performance and the organisation of work are, for all purposes, employing companies. Recognising platforms as companies is a necessary step to determine the obligations they have as employers towards workers but also as businesses towards genuine self-employed, customers, and other businesses. It is impossible to advance on the topic of collective bargaining without determining who exerts the employer function. The function of employer would resolve issues such as illegal subcontracting of accounts, which are being overlooked by the Commission. Finally, it is of utmost importance that the activities and obligations of platform companies are also regulated by reference to the sector of activity where they organise work. This is the only way to ensure a level-playing field and fight unfair competition in sectors where these platform companies are developing. A sectoral approach would also allow the application of provisions fixed and/or bargained collectively specifically for each sector (health & safety provisions, adequate wages, training provisions…).

I.1 Employment status

The ETUC agrees that “when addressing working conditions and social protection challenges for people working through platforms at national and EU levels, the core issue is the employment status” (p.7).

The ETUC insists that there are different measures and ways to tackle these problems. Furthermore, the Commission rightly underlines that many issues are shared with workers in other non-standard forms of work.

The ETUC acknowledges that a “basic distinction is made between on-location labour platforms (like passenger transport, deliveries, domestic work) and online labour platforms (where the tasks are not location-dependent, e.g. encoding data, translation work, tagging pictures, IT or design projects)” (p.5). The distinction between onsite - online is in many cases difficult to make and not a criterion to deny labour and social rights. This distinction, however, cannot imply that workers active in some type of platform company continue to be denied their labour and social rights. It can only help identifying additional challenges and issues that must be tackled over and beyond the minimum level of rights. The ETUC disagrees when the Commission seems to claim that, in contrast with temporary agency work, “platform work, in which many people are currently classified as self-employed’ does not entail a clear employment relationship” (p.4). The ETUC agrees with the Commission that the way to classify a relationship as an employment relationship can and should be improved.

When the Commission underlines that “less than half of Member States have taken actions directly relevant to employment status” (p.8). It must bear in mind that the Commission in its Communications on collaborative economy advised the Member States to abstain from action: "Member States are advised to take the opportunity to review, simplify and modernise market access requirements that are generally applicable to market operators. They should aim to relieve operators from unnecessary regulatory burden, regardless of the business model adopted, and to avoid fragmentation of the Single Market".[3] Under the growing pressure of precarious working conditions and their impact on the future of work, some Member States decided that they could not wait any longer. The ETUC regrets that the Commission remains ambiguous on the issue of a ‘third status’ which is a clear redline for the trade union movement (p. 9).

I.2 Working conditions

Digitalisation presents opportunities, so do platforms: opportunities in terms of job creation, albeit often precarious in nature; additional income, oftentimes unpredictable and discontinuous.  They also have risks: sub-standard working conditions; unfair pay; lack of rights or collective representation; unfair work-life-balance; health and safety hazards; inadequate social protection; misclassification as self-employed with all the consequences in terms of social security; and pension poverty. Furthermore working hours are often long, wages low, and, in some platforms, a significant gender pay gap and other types of discrimination are either tolerated or reinforced by algorithmic bias.

Presenting platform work as a new type of organising work, not bound by regular employment rules, is a smart way to exploit the risks to achieve the opportunities.

The Commission highlights the findings of the Joint Research Centre that 1,4 % of the workforce provide services as a main job, 4,1 % as a secondary source of income.

The assessment “Nevertheless, most people working through platforms are likely to be engaged in other professional activity outside of platform work, and/or may be active on a number of different digital labour platforms. They may therefore have rights (including in relation to social protection) stemming from another professional activity. However, platform work as a supplementary income, may not open up entitlements to social protection” (p. 15) does not consider the situation of the most vulnerable workers in platforms. The platform companies regularly attract vulnerable groups in the labour market and often exploit people with lower levels of incomes, students and migrants: audiences towards whom it is easy for the platforms to offer low remuneration, otherwise someone else will take the job. Also, it does not consider the responsibility of the platform to observe their social obligations to the public social security systems.

The document reads “In on-location platform work, where people working through platforms can set their own rates and largely determine their own working time, the earnings may be quite predictable. In online platform work where tasks are distributed through contests, earnings may be (relatively) good but may also be unpredictable” (p. 11). Nothing is said about the time on hold or searching for work. According to a recent ILO report, for every hour of paid tasks, workers spend about 23 minutes on freelance platforms and 20 minutes on microtask platforms doing unpaid work;[4] the time these workers are at the disposal of the platform (employer) is to be considered as working time. The key issue of monitoring working time (p.13) remains largely unresolved due to the unwillingness of the platform companies. However, the very nature of the algorithmic tool would make it easy to resolve this issue; time of connection and of activity ‘when the app is on’, would be easy to determine if the platform companies had the willingness – or were required – to take on these responsibilities.

Besides in reality people working through platforms cannot set their own rates and determine their own time.

We can nevertheless observe that the predictability of earnings for on-location platform work should not be overstated. First, unfair low wages remain unfair low wages even if they can be predicted, as a person would not deliberately choose the least paid task between two options and in consideration of the fact that workers are not in a position of strength and are unable to set or demand a different, higher, ‘price’. Secondly, labelling as predictability the mere fact that, say, a driver is made aware of her or his earning or fee at the exact moment a ride is accepted, falls dramatically short of the concept of ‘predictable working conditions’ as understood by workers and the trade union movement. The unacceptable reality for millions of workers working through a digital platform is that each and every day they begin to work, they have no clue at all of the amount of money they will have when they log-off.

Unfortunately, the consultation document is also seemingly oblivious to the OSH risks for on-location work in the light of the pandemic, but also outside. And we fear that the assumption that "physical environment challenges are partially addressed for people working through on-location labour platforms" (p.14) does not sufficiently take into account the risks that the algorithm places on workers by means of the managerial performance tools correctly identified by the document (p.13). It should be noted that if platforms were mere intermediaries, productivity and incentivising workers to do more would not be of any interest to them.

OSH in online work is also an issue – especially when it comes to psychological / mental health / safety issues (task intensity/ glorifying violence content etc.).

I.3 Access to social protection

Challenges related to social protection are correctly identified and should be understood as applying all non-standard workers (including the self-employed), whether work is performed online and/or offline. However once cannot overlook the issue that opening access to social protection for all would be a mere mirage unless the critical question of who will finance it is adequately addressed. Bogus self-employed living with less than the minimum wage is unable to contribute to or finance their social protection. In fact, they are far more likely to be net recipients of welfare state support measures and social security subsidies. Therefore, work needs to be done on ensuring a level playing field with just wages for employees and adequate tariffs for the self-employed. Finally, access to social protection for workers active in platform work as their main occupation is still an issue that remains inadequately addressed in many Member States.

I.4 Access to collective representation and bargaining

The challenges here are correctly identified. Support for social partners to be able to represent the actors in the platform economy will be key to give a future of work with technologies that is compliant with human rights and trade union rights. At this stage, the ETUC would like to emphasise the need for collective bargaining at the following levels:

  • at platform company level, with workers representatives and their trade union,
  • at sectoral level, connected to the sector of activity in which companies are active in, to ensure a level-playing field, respect of health & safety measures and working conditions (a food delivery platform company is thus active in the transport sector for e.g.)
  • at national and transnational level when relevant (rules for algorithm, GDPR, cross-border issues and workers representation in EWC, for e.g.)
  • in respect of the transparency of algorithmic management when its effects fall within the traditional scope of collective bargaining activities.

Collective bargaining is the exclusive competence of social partners, representing employers’ associations/single employers and trade union organisations, in full respect to national labour and industrial practices.

The ETUC also insists on the spill-over effects of platform work, e.g. algorithmic management, a characteristic of platform work, which is spreading to ‘traditional’ workplaces. There is also a need to address the issue of the lack of an “inclusive” workplace: platforms must put at disposal secure and non-intrusive tools of communication to workers and their representatives. The Commission rightly points to the fact that algorithms can carry gender, ethnical or other bias, which needs to be eliminated pro-actively by the platform companies instead of putting the burden of proof on the worker. However, it remains unclear how the Commission wants to address the issue that many labour platforms have “developed invasive, if subtle, forms of modern electronic monitoring” (p. 17). Algorithmic technologies have provided employers with new tools to exercise power in labour relations through control and surveillance, and there is a far-reaching spill-over effect into traditional economy. The Commission overlooks the fact that algorithmic management is spreading from digital platforms to conventional companies and the public sector. The new control mechanisms are used for recruitment, direction, evaluation, performance management, and discipline. Access to algorithms is often restricted, so that it is difficult to counteract algorithmic control.

In page 15, the Commission document characterised platform work as a triangular relationship with platforms that can potentially be “de facto” employers. In that particular case, there is no triangular relationship at all. If there is an employer, then there is only a service relationship between the platform company and either another business or customers but the worker is only performing the task for the company. When footnote 56 points to a quadrangular relationship, this is a complete mistake, a mischaracterisation of the underlying relationships. The worker is not a service provider but a bogus self-employed, and the client and the restaurant are not involved at all in the matching of the supply and the demand organised by the platform company itself.

I.5 Training and professional opportunities for people working through platforms

When talking about training provisions, it is again impossible to speak about platforms without linking them to their sector of activity and without recognising them as companies. The content, financing and transportability of training for workers are widely determined by provisions in the sector of activity and bargained collectively with trade union representatives. When talking about workers in platform companies, the training needs of a worker in a food-delivery platform company or a freelance active in an online app have nothing in common. Training is also often a joint commitment of social partners to provide a quality insertion in the labour market and continuous upskilling of the labour force, leading to career development. On some on-location platform companies, based on low-skill needs, the business model of pressuring labour costs does not allow any quality insertion or career development in the labour market, due to drop-out of fatigued and drained workers. It goes without saying that when one is active for 60 hours a week to make a living, there is little time for training at the end of the “day”. For workers in online platforms, the challenges are rather different. High-skilled workers that are genuine self-employed need to access lifelong learning to maintain their skills. Policies and recommendations concerning arrangements or system of education, lifelong learning, validation, and individual learning accounts must be tailored to fit the needs and the context of each Member State.

Some of these jobs can be considered as a “dead end”. It must be recognised that the low-entry requirement jobs allow some part of the population to access these jobs, but this is the end of the story. There is no perspective of development for these workers, they do not hone new skills, and if they want to change and work for another platform, they do not have any recommendation. The time that these workers are spending working for a platform company as “self-employed” is a time they will hardly build upon.

I.6 Working time directive

Concerning the Working Time Directive, the Commission only refers to case C‑692/19 Yodel judgement in the footnote, but does not elaborate on how the CJEU reasoned on whether the Working Time Directive also applies to workers (claimed to be self-employed) providing their work and services though digital platforms. In its judgment, the CJEU not only pointed to the requirements of being “physically present at the place determined by the employer and that they are available to the employer to provide the appropriate services immediately, if needed”. In addition, it also stated that such “self-employed” persons can only be excluded from the Directive on the condition that “the independence of that person does not appear to be fictitious and, second, it is not possible to establish the existence of a relationship of subordination between that person and his putative employer” (§ 45)[5].

The notion of working time is a subject of several ECJ-decisions and it is clear that a worker, even if he/she is waiting for the task but staying under the direction /disposition of the employer, is within working time. The problem lies in the enforcement of working time rules.

I.7 Court cases

Whereas there are some crystal-clear Court decisions, the Commission continues to declare that “To date, jurisprudence has not removed possible legal uncertainty of employment status for people working through platforms, or more broadly for people who are misclassified as self-employed.” (p. 9). However, as the document itself recognises (p. 25), in a growing number of Member States, employment tribunals and courts (including some Supreme Courts) have had no qualms at establishing that a clear employment relationship exists (and usually one of full employment). While some more junior courts have provided contrasting opinions, there is no national supreme court that has recognised these relationships as amounting to genuine self-employment.

The Commission seems to regret that some “bottom-up initiatives in a few Member States” including collective agreements are narrow in scope, but it forgets to say that Competition Authorities in line with EU competition law invalidated some collective agreements, which has had a severe setback on the spread of collective bargaining with platform companies (p. 10). The Commission mentions Hilfr but leaves out the fact that the Danish competition court invalidated the agreement. The Commission underlines that this “consultation does not address the issue of potential impacts of EU competition law” (p. 20). However, the consultation includes “collective bargaining aspects in platform work that go beyond the competition law dimension” so to “support social partners’ coverage of platform work”, which is welcome. The clarification of the objective of not bringing collective bargaining within the scope of competition law seems helpful, although the narrowing down on ‘certain solo self-employed’ is undesirable as it excludes some categories of self-employed persons that should be covered by collective agreements.

The ETUC believes that the examples of national case law provided in the consultation document have been selectively identified to provide a false perception of court rulings equally in favour of both workers and platforms. In the last two months there have been major landmark court cases calling platforms on the reclassification of their workers throughout Europe.[6]

With regards to the three cases of CJEU jurisprudence, although none of these references to the CJEU directly concern the labour dimension, their contribution to the discussion on the responsibility of platforms might be more important than what the Commission outlines in their brief “analysis”. The logic of these CJEU rulings is basically as follows: if the intermediation is not a passive information society service, but closely linked to the provision of a physical service over which the platform exercises control, then the platform is liable to comply with sectoral rules (e.g. on transport). So, by analogy, one could wonder – if the platform company is responsible for the physical service, should the platform company not be responsible also for the person who provides the service, i.e. the worker?

In C‑434/15 Elite Taxi v Uber, the CJEU held that “Uber exercises decisive influence over the conditions under which that service is provided by those drivers. On the latter point, it appears, inter alia, that Uber determines at least the maximum fare by means of the eponymous application, that the company receives that amount from the client before paying part of it to the non-professional driver of the vehicle, and that it exercises a certain control over the quality of the vehicles, the drivers and their conduct, which can, in some circumstances, result in their exclusion.” (§ 39) Given the clear control exercised in this case, Uber was held liable for the transport service.

In C‑320/16 Criminal proceedings against Uber France, the CJEU referred to its judgment in Elite Taxi and once again confirmed that “The same holds true, for the same reasons, with regard to the intermediation service at issue in the main proceedings” (§ 24). Given the clear control exercised in the case, Uber was held liable for the transport service. However, in C‑62/19 Star Taxi the CJEU compared the facts at hand with the previous two cases and on the contrary concluded that the Star Taxi app was different as it did not exercise the same degree of control and therefore "constitue un « service de la société de l’information », au sens de ces dispositions, un service d’intermédiation consistant, au moyen d’une application pour téléphone intelligent, à mettre en relation, contre rémunération, des personnes qui souhaitent effectuer un déplacement urbain et des chauffeurs de taxi autorisés, pour lequel le prestataire dudit service a conclu à cette fin des contrats de fourniture de services avec ces chauffeurs en contrepartie du paiement d’un abonnement mensuel, mais ne leur transmet pas les commandes, ne fixe pas le prix de la course ni n’en assure la perception auprès de ces personnes, qui paient celui-ci directement au chauffeur de taxi, et n’exerce pas davantage de contrôle sur la qualité des véhicules et de leurs chauffeurs ainsi que sur le comportement de ces derniers." (§ 55) In other words, in this case the CJEU held that the app had more of a passive and intermediating nature, and therefore this platform was not responsible in the same way as in the Uber case.

I.8 Additional considerations

Calls against restrictive regulations or complaints about the uncertain regulatory situation cannot be used as an alibi to avoid responsibility. The challenge Europe faces today is not the unpredictability hampering the ability of platform companies to innovate and expand. The task at hand is that of addressing the vulnerability of millions of workers emerging from both the lack of enforcement of existing regulations and from absence of adequate rules. Exploitative business models cannot be preserved under the illusion of economic growth or the illusion of taking workers out of undeclared work. With sound regulations, existing platform companies will either have to adjust their business practices to comply with the rules, or other more virtuous platforms will develop and occupy their share of the market.

The possibility of a European initiative on workers in platform companies should be built on sound scientific background and a rights-based approach related to the social and economic impact of digital labour platforms. Platforms have been using a series of arguments in their advocacy strategies in the hope of avoiding any regulation which may hinder their business model, based on setting the risk on the shoulders of the workers and operating in a legislative vacuum to maximise benefits. Some of these claims are mentioned hereunder, but we stress that none of them is substantiated by facts or methodologically robust empirical evidence:

Food delivery apps and other delivery platforms have been spreading the myth that they are saving restaurants and commerce, when instead they are charging substantial fees that are eating into the already reduced margins of the sector.

Platforms have been lobbying about the need to avoid a "one-size-fits-all" legislative approach, considering the variety of platforms operating in the labour market. The ETUC refutes this argument. Labour legislation provides a basis of the rights for workers, not a "one-size-fits-all" burden. It should be recalled that the complexity of the labour market in matters of sectors and occupations has never prevented the possibility of providing tailored solutions to its different realities through social dialogue and collective bargaining.

Whereas platform companies alleged that they are creating jobs and economic growth, accurate information on the qualitative and quantitative impact of these companies in terms of jobs and gross and net amounts is still missing[7]. Besides, this analysis should also consider the tax revenues and contributions to the social security systems evaded by platform companies as well as the decent jobs that are often ‘priced out’ of various markets and are most of the times replaced by the less protected, and comparatively cheaper, forms of work provided by digital platforms.

The narrative that workers on platforms appreciate their freedom, and this can only be satisfied by signing commercial agreements with them as self-employed workers is based on a highly distorted vision of reality. The employment status can provide for the flexibility in terms of working time that may be requested by the workers. Recent research[8] indicates that these workers perceive themselves as employees, a fact that is further corroborated by the growing and increasingly successful number of cases brought before national tribunals and courts by nominally self-employed workers requesting to be reclassified as employees or workers.

We therefore ask the Commission to provide serious background studies in the impact assessment scheduled with the second stage consultation, if some of these arguments are to be considered important for policy proposal.

II. Do you consider that EU action is needed to effectively address the identified issues and achieve the objectives presented?

The comprehensive overview of the relevant EU legislation in the Commission document demonstrates that “existing EU-level instruments only partially impact the challenges posed by platform work” (p.24), notably the employment status, underlining the need for EU action. On issues such as surveillance, direction and performance appraisal, and algorithmic management, EU law is either lacking or, where it exists, it fails to provide specific and sufficient protection.

In addition to this the ETUC believes that, as platform work is an international phenomenon with a strong cross-border dimension, it would be a step in the right direction to establish a European framework to ensure a level playing field, equal treatment and avoid unfair competition. The European framework, however, must take into consideration the variety of industrial relation systems and also the need not to undermine their fundamental institutions. The ETUC notes that the document makes no reference to the fundamental principle that platforms (regardless of their location for company law purposes) must abide with the laws of the country where the service takes place, a point that is increasingly being accepted by case law of the CJEU.

The added value of EU action is obvious at several levels:

  • EU action could improve working conditions for workers in platform companies in the EU;
  • the cross-border dimension of platform work makes a common EU approach appropriate to establish a level-playing field and to avoid fragmentation;
  • national regulatory differences in platform work may induce regime shopping for the platform companies choosing the least regulated Member States to drive their business, the patchwork situation would continue;
  • addressing working conditions in platform work is a first condition to create a level playing field and avoid unequal treatment of workers, and reducing the risk of cheaper precarious labour that substitutes decent and well remunerated employment;
  • the initiative can strengthen the European way of life and bring about strong workers’ rights; and
  • EU action addressing the misclassification of the employment status would bring many people working in platforms within the scope of EU legislation.

III. If so, should the action cover all people working in platforms, whether workers or self-employed? Should it focus on specific types of digital labour platforms, and if yes which ones?

First of all, the EU action should be based on a presumption of an employment relationship complemented by a reversed burden of proof in respect of the recognition of platforms as companies with all the obligations this entails. The ETUC is in favour of defining the employing entity that would impose labour law obligations on the party that in practice largely determines the terms and conditions. The employer should be the party that largely determines the terms and conditions of engagement or employment of a worker. This action should target all digital labour platforms, including on-location and online ones. This will allow for digitalisation with a rights-based approach. It should prevent sham constructions from being legalised and should also prevent (digital) platforms from being given a separate status. A separate status is unnecessary and undesirable.

Additionally, the added value of EU action would consist in guaranteeing fair working conditions for all types of non-standard workers and workers on platform companies (including the self-employed) regardless of their employment status, and would ensure fair competition within the internal market. The level of rights which would also apply to self-employed workers is a national competence, which should be decided in each Member State with the participation of social partners.

IV. If EU action is deemed necessary, what rights and obligations should be included in that action? Do the objectives presented in Section 5 of this document present a comprehensive overview of actions needed?

In this section, the ETUC provides a comprehensive answer about the regulatory action at European level that is needed for each of the challenges identified. The potential European action should respect national industrial relation systems and collective bargaining systems. The challenges are the following: addressing misclassification of employment status in platform work; ensuring fair working conditions for all; guaranteeing protection against economic and social risks for workers in platform companies; promoting an approach to automated decision-making in platform work based on transparency, human oversight and accountability and full respect of data protection rules; addressing access to collective bargaining and to collective rights; promoting cross-border fairness in platform work; and equipping workers in platform companies with the tools to steer their career and have access to professional development. Lastly, the ETUC also provides an answer concerning the personal scope of the EU initiative and the EU instruments which should be considered. The answers provided are articulated around the seven priorities of the ETUC explained at the beginning of the reply.

The EU actions presented are quite comprehensive. As mentioned earlier in the key challenges arising in the platform work context, the failure to recognise platforms as employing companies with all the obligations (including the sector specific ones) it entails, would weaken any initiative and distort the situation even more.

Hereunder are the ETUC demands for the upcoming legislative initiative:

IV.1 Addressing misclassification of employment status in platform work

The ETUC wants to put an end to the misclassification of workers, which deprives them of their rights. The direction exercised by a platform company may differ from the traditional way as it is mediated via a digital tool, the platform. What is relevant for the purposes of establishing employment status, however, is not the intention of the company, but the actual design of the employment relationship. A presumption of an employment status should be the starting point. It should be complemented by a reversal of the burden of the proof for platform companies seeking to establish that they are not the worker’s employer. A worker who performs work under the same conditions as "normal" workers (meaning a worker with a full-time open-ended contract in the sector of activity of the platform) should be classified as such according to the definitions used in the respective industrial relation systems. Many digital platforms exert the prerogatives of employers and, in these cases, the application of this regulation should make it clear that if a platform acts as an employing entity, by largely determining the terms of engagement or employment of a worker, then the workers will also be presumed to be employees until the platform proves the opposite.

These two instruments (presumption of employment relationship and reversal of burden of the proof) don’t need changes in the definition of a worker set by Member States nor do they express a preference, or otherwise, in respect of particular employment tests or indicators or the definition of subordination criteria. It thus respects the principle of subsidiarity and the autonomy of social partners.

The presumption of employment relationship means that any natural or legal person (e.g. labour platform) who has responsibility for the undertaking and/or the establishment, is considered to have an employment relationship with the worker. Under this condition, the labour platform shall grant its workers all the existing employment rights. The reversal of the burden of proof places the responsibility to prove that the presumption of employment does not apply on the employer/company’s side. Therefore, if the labour platform has been hiring genuine self-employed, they can prove it and their business model will not be affected. However, if the platform cannot reverse the presumption of employment, it means that workers have been employed by them from the beginning. Thus, the reversal of burden of the proof only shifts the weight of the burden of proof from the most vulnerable in the labour relationship (the worker) to the most able to prove the opposite (the company). Their vulnerabilities have been underlined by previous research showing significant challenges for workers in a bogus self-employment situation to start legal action.

The administrative access and the barriers to submit a claim are considerable[9],  even more for vulnerable workers. Indeed, it is difficult for individual workers to assert employment and/or to insert their employment rights considering the threat of deactivation. Without proper protection and without Trade Union support, it is unfeasible to bring a claim while still working. The vulnerability of the workers should also be taken into consideration: the people who would need to start a legal action to have a recognition of the rights they are entitled to are the ones in a situation of weakness in their relationship with the employer. Some of them are concerned with meeting the needs of the day or of the week; courts are the last thing that will cross their minds. The uncertainty of the courts discourages attempts to take action. Moreover, the platforms, via the data of the algorithm, are in the best position to prove or not the existence of the employment relationship Therefore, the resources of the Labour inspectorates and other labour authorities should be supported to provide for a greater involvement in the administrative enforcement of the legislation.

For all these reasons, there is a need to adopt the presumption of employment and a reversal of the burden of proof. Limiting the action to easing the burden of proof by setting criteria to address misclassification would still require the worker to start legal action and this is not appropriate. It would facilitate the case for workers who are taking that route, but it will not address the structural problem. Systems of enforcement should never rely on the worker to bring individual claims. Furthermore, the platforms, through the algorithm, are the best placed to prove the absence of an employment relationship, if it is the case.

In page 27 of the consultation document, the Commission states that "EU action could also aim at facilitating the enforcement of existing labour legislation and strengthening controls and inspections of digital labour platforms, to detect and pursue possible case of misclassification". Firstly, the ETUC would like to state that the reversal of the burden of proof will facilitate the action of the labour inspectorate. This said, the labour inspectorate and other forms of administrative enforcement offer advantages in terms of a comprehensive solution at company level, whereas court decisions so far seemed to only have offered economic compensation for the individual litigants (and not to all workers of the platform), and no changes in the business model of platform companies either. The ETUC would like to point out, however, that Labour Inspectorates are already challenged by the complexities brought by digitalisation and the fragmentation of the workplaces[10]. They should therefore be better equipped in terms of human, technical and financial resources to undertake this new task.

IV.2 Ensuring fair working conditions for all

The new Directive 2019/1152 on transparent and predictable working conditions is a result of the European Pillar of Social Rights. Once the Directive will be transposed in EU Member States, all workers will have the right to more comprehensive information on essential aspects of work, which they will receive promptly in writing. This includes: a limitation of the duration of probationary periods at the beginning of employment; the right to seek additional employment, with a prohibition of exclusivity clauses and a limitation of incompatibility clauses; the right to knowing a reasonable period in advance when work is taking place, for workers whose working schedules are very unpredictable, such as in the case of on-demand work; the right to anti-abuse legislation for zero hour contract work; the right to receive a  written reply to a request to transfer to another more secure job; and, the right to receive free mandatory training that the employer  has a duty to provide. The Directive has a wide personal scope. It aims to ensure that these rights cover all workers in all forms of work, including work in platform companies.

Work in digital platform companies involves risks such as exposure to electromagnetic fields, visual fatigue, musculoskeletal problems and other health risks related to specific sectors and to Covid-19. Psychosocial risks include isolation, stress, technostress, technology dependency, information overload, burnout, posture disorders, online harassment, and overall precarious working conditions. Finally, job insecurity, which is known to contribute to the overall poor health of non-standard workers and workers in platform companies (including the self-employed), is a characteristic of working on an online platform. These risks would make the enforcement of OSH regulations for work carried out through online platform companies of the highest importance. The application of OSH rules and labour law in general is contested by the platform companies, as the involvement of online platforms in work organisation tends to complicate the classification and regulation of responsibilities for the work in question. Rules and practices of the host country where the platform work is performed should apply. National labour inspectorates shall develop tools and strategies to effectively enforce existing OHS rules and labour law at the place of work.

In matters of the application of the regulation (EU) 2019/115095 on promoting fairness and transparency for business users of online intermediation services (the so-called ‘Platform-to-Business’ or ‘P2B’ regulation), the Commission seems to forget to mention that the scope of this Regulation is limited to “information society services”. In other words, since according to the case law of the CJEU platforms such as Uber do not provide information society services, but physical transport services, this Regulation does not cover Uber.

Similarly, the scope of the proposed Digital Services Act is limited to online intermediation services, which are defined as “information society services”. The Commission proposal implies that platforms such as Uber will not be covered, since they provide physical services and not information society services.

Conclusion 1: Digital labour platforms are not online intermediaries, but companies, and therefore they should also be considered employers. Recognising platforms as companies allows to determine the obligations they have as employers when the employment relationship applies.

Conclusion 2: If platforms such as Uber fall outside the scope of both the P2B Regulation and the DSA, they must be regulated as nothing less than employers through the up-coming initiative on workers on platforms. A situation where all online platforms are regulated except for digital labour platforms - which continue to operate in a vacuum – is unsustainable and unjustifiable.

IV.3 Guaranteeing protection against economic and social risks for people working through platforms

Platform companies transfer the costs of the social protection that they are not granting to their workers, to society as a whole. The present situation implies that companies that use the ordinary and proper employment relationship are subsidising the platform companies; if this becomes more mainstream it would put enormous pressure on the sustainability of the redistributive institutions that characterise the welfare state. We cannot ignore the legislative loopholes that permit businesses to elude their legal and societal obligations de facto social protection for all workers, including non-standard workers. As regards to the scope of social protection for non-standard workers and workers in platform companies (including the self-employed), a comprehensive approach should be taken in which non-standard workers enjoy the same protection as ordinary workers. An ‘à la carte’ solution would ultimately lead to discriminatory practices against specific groups in society, thus jeopardizing the social acquis and the project of an equitable society where everyone has the right to live and work in dignity.

Many workers in platform companies are undocumented migrant workers as well as asylum-seekers. Their precarious and vulnerable position needs to be taken into account in any EU initiative

IV.4 Promoting an approach to automated decision-making in platform work based on transparency, human oversight and accountability, and full respect of data protection rules

Information, consultation and participation rights at EU level must be respected, thus granting workers and trade unions access to the algorithms of digital platforms. The workers should know what data is gathered, why it is collected, where it is stored and how it is used to control their labour. This access to data should be permitted in the country where the services of the platform are provided and in the language of the worker. 

The democratic control and transparency of the operation of the algorithm of intermediate work applications (including rating of workers) and platforms, the application and enforcement of the right to disconnect, and the protection of the data of workers must be at the heart of the public debate on digitalisation.   They should also be discussed through information, consultation and participation of workers in full compliance with principles of non-discrimination. The platform companies’ decision-making processes concerning workers’ grievances must be under human control. The European Commission and the Member States should promote the creation of a public register that displays a complete list of online platforms companies. The general employers’ obligations of a company should be a requisite for their operation in an EU member state.

IV.5 Addressing access to collective bargaining and to collective rights

The misclassification of workers in platform companies as "independent contractors" limits their collective representation, as this status is often considered incompatible with trade union membership. Collective representation should be possible also for this category of workers regardless of their employment status. Although most digital work platform companies are unsurprisingly hostile to any effort to organise workers representation, some models of collective representation of workers on platforms are emerging.

Trade Unions shall be given digital access rights to communication channels between the app and the workers on platforms and to get directly in contact with these workers. As real meetings become more challenging, virtual communication and mobilisation networks become more important.

Non-standard workers and workers in platform companies (including the self-employed) represented by a trade union when bargaining collectively must not be considered undertakings for the purpose of competition law. Trade unions are not cartels and collective agreements are not agreements between undertakings resulting in anti-competitive business practices. Wage setting should never be seen as price-fixing. Minimum wage conditions are also required for solo self-employed (workers). To this end, it is particularly necessary to improve the legal options for solo self-employed workers to conclude collective agreements in order to achieve industry-specific minimum wages.

It is good that the Commission makes a clear distinction between the social policy initiative on workers of platforms and the competition policy initiative on self-employed and their access to collective bargaining. Mixing these two initiatives up would imply that all workers on platforms are self-employed. Any worker on a platform who can be qualified as employed or falsely self-employed has the already right to bargain, and therefore does not face tensions between competition and collective bargaining. 

IV.6 Promoting cross-border fairness in platform work.

Border cooperation between the labour inspectorates will be of utmost importance. The European Commission’s first phase social partner consultation document on addressing the challenges related to working conditions in platform work (24 February 2021), raises a number of cross-border aspects, including issues such as jurisdiction, applicable law, taxation and social security.

Below are some observations and thoughts regarding labour relations and social security coordination in transnational situations. However, these points do not address the issue of taxation. (Whereas income taxation for workers on platforms is likely to be regulated by applicable local laws or by bilateral taxation agreements between Member States in cross-border situations, the taxation of digital platforms is a very different issue).

To clarify how the reversal of burden of proof mechanism can work in a cross-border situation, we ask the Commission to make clear that the possibility to reverse the presumption of employment relationship should be made in the country where the worker operates, based on the legislation of that country. This to avoid that the presumption is rebutted and a worker considered as a self-employed based on criteria for employment status in a country A whereas this worker operates in country B where the criteria for employment status are different (and where the presumption could not be rebutted based on this country’s criteria). Another reason is that presumption will be rebutted based on the real working conditions, and those can only be investigated and stated in the country where the work takes place.

IV. 7 Equipping people working through platforms with the tools to steer their career and have access to professional development

Workers in platform companies should have access to the same rights to employee training and/or lifelong learning, and to public employment services (PES), that their status gives them access to in different Member States.

We agree with the consultation paper that irrespective of employment status, workers in platform companies should benefit from support for training and upskilling. The Commission should also respond on how to ensure the right to training to workers under the effective implementation and monitoring of the first principle of the European Pillar of Social Rights. Access to trainings offered by public employment services could be ensured via the implementation of the Upskilling Pathways initiative. In order to improve access to trainings, workers could be supported by the PES and other validation agencies to validate their skills and competences.  

As the consultation paper mentions, over 80% of job-related training programs in the EU Member States are employer sponsored, employers take their fair share to support upskilling and reskilling the workers. This has been helped by the implementation of the EU Social Partners agreement on digitalisation which says that trainings in relation to job-related skills need to be paid by the employers.[11] Also, employers of workers on platforms could join the Pact for Skills to engage for upskilling their workers.

However, it would be necessary to prioritise them so that the most important actions come first and the less important one can follow at a later stage.

IV.8 Several options could be envisaged for the personal scope of the EU initiative

The ETUC would be in favour of applying the EU action to all non-standard workers and workers in platform companies (including the self-employed).

The ETUC would be in favour of an EU action including all platforms.

IV.9 A range of EU instruments could be considered in the preparation of such an EU initiative

The ETUC is convinced that an exchange of good practice (mutual learning), provision of guidance or monitoring the development of platforms are no longer viable options. Only a legislative framework can achieve the goals outlined in the Commission’s document.

V. Would you consider initiating a dialogue under Article 155 TFEU on any of the issues identified in this consultation?

As most of the objectives laid down in the Commission’s document can only be achieved through legislation and as BusinessEurope refuses since more than a decade to enter discussion on legal frameworks, the ETUC is most reluctant to start social dialogue proceedings that might end after many delays in a voluntary agreement ,which is unable to fulfil the expectations of people working on platform. Another argument is that most platform companies deny that they are employers and are not members of employer association; a voluntary agreement with a traditional employer association would not cover many platform companies and so its implementation would fail. Moreover, most recent court cases acknowledge that there is an employment relationship.

Last but not least, during the social dialogue on digitalisation it was impossible to raise the question of workers in platform companies due to a rigorous opposition by BusinessEurope. Under these very specific conditions it would be preferable that the European legislator takes the initiative to avoid lengthy empty talks, which will only delay legislation and harm workers in platform companies instead of giving them the necessary support.

Were platform companies to observe their responsibilities as employers and be affiliated to the recognised European employers’ organisation, the ETUC would be open to consider engaging in fruitful social partners consultation in virtue of the Fundamental Treaty of the European Union.



Existing instruments for determining applicable law in cross-border situations

This table develops some arguments around the scope of applications for platform companies of cross-border EU legislation, with a focus on the applicable law on employment contracts, the access to justice, and the coordination of social security.

As a general rule, the aim of this initiative shouldn’t be to create new rights or categories (‘third category’ of worker) but to empower workers on platforms to claim their legitimate rights and hold their de facto employers (the platform companies) accountable. Against this background, the suggestion, in the consultation document, to elaborate interpretation and guidance regarding existing EU legislation regarding the implications of cross-border platform work, seems reasonable. The existing rules should apply equally to online and offline cross-border situations. As a general rule, any EU action should depart from the presumption of an employment relationship between the worker and the platform company to ensure coherence and legal certainty.

    1. As regards the freedom of movement and applicable legislation, we should first ask ourselves who is operating cross-border and in what kind of activities. When it comes to the free movement of workers, the applicable rules are governed by the Member State of destination or residence of the worker.
    2. When it comes to the free movement of services, it depends on whether the service is delivered in a given location or digitally. Digital services (information society services) traditionally follow the country-of-origin principle, whereas physical services are governed by the rules of the country of destination.
      • However, in this regard, it should be noted that the Commission’s recent proposal for a Digital Services Act (DSA) also strengthens the country-of-destination principle, by aiming to make sure that destination Member States are able to define what constitutes illegal or harmful services in line with national rules that apply to offline content, products or activities on their territory.
      • Also, when it comes to platforms established outside the EU, such challenges could perhaps be solved by introducing similar requirements as set out in the DSA, such as the obligation to have a legal representative in at least one EU Member State.
    3. In accordance with the CJEU case law, digital labour platforms such as Uber do not provide information society services, but physical services such as transport. In other words, such a transport service is governed by the national/regional/local rules in the country of destination.
    4. Consequently, any attempt by digital labour platforms to enjoy benefits based on the country of origin (e.g. more favourable rules in the country where the platform is registered) should be rejected. This should hold true when it comes to the regulation of not only services, but also employment relationships.

Assuming there is an employment relationship between the platform and the individual provider of labour, this raises the question of applicable law. Is it the law of the Member State where the platform is registered? Or the Member State where the worker works or lives?

  1. The main rule of the Rome I Regulation, when it comes to applicable laws on employment contracts, is the law chosen by the parties (on the condition that this does not lead to abuse). Consequently, in the case that platform companies are considered as employers, they should not be allowed to apply any jurisdiction, such as the jurisdiction of their legal seat to the employment contract, and especially in case the employee has no connection to that country. Any EU rules protecting workers working in the platform economy should be formulated in a way that cannot be derogated from by agreement with the employer.

Article 8 – Individual employment contracts

1.   An individual employment contract shall be governed by the law chosen by the parties in accordance with Article 3. Such a choice of law may not, however, have the result of depriving the employee of the protection afforded to him by provisions that cannot be derogated by agreement under the law that, in the absence of choice, would have been applicable pursuant to paragraphs 2, 3 and 4 of this Article.

2.   To the extent that the law applicable to the individual employment contract has not been chosen by the parties, the contract shall be governed by the law of the country in which or, failing that, from which the employee habitually carries out his work in performance of the contract. The country where the work is habitually carried out shall not be deemed to have changed if he is temporarily employed in another country.

3.   Where the law applicable cannot be determined pursuant to paragraph 2, the contract shall be governed by the law of the country where the place of business through which the employee was engaged is situated.

4.   Where it appears from the circumstances as a whole that the contract is more closely connected with a country other than that indicated in paragraphs 2 or 3, the law of that other country shall apply.

  1. Similarly, the Brussels I Regulation aims to ensure access to justice in cross-border situations by regulating issues such as jurisdiction and the recognition and enforcement of judgments in civil and commercial matters. These rules must be used so as to empower workers to bring cases against their employer also in situations of cross-border conflicts. They must not be deprived of any such rights that standard offline workers already have.
  2. The consultation document rightly points out that both the Rome I Regulation (Article 8) and Brussels I Regulation (Article 23) ensure that for employees, the choice of jurisdiction and/or law applicable to their individual employment contract cannot lead to them being deprived of protections that they would otherwise have in the absence of such choice in their contract. For this reason, it is important that the future EU rules on platform work clearly recognise workers as employees with all the rights that follow from this employment status. It should not be possible to derogate from such an employment status or from certain rights “through an agreement” between the employer and the worker.

When it comes to social security coordination in cross-border situations, the 883/2004 Regulation on the coordination of social security systems applies to activities of both employees and self-employed persons.

    1. As a general rule, according to Article 11(3)(a), any person pursuing an activity as an employed or self-employed person in a Member State shall be subject, in the first place, to the legislation of the State where the economic activity is effectively pursued.
    2. Other social security coordination rules may of course become applicable, in case also the worker himself is moving across border, such as in the case of frontier, posted, seasonal or multi-activity workers.

The consultation document also mentions challenges linked to the applicability of the principle of non-discrimination of EU nationals in cross-border situations, in case there are no local workers to compare with in order to determine what should be considered to constitute equal treatment in a given situation. Should such a situation arise, any possible conflict should primarily be solved by using the normal social security rules, presuming that the worker in a platform company is an employee. In any case, the basic rule must always be that someone exercising their freedom of movement (whether physically or digitally) must never be treated less favourably only for this reason.

As regards the enforcement of labour and social security rights in cross-border situations, cooperation and coordination is key. In this regard, the European Labour Authority is competent at least when it comes to social security coordination, as well as for the tackling of undeclared work (e.g. the European Platform tackling Undeclared Work has recently held discussions on platform work from the perspective of undeclared work, as regards labour rights, social security and tax).

However, enforcement may encounter not only legal challenges, but also practical obstacles as a result of the digital ecosystems in which platforms operate. E.g the Swedish Occupational Health and Safety Authority wanted to inspect the working conditions at Uber and Uber Eats, as part of an investigation of whether Uber should be classified as an employer. The inspectors went to the Uber headquarter in Stockholm to request more information, but it turned out Uber was under no legal obligation to cooperate, because the tasks distributed to the drivers/riders were in practice/technologically distributed by another online company, registered in the Netherlands. This demonstrates that the issue of jurisdiction might be an obstacle not only for legal proceedings or the protection of working conditions, but also when it comes to conducting labour inspections.


[1] Non- standard forms of work may, for instance, comprise temporary work, part-time work (especially part-time work performed on an involuntary basis), on-call work and contracts with zero- or variable working hours, casual work, agency work, digital platform work and disguised/dependent self-employment.

[2] As Shawn Carolan, early Uber investor, wrote in an op-ed stressing the potential to spread Prop 22’s (California ballot of Nov 3, 2020 defining workers in platform companies as independent contractors) vision of work “from agriculture to zookeeping,” including to “nursing, executive assistance, tutoring, programming, restaurant work and design.”

[3] Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions - A European agenda for the collaborative economy COM/2016/0356 final, available at: .

[4]  World Employment and Social Outlook 2021: The role of digital labour platforms in transforming the world of work. International Labour Office, Geneva: ILO, 2021, available at:

[5] As a matter of law, the Working Time Directive, as most Health and Safety instruments, has a broad personal scope of application, as also evidenced by CJEU decisions such as C-428/09, Union syndicale Solidaires Isère; Case C-316/13, Fenoll; and Case C-518/15, Matzak.

[6]; see also German Federal Court, 1. Dec. 2020, 9 AZR 102/20 which classifies a crowdworker as an employee (dependent worker)

[7] DE CALIGNON, Guillaume (2021), Des créations d'entreprises record grâce aux petits boulots. Les Echos, available at:

[8] Urzì Brancati, C., Pesole, A., Fernández-Macías, E. (2019), Digital Labour Platforms in Europe: Numbers, Profiles, and Employment Status of Platform Workers, EUR 29810 EN, Publications Office of the European Union, Luxembourg, 2019

[9] Kirk E., Contesting ‘bogus self-employment’ via legal mobilisation: the case of foster care workers. (2020) Capital and Class, 44(4), pp. 531. (doi: 10.1177/0309816820906355)  

[10] Opinion on future EU OSH Enforcement Priorities contributing to a renewed EU OSH Strategy, A submission from the Senior Labour Inspectors' Committee (SLIC); EU OSHA (2018) Foresight on new and emerging occupational safety and health risks associated with digitalisation 2025 (p.14).