The new EU Directive on Transparent and Predictable Working Conditions in the context of new forms of employment

Abstract

The article examines the reach, protective effects and limitations of the recently adopted European Union (EU) Directive on Transparent and Predictable Working Conditions. After explaining the need for a new instrument, the article analyses the Directive’s protective provisions. Cases of the European Court of Justice are presented to provide the wider context and explain how the EU social acquis impacts upon the implementation of the Directive. Finally, new developments in the EU labour and social field are discussed, making recommendations of possible avenues for providing protection to a larger category of workers.

Introduction

On the 13th of June 2019, Directive 2019/1152 on Transparent and Predictable Working Conditions (‘DTPWC’) was adopted by the Council. The news that the European Union (‘EU’) had passed legislation on the social field after 20 years of inactivity quickly circulated the internet, sparking excitement about European developments. As Calvet Chambon, the Spanish Rapporteur responsible for the negotiations, said:

‘[t]his is for the people of on-demand contracts, zero-hour contracts, Uber, Deliveroo, Cabify, these kinds of new jobs. The clickers for instance that work at home (or) at different times. […] Thanks to this Directive […] all these workers who have been in limbo will now be granted minimum rights [and], from now on, no employer will be able to abuse the flexibility in the labour market’ (Multimedia Centre, 2018).

How far does the protection of the DTPWC extend? Has it succeeded in updating the European labour framework? What are its shortcomings and how can they be mitigated? This article sets out to examine the recent DTPWC with a view to determining its potential impact upon EU labour and social protection legislation. After giving a brief overview of how the Directive came about and the reasons that led to its adoption (The Need for Regulation), the article critically discusses the Directive’s protective provisions (The Personal Scope of the DTPWC: Reach, Protective Effects and Limitations). Particular emphasis is paid to the personal scope of the DTPWC and its deficiencies. In this context, cases from the European Court of Justice (‘ECJ’ or ‘the Court’) are analysed to demonstrate the social acquis in the area and explain how it impacts the implementation of the Directive. Finally, the article discusses recent regulatory developments in the labour field and proposes ways forward.

The need for regulation

The previous instrument regulating workers’ right of information was the Written Statement Directive (‘WSD’) 91/533/EC. Adopted back in 1991, it stipulated that employers have to provide employees with core information on the conditions applicable to their employment relationship within 2 months of their starting date. That instrument, however, allowed for a lot of leeway and was fit for a different era when most workers satisfied the archetypical paradigm of an employee who works under an open-ended contract of employment for one principal throughout their life. As the REFIT evaluation (European Commission, 2017b) revealed, the WSD was no longer effective due to the following reasons: (i) the 2-month deadline impeded transparency, increasing the potential for undeclared work and abuse of employee rights; (ii) the enforcement mechanism was ineffective as it did not regulate means of redress or sanctions for non-compliance and (iii) the information package prescribed was insufficient. Particular problems were created by the lack of information regarding precise working time.

The most pressing issue, however, identified in the REFIT document was the lack of clarity regarding the categories of workers covered by the Directive. As the Impact Assessment noted, ‘the EU labour law acquis, including but not limited to the WSD, does not apply uniformly to all workers, creating disparities and leading to inequalities in terms of working conditions and protection in general’ (European Commission, 2017c: 17). For instance, even though positive developments like the ones in Danosa (C-232/09) and Balkaya (C-229/14) suggest that senior managerial executives should be treated as ‘workers’ for the purposes of the Pregnant Workers (92/85/EEC) and Collective Redundancies (98/59/EC) Directives, comparative legal research shows that people in management functions are not considered employees in Sweden (European Commission, 2017b). Moreover, despite the ECJ’s rulings requiring Member States (‘MS’) to apply the Directives consistently across the private and the public sector (C-212/04, Adeneler; C-53/04, Marrosu and Sardino), public servants were found not to fall within the definition of employee in Lithuania and Austria (European Commission, 2017b). Finally, while trainees (C-229/14, Balkaya) and researchers on a doctoral grant (C-94/07, Raccanelli) have at times been treated as workers for the purposes of free movement and equal treatment provisions, there is still a long way to go before these and other categories of vulnerable workers (i.e. apprentices and domestic workers) receive any labour law protection worthy of the name.

While, the EU’s competence in the social policy field is limited to the adoption of minimum requirements for gradual implementation (Article 153 TFEU), it is evident that the vast heterogeneity of national approaches apropos who is covered by the Directive can lead to the invigoration of social dumping practices and the creation of two-tier markets within the EU. As the Impact Assessment noted,

‘where the regulatory framework is weak and patchy across MS […] there is a risk of race to the bottom in standards applying to new forms of work. [The] increasingly divergent and even contradictory national solutions, create regulatory loopholes when viewed from an EU perspective, leading to inequality in the protection of workers and their living conditions. Eventually it could affect the quality of the workforce, the relative competitiveness of employers, companies and MS, and the functioning of the EU internal market’ (European Commission, 2017c: 30).

The uncertainty regarding coverage has been exacerbated in recent years by the emergence of new and atypical forms of work. New types of employment have been introduced (or have become more prevalent) such as zero-hours contracts, employee sharing, ICT-based mobile work, voucher-based work, interim management, portfolio and crowd work (Eurofound, 2015). While the standard contract of employment still remains the predominant form of work organisation, its social importance has been declining as an increasing number of firms are resorting to franchising and outsourcing. The data are revealing: half of all the new jobs created in the last 10 years have been non-standard, leaving over 25% of the EU workforce engaged in casual forms of work (European Commission, 2017c). The number of workers on contracts lasting less than a month, for instance, has increased from 373,000 in 2002 to 1.3 million in 2016, with a large number of people (3.8 million) working less than 8 h per week (European Commission, 2017c).

The advent of the platform economy has made matters even more complicated. Millions of on-demand workers are nowadays channelled through online platforms for the execution of all kinds of tasks. The platforms often claim to be nothing more than mere labour market intermediaries, bringing together demand with supply. By classifying individuals as self-employed instead of workers they are able to gain significant profits for themselves while bypassing labour and social security legislation. In reality, the platforms exercise ‘decisive influence’ over the providers’ activities (C-434/15, Elite Taxi; C-320/16, Criminal Proceedings Against Uber France). More particularly, they often determine the price of the provided services and can unilaterally alter the workers’ expected rate of return by altering ad hoc the level of commission retained (Todolí-Signes, 2017). Furthermore, platforms often insert cumbersome exclusivity or non-circumvention clauses and usually set minimum quality standards for service, strict branding requirements and rules of conduct (Cherry and Aloisi, 2017; Irani, 2015). To monitor workers, platforms use geolocation devices and customer-led ratings (Aloisi, 2016). If individuals fall below a certain threshold set by the platform, they might get penalised. MTurk and Uber, for instance, are known to suspend workers who do not maintain a high acceptance rate (Sanders and Pattison, 2016; Stefano De, 2016). Even if one’s account is not deactivated, however, there are other ways in which platforms can penalise individuals. Algorithmic management can push workers who decline task offers to the bottom of the option list, meaning that they get less work and get paid less for it (Prassl, 2018).

Apart from controlling workers, platforms also use their power to pass down business-related costs to them. Under many contracts for platform work, individuals are called upon to shoulder material and human capital investment, redeployment and maintenance costs (Georgiou, 2021). More particularly, individuals often have to provide and maintain their own equipment (i.e. vehicle and cleaning equipment) at a certain standard set by the platform and also have to pay for their own petrol, insurance, tax and potential leasing costs (Aloisi, 2016). Furthermore, platform workers are 47% less likely to receive training than their permanently employed counterparts (Eurofound, 2015) and hence have to invest in their own training in order to remain competitive in this volatile labour market. Moreover, under many of these contracts, individuals are called upon to shoulder heightened health and safety costs as well as third-party liability costs (Georgiou, 2021). The low and unstable incomes associated with platform work combined with the long and irregular working hours have been linked to high levels of stress, anxiety and depression (Kleppa et al., 2008). These psychological health conditions place extra strain on workers, often leading to physical health problems and hence, heightened health and safety expenses (Caroli and Godard, 2016). ICT-based mobile workers, for instance, are often responsible for the health and safety conditions of the environment they work in, while platforms workers are responsible not only for their own health and safety, but also for that of their ‘customers’ (Eurofound, 2015; Kuhn, 2016). Some contracts for platform work, for instance, include clauses under which the platform absolves itself from any liability for damages incurred to third parties and to the workers in the course of the performance of their tasks (Todolí-Signes, 2017).

Those affected the most by the recent shift towards casualisation are the most vulnerable: the less-educated and lower-skilled, the women and the young (European Commission, 2017c). With little bargaining power and very few viable alternatives, these workers often have no choice but to accept the employers’ exploitative terms. Empirical research, for instance, shows that involuntary casual work has risen 5.3% in the last 9 years (European Commission, 2017c), as an increasing number of enterprises are taking advantage of these newly developed forms of work to push the costs of doing business onto the shoulders of workers. It is not an uncommon phenomenon, for instance, for people working under permanent contracts of employment to be asked to set up a limited company as the legal vehicle for the supply of their services in order not to lose their jobs completely. Therefore, a new category of workers has emerged that consists of those who formally appear to be self-employed but are rather substantively and economically dependent on a single principal for subsistence (Böheim and Muehlberger, 2006; OECD, 2014).

These labour market developments have superceded the EU legal framework, leaving many persons standing uncomfortably in the ‘grey area’ between employment and self-employment (Perulli, 2003; Supiot, 2001). To mitigate these issues, the Commission put forward in 2017 a proposal for the revision of the WSD with a view to ‘clarify its scope of application’ and with it the concept of ‘worker’, ‘[and] to define core labour standards for all workers’ (European Commission, 2017a: 8, 4). After a short period of gestation, the new DTPWC was adopted in June 2019, repealing the old instrument. The next section analyses the scope and protective effects of the new Directive.

The personal scope of the DTPWC: Reach, protective effects and limitations

According to Article 1(2), the Directive applies to ‘every worker in the Union who has an employment contract or employment relationship as defined by the law, collective agreements or practice in force in each MS’. Like most Directives in the social field, the DTPWC reserves the scope-defining exercise for domestic jurisdictions: each MS is responsible for determining the categories of workers the Directive applies to. An important caveat, however, is introduced: when assessing ‘worker’ status, national judges and legislators have to take into account the case law of the ECJ. By recalling some of the Court’s most prominent judgments (C-66/85, Lawrie-Blum; C-428/09, Union Syndicale Solidaire Isère; C-229/14, Balkaya; C-413/13, FNV Kunsten; C-216/15, Ruhrlandklinik), the DTPWC emphasises the importance of the ECJ’s jurisprudence for the transposition of the Directive into domestic jurisdictions. As the Preamble notes, provided individuals meet the ‘worker’ criteria set by the Court, national trainees, apprentices and portfolio, on-demand, intermittent, voucher-based and platform workers fall under the protective ambit of the Directive. What are the criteria individuals have to meet to satisfy the ECJ’s ‘worker’ definition?

The social acquis

Even though the ECJ has warned that the term ‘worker’ may not always have the same meaning when it is used in the Treaties, ever since Hoekstra (C-75/63) the same definition has been consistently used in the Free Movement of Workers (‘FMW’) field (Article 45 TFEU). In the seminal case of Lawrie-Blum (C-66/85) the Court held that ‘the essential feature of an employment relationship is that, for a certain period of time, a person performs services for and under the direction of another in return for which he receives remuneration’ (para 17). Three criteria have thus emerged as relevant for the assessment: (a) the performance of effective and genuine economic activities; (b) remuneration and (c) subordination. Most activities would satisfy these requirements provided that they are not carried out on such a small scale so as to be regarded as purely marginal or ancillary (C-53/81, Levin) or merely as a means of rehabilitation (C-344/87, Bettray). Remuneration is also needed, but the level of income or its origin is not important (C-139/85, Kempf). In general, the ECJ has developed a broad and autonomous definition of ‘worker’ in the FMW context that cannot be interpreted narrowly.

It has also brought this definition into other areas of EU law such as the equal pay provisions of the Treaty (Article 157 TFEU). In Preston (C-78/98), for instance, a case concerning equal pay claims for work of equal value, the ECJ accorded ‘worker’ status to persons holding intermittent contracts of employment, while in Allonby (C-256/01) it did not hesitate to deal with bogus self-employment. The case concerned the working status of an hourly-paid lecturer who was made redundant from the University she was working at. She was then immediately re-hired by the same institution as ‘self-employed’ through an intermediary agency, only to find out that her pay had diminished and was now lower than that of the male lecturers employed by the University. The ECJ held that attention should be paid to the de facto working relationship between the parties and not its legal characterisation in the contract. Persons who are formally classified as ‘self-employed’ under domestic legislation can still be considered ‘workers’ under Article 157 TFEU if their independence is merely notional. This would be the case, in particular, if the persons do not have the freedom to choose the timetable, place and content of their work. If, after looking at all the facts of the case, national judges think that the persons are, in reality, subordinate to their principal, they should re-classify them as ‘workers’.

Apart from the FMW and equal pay provisions of the Treaty, the Lawrie-Blum ‘worker’ definition has also been used to inform the personal scope of those labour and social law Directives that do not expressly defer to domestic definitions. In cases concerning the personal scope of the Pregnant Workers (92/85/EEC), Working Time (2003/88/EC), Information and Consultation (2002/14/EC) and Collective Redundancies (98/59/EC) Directives, the Court has relied on the Lawrie-Blum formula to provide protection to categories of workers that have unjustifiably been excluded from national ‘worker’ definitions.

Expansive rulings have also been delivered on the personal scope of most social law Directives that explicitly defer to domestic definitions. O’Brien (C-393/10), for instance, was a case concerning the exclusion of a judge from the personal scope of British legislation implementing the Part-Time Work Directive (97/81/EC). After recognising that Directive 97/81/EC allows MS to determine the beneficiaries of its protective provisions, the Court noted that this discretion ‘is not unlimited’ but is fettered by the MS’ obligation to ‘respect the effectiveness of the Directive and the general principles of European law’ (para 34). MS hence cannot arbitrarily remove protection from categories of workers that would, otherwise, be protected.

The effectiveness principle has also been used in cases concerning the personal scope of the Temporary Agency Work Directive (2008/104/EC). In Ruhrlandklinik, for instance, a case concerning the exclusion of Red Cross nurses from the German legislation transposing Directive 2008/104/EC, the ECJ held that ‘the EU legislature did not leave it to the Member States to define [the] concept [of ‘worker’] unilaterally but specified the contours thereof in Articles 3(1) (a) and 3(1) (c) of the Directive’ (para 32). MS, thus, could not inordinately and unjustifiably restrict the scope of its protective provisions by adopting a very narrow ‘worker’ definition. Recalling its earlier judgment in Danosa (C-232/09), the ECJ concluded that the concept of ‘worker’ under the Temporary Agency Work Directive ‘must be interpreted as covering any person who […] for a certain period of time, performs services for and under the direction of another person, in return for which he receives remuneration’ (para 43).

Similar expansive judgments have been delivered in the context of the Fixed-Term Work Directive (1999/70/EC). In Adeneler (C-212/04), for instance, the Court employed l’ effet utile principle to accord ‘worker’ status to persons engaged in successive fixed-term contracts of employment in the public sector. In general, and despite the strong textualist argument, the ECJ’s action in this area has been, overall, worker-protective. As Kountouris notes, ‘the Court is increasingly steering the scoping provisions [of the Directives] in ways that are pushing towards convergence’ (Kountouris, 2018: 202) and does so, notably, whenever the national nomenclature deprives de facto workers of their labour rights rendering the transposition of the relative Directive incomplete.

Finally, the expansive FMW ‘worker’ definition has also been used in other areas of EU law such as the Free Movement of Services and Competition Law provisions of the Treaty. After repeating verbatim that for the purposes of Articles 56 and 101 TFEU, a ‘worker’ is also ‘any person who performs services for and under the direction of another in return for which he receives remuneration’ (C-151/04 and C-152/04, Nadin and Durré: para 3; C-22/98, Becu: para 26, respectively); the Court draws a distinction between ‘workers’ and ‘self-employed persons–undertakings’. If one is a ‘worker’, they are not a ‘self-employed person–undertaking’ and vice versa. Hence, for the purposes of EU Labour Law, Free Movement of Services and Competition Law provisions of the Treaty, a common and autonomous definition of ‘worker’ has emerged that relies on the Lawrie-Blum formula of subordination (Georgiou, 2021; Lianos et al., 2019).

In this context, ‘subordination’ has traditionally been understood as referring solely to the control of the employer over the workers’ activities. More particularly, the employment relationship has been viewed as a hierarchical structure with democratic deficits under which the employer must be in a position of power so as to dictate ‘the type of work and tasks to be executed, the manner in which that work or those tasks are to be performed, and the time and place of work’ (C-270/13, Haralambidis: para 33). The ‘control’ exercised by the employer, however, does not have to be direct or absolute – more subtle forms of power relations have, at times, been taken into consideration as well. In Danosa (C-232/09), for example, a Latvian case concerning the exclusion of a pregnant member of the board of directors of a capital company from the personal scope of Directive 92/85, the Court interpreted the notion of ‘subordination’ as more than just the superior’s right to micro-manage how work is being done. It stressed that it can effectively amount to a power of ‘direction or supervision’, especially when such workers are ‘an integral part’ of the company they provide services to. Miss Danosa thus had to be considered a ‘worker’ because she had to report to the supervisory board and to cooperate with it.

On top of ‘control’, the ECJ has more recently been paying attention to another criterion, namely the ‘assumption of business risks’ (Georgiou, 2021). Agegate (C-3/87), for instance, was a FMW case concerning fishermen on British vessels whose salary was calculated on the basis of a share from the proceeds of their catches. The Court held that the sole fact that persons are paid a ‘share’ and that their remuneration may be calculated on a collective basis is not enough so as to deprive them of their ‘worker’ status. Instead, what matters is whether the persons are ‘sharing the commercial risks of the business’ (para 36). Similar pronouncements were later made in Becu (C-22/98). Following AG Colomer’s suggestion, the ECJ decided that the dockers in question were ‘workers’ and not ‘self-employed persons–undertakings’ because they were under the control of their principal and did not take on any business risks. The criterion of ‘business risk-assumption’ has since been used in a number of cases concerning the classification status of customs agents (C-35/96, Commission v Italy), doctors (C-180 to 184/98, Pavlov), lawyers (C-309/99, Wouters) and chartered accountants (C-1/12, Ordem dos Técnicos Oficiais De Contas). In all these instances, the ECJ found that the persons in question were not ‘subordinate’ to their principal because they ‘assumed the financial risks involved in the exercise of [their] activity’, meaning that, if there was ‘an imbalance between expenditure and receipts, they [had to] bear the deficit’ (C-35/96, Commission v Italy: para 37).

The same criterion has more recently been used in competition and labour law cases concerning disguised employment. FNV Kunsten (C-413/13), for instance, was an Article 101 TFEU case concerning, inter alia, the status of substitute orchestra musicians as ‘workers’ or ‘self-employed persons–undertakings’. The Court took the time to give instructions to national judges on how to spot bogus self-employment. Recalling its previous jurisprudence in the labour field (C-270/13, Haralambidis; C-256/01, Allonby; C-3/87, Agegate), the ECJ held that ‘a service provider can lose his status of an independent trader, and hence of an undertaking, if he does not determine independently his own conduct on the market, but is entirely dependent on his principal, because he does not bear any of the financial or commercial risks arising out of the latter’s activity’ (para 33). Similar pronouncements were recently made in B v Yodel (C-692/19). The case concerned the working status of B, a parcel delivery courier for Yodel, under the Working Time Directive (2003/88/EC). The ECJ found that, as a rule, persons will be considered ‘self-employed’ if they have discretion to: (a) use subcontractors or substitutes in the provision of their services; (b) accept or not accept the various tasks offered by their putative employer, or unilaterally set the maximum number of those tasks; (c) provide their services to any third-party, including direct competitors of the putative employer and (d) fix their own hours of ‘work’ within certain parameters and tailor their time to suit their personal convenience rather than solely the interests of the putative employer (para 45). A caveat, however, was introduced. Recalling its previous judgments in Allonby (C-256/01) and FNV Kunsten (C-413/13), the Court clarified that persons will lose their status as ‘independent traders’ ‘if their independence is merely notional’ (para 30). This would be the case, in particular, if the persons ‘act under the direction of [their] employer as regards [their] freedom to choose the time, place and content of [their] work [and] do not share in the employer’s commercial risks’ (para 31). It was left up to domestic courts to ascertain whether the independence of B was real or whether he was, in fact, in a relationship of subordination with the putative employer.

Assessment of the personal scope of the DTPWC

Having analysed the social acquis in the area, one can make the following observations. On the one hand, the new hybrid ‘worker’ definition has the possibility to broaden the scope of protection to include new categories of workers that were previously excluded from the personal scope of the WSD. The reliance on the ECJ’s jurisprudence means that MS will not be able to unjustifiably remove protection from categories of workers that satisfy the ‘worker’ criteria set by the Court. National judges and legislators would have to take into account the ECJ’s jurisprudence in the area and interpret the national legislation implementing the Directive accordingly. This would increase coverage and homogeneity between national approaches, providing for broader and more convergent definitions and hence, increased coherence, certainty, transparency and predictability within the EU labour market.

On the other, the ‘hybrid’ approach adopted by the Directive might create issues when it comes to its implementation. As Bednarowicz (2019: 612) notes, ‘the wording of this provision remains rather ambiguous and may prove difficult for the Member States not only to implement it properly in its full capacity but also for the judiciary to effectively have it enforced’. The EU is comprised of 27 MS that have vastly different labour legal systems and traditions. Labour and social security laws in the Nordic MS, for instance, such as Denmark, Finland and Sweden, are governed by the social partners who do not like EU interference with internal affairs (Bednarowicz, 2019). Other countries rely on tripartite systems of classification that do not follow the EU’s dualistic approach. Italy, Germany and Spain, for instance, have introduced intermediate categories of working persons (‘para-subordinate’, ‘arbeitnehmerähnliche personen’ and ‘trabajadores autónomos económicamente dependientes’, respectively) that fall in-between the two polar opposites. The application of the ECJ’s jurisprudence in these MS is not as straight forward and might require adjustments. Finally, certain domestic jurisdictions have idiosyncratic ‘worker’ definitions while others use a plethora of different indicia to classify working persons that are not usually mentioned in the ECJ’s jurisprudence (Davidov et al., 2015). Without robust guidance on the EU ‘worker’ definition and the criteria used for the identification of ‘subordinate’ working persons, national judges and legislators might struggle to implement the DTPWC correctly.

More importantly, the reliance on the ECJ’s formula of ‘subordination’ and the criteria of ‘control’ and ‘business risk-assumption’ used thereof means that many casual workers will not be granted protection. As was shown in the section The Need for Regulation, under many modern-day casual work contracts, individuals are called upon to shoulder a plethora of business risks. Under the current regime, the assumption of such risks on their part would automatically render these persons ‘self-employed’ even though they do not have the ability to benefit from the upside payoffs of their risk-taking activity (Georgiou, 2021). Furthermore, employers have nowadays come up with various innovative ways to control workers. Even though the ECJ has at times recognised more subtle forms of power relations, it is not clear how it will decide in cases that involve casual forms of work. Hence, while bogus self-employed persons, meaning persons who are under the complete control of their employer and do not assume any business risks, will be (re-)classified as ‘workers’ and will be afforded protection under the DTPWC; the same is not necessarily true for quasi-subordinate workers. Those who are not under the strict control of their principal and have been made to assume a level of business risks might not be covered by domestic legislation implementing the DTPWC. Hence, even though initial estimates wanted the Directive to cover 2–3 million workers that were previously neglected by EU legislation (including 3% of platform workers) (European Commission, 2017c), in reality, that figure will be much lower.

Moreover, it should be noted that some workers will not be able to take advantage of the Directive’s protective provisions despite satisfying, in principle, the criteria set by the ECJ. MS can remove, on objective grounds, protection from civil servants, policemen, judges, prosecutors, investigators and other persons engaged in law enforcement services. Voluntary derogations have also been introduced for domestic workers. According to Article 1(7), MS may establish specific rules to exclude individuals acting as employers for domestic workers from the requirement to: (i) consider and respond to requests for different types of employment; (ii) provide mandatory training free of cost and (iii) provide redress mechanisms that are based on favourable presumptions in the case of information missing from the provided documents. These derogations have been introduced to accommodate the particular nature of domestic work. In these instances, the employer is not a large firm but an individual that does not usually have the ability to provide training, redress mechanisms or alternative sorts of employment. Furthermore, according to Article 1(3), MS can remove protection from extremely casual workers, defined as persons working on average less than 3 h per week in a reference period of four consecutive weeks.

Finally, a comment should be made about the counterparty of the employment relationship, namely the ‘employer’. The DTPWC does not provide an ‘employer’ definition but allows Member States to determine the persons responsible for the execution of the employer’s obligations as they are set out in the Directive. As it is stipulated, MS ‘may decide that all or part of [the employer’s] obligations are to be assigned to a natural or legal person who is not party to the employment relationship’ (Article 1(5)).

The material scope of the DTPWC

The DTPWC was introduced with the objective of raising employment standards through the provision of more transparent and predictable conditions regarding work. To this end, the Directive provides for an updated and more complete information package. The list of essential information employers have to provide workers with has been extended while the deadline for doing so has been reduced. While under the WSD employers had 2 months to supply workers with the relevant core information on their working conditions; under the new instrument, they have to do so on the first working day and, in any case, no later than the seventh day of work. The other secondary information referred to in Article 4(2) can be provided to the worker within the first month of starting the job. The information must be made available to the worker free of charge and must be written in a clear, transparent, comprehensible and easily accessible manner. If any changes occur in the terms of employment, the employer must inform the worker at the earliest opportunity and, at the latest, on the day on which the changes take effect. Where appropriate, employers must also include information regarding the laws, regulations, administrative or statutory provisions and collective agreements applicable to the employment relationship.

Additional information is to be provided to posted workers (Article 7) and to casual workers who are engaged in on-demand contracts or contracts for work under which the working pattern is entirely or mostly unpredictable (Article 4(2)(m)). In the latter case, the employer must inform the worker of: (i) the number of guaranteed paid hours; (ii) the pay for work performed in addition to those guaranteed hours; (iii) the reference hours and days within which the worker may be required to work; (iv) the minimum notice period the worker is entitled to before the start of a job and (v) the deadline for the employer to cancel a job assignment. If the employer fails to give reasonable advance notice, workers can refuse the job assignment without harming their employment status. Workers might also be entitled to compensation in the case of a late cancellation of a pre-booked assignment (Article 10(3)). The employer cannot penalise workers for turning down a job assignment offered outside the notice period (i.e. by firing them or not re-hiring them for work). If the employer subjects the workers to any adverse consequences, the latter can exercise their right to redress under Chapter IV of the DTPWC (see below).

Furthermore, the new Directive has introduced additional measures to protect zero-hour workers. When no guaranteed amount of paid work is predetermined, MS have to adopt one (or more) of the following approaches to prevent abuse: (i) they can introduce limitations to the use and duration of on-demand or similar work contracts; (ii) they can establish a rebuttable presumption of an employment relationship with a minimum amount of paid hours based on the average hours worked during a given period and/or (iii) they can adopt alternative equivalent measures that would ensure the effective prevention of employers’ exploitative practices (Article 11).

Exploitation is also prevented by the introduction of protective provisions that prohibit the widespread use of exclusivity or non-competition clauses. As the Impact Assessment (European Commission, 2017c) highlighted, there are currently 0.5–1.5 million workers in the EU that are burdened by these sorts of clauses. By restricting individuals from searching for parallel work with another employer, these clauses create hold-up situations where the worker becomes dependent upon a single employer for subsistence. To rectify this situation, the DTPWC bans the use of these types of clauses. As Article 9(1) stipulates, ‘MS shall ensure that an employer neither prohibits a worker from taking up employment with other employers nor subjects a worker to adverse treatment for doing so’. MS, however, can lay down conditions for the use of incompatibility restrictions by employers on the basis of objective grounds, such as health and safety, the protection of business confidentiality, the integrity of the public service or the avoidance of conflicts of interests.

Furthermore, the DTPWC puts in place maximum limits on probationary periods (Article 8). More particularly, it is stipulated that probation periods should not exceed 6 months, apart from cases where this is in the interest of workers (i.e. absence from work due to sickness or leave) or when it is justified by the nature of the employment (i.e. executive positions or public service posts). In the case of fixed-term contracts, MS must ensure that the length of the probationary period is proportionate to the expected duration of the contract and the nature of the work. If the contract gets renewed, the worker should not be subjected to a new probationary period.

Moreover, the DTPWC establishes a right to cost-free training (when this is required by EU or national law) which shall be performed during working hours and count as working time (Article 13) and introduces a right to request to transition to a more predictable and secure position, provided that opportunity is available. More precisely, according to Article 12: ‘MS shall ensure that a worker with at least six months’ service with the same employer […] may request a form of employment with more predictable and secure working conditions where available and receive a reasonable written reply’. If such a request is received, the employer must consider both the needs of its business as well as those of the worker. A short, generic, negative reply claiming ‘limited financial capacity’ would not suffice; the employer must provide detailed reasons, explaining the rationale behind their decision. However, as Lynch notes, ‘a right to request a more secure and predictable form of work is not really a meaningful right for the many workers trapped in precarious and zero-hours type contracts. Much stronger provisions are needed if workers are to have a realistic route to securing a higher number of guaranteed paid hours or a less variable work schedule’ (ETUC, 2017).

Finally, the DTPWC rectifies the previous instrument’s lack of a comprehensive enforcement mechanism by establishing a right to redress (Chapter IV). When workers do not receive from the employer the relevant information stipulated by the Directive in due time, they shall be able either to benefit from favourable rebuttable presumptions introduced by the MS or to submit a complaint to a competent authority to receive adequate redress in a timely and effective manner (Article 15). Redress should also be provided for any other infringement of the rights arising from the Directive (Article 16). If the employer dismisses a worker for exercising their rights under the Directive or if they submit them to any other adverse treatment due to this reason, the worker shall be entitled to redress (Articles 17 and 18). MS are responsible for laying down the rules on penalties for non-compliance with the Directive’s provisions. These penalties, however, must be effective, proportionate and dissuasive (Article 19). MS are allowed to introduce more favourable provisions to workers than those stipulated by the Directive but cannot use the instrument as grounds to reduce the level of protection already afforded to workers by national legislation (Article 20).

Overall, the DTPWC boosts workers’ protections by providing them with a more comprehensive set of day-one rights. The information package provided to workers has been updated to increase clarity vis-à-vis working conditions. This way, employment standards are raised through the provision of more transparent and predictable conditions regarding employment. The introduction of complementary measures for zero-hour workers, the prohibition of burdensome exclusivity clauses and abusive probationary periods, the establishment of an enforcement mechanism and the introduction of a right to request to transition to a more secure and predictable type of employment are all positive steps towards combating the extreme casualisation of work.

Concluding remarks and ways forward

The adoption of a new legally binding instrument in the field of employment should be welcomed as a big accomplishment. The DTPWC rectifies some of the deficiencies of the WSD by increasing workers’ rights. More precisely, the material scope of the Directive has been extended to provide greater protection to persons engaged in new and casual forms of employment. Complementary measures have been introduced for on-demand workers, while enforcement mechanisms have been put in place to ensure that workers have a right to redress. At the same time, the personal scope of the Directive remains rather ambiguous, making it unclear how far the coverage extends. While the reference to the ECJ’s jurisprudence may provide protection to a larger category of workers, it does not go far enough so as to capture individuals who are not under the strict control of the employer and have been made to assume business risks. Hence, despite its seemingly broad personal scope, the DTPWC may not cover many of the casual workers it originally aspired to.

Since the adoption of the DTPWC, the EU has introduced other soft law and hard law instruments that boost workers’ protection. More precisely, the EU has adopted the revised Work-Life Balance Directive (2019/1158) and the Council Recommendation on Access to Social Protection for Workers and the Self-Employed. The latter instrument increases platform workers’ protection by asking MS to commit to extend the coverage of their social security systems to include all types of employment, including self-employment. More precisely, the Council Recommendation calls upon Member States to: (i) allow non-standard workers and the self-employed to enrol into social security schemes; (ii) take measures to ensure that non-standard workers and the self-employed can build up and take up adequate social benefits so that they are effectively and adequately covered and (iii) increase transparency regarding social security systems and rights. In achieving these goals, the Commission will support the MS through dialogue and mutual learning activities as well as through the establishment of a monitoring framework that will follow the implementation of the Recommendation in the different MS.

Moreover, the Commission has recently launched two public consultations on: (i) Collective Bargaining for the Self-employed and (ii) Improving the Working Conditions in Platform Work. Under the current EU framework, only associations of workers are allowed to bargain collectively for the amelioration of their working conditions (C-67/96, Albany). If associations of self-employed persons conclude collective bargaining agreements, they face the risk of being found liable for breaching competition law provisions and might be subjected to large fines (C-180 to 184/98, Pavlov). The only exception allowed is for those who are falsely self-employed. As the ECJ found in FNV Kunsten (C-413/13), the self-employed who ‘cannot independently determine their own conduct on the market’ and ‘do not bear any of the financial or commercial risks of the activity’ should be re-classified as ‘workers’ and should be allowed to bargain collectively for their rights.

Even though, with its decision in FNV Kunsten the ECJ accorded collective labour rights to fully dependent self-employed workers (de facto employees), it did not go far enough so as to guarantee protection for quasi-subordinate workers. The European judges prompted national courts to re-classify only those individuals who are blatantly ‘bogus self-employed’. Those, on the other hand, who are not under the direct control of their principal and bear a level of business risks will not escape the ‘antitrust menace’ (Stefano De, 2017: 190). As Lao observes, the current exclusion of independent contractors from collective bargaining ‘emerged in an era that did not contemplate the new hybrid work relationships that are characteristic of the gig economy’ (Lao, 2018: 1547). Under the current framework, platform workers and other workers engaged in casual forms of work will not be able to take advantage of the ‘FNV Kunsten exception’. Hence, not only are these individuals usually excluded from EU labour and social protection legislation, but they are also unable – because of competition law restraints – to bargain collectively for their rights (Doherty and Franca, 2020; Schiek and Gideon, 2018). Aware of these challenges, the Commission launched on 30 June 2020 a public consultation to address the issue. Subject to the outcome of the impact assessment, the Commission is expected to adopt an initiative by the end of 2021.

Furthermore, in its 2021 Work Programme, the Commission recognised that ‘there is a growing uncertainty on a number of issues [concerning] platform-based work, […] including employment status, working conditions, access to social protection, and access to representation and collective bargaining’. It, therefore, committed itself to propose ways to improve the labour conditions for platform workers. On 24 February 2021, the Commission launched a first-stage consultation of the social partners on how to improve the working conditions for people working through digital labour platforms. Article 154(2) TFEU provides for a second stage consultation of the social partners for proposals in the social policy field based on Article 153 TFEU. As the Commission announced, unless social partners decide to enter into negotiations among themselves, following the first or the second stage of the consultation, it intends to put forward a legislative initiative by the end of 2021.

This can take one (or more) of the following approaches. First, the Commission could propose the adoption of a Directive that focuses solely on the responsibilities of platforms towards the persons who provide services through their site or app, regardless of their employment status. This way, the question of whether platform workers are ‘workers’ or ‘self-employed’ is avoided altogether and instead, attention is paid to the rights of the persons who provide services through platforms (Prassl and Risak, 2016).

A second option is for the Commission to introduce a classification test similar to the ‘ABC test’ adopted by the Californian Supreme Court in Dynamex (Dynamex Operations West, 2018). Under the ABC test, a person would be classified as a ‘worker’ unless the putative employer can prove (cumulatively) that (a) the person is free from the control and direction of the hiring entity in connection with the performance of the work, both under the contract and in fact; (b) the person performs work that is outside the usual course of the hiring entity’s business and (c) that the person is customarily engaged in an independently established trade, occupation or business of the same nature as the work performed. Essentially, the ‘ABC test’ would reverse the onus of proof by establishing a rebuttable presumption of employment status for all platform workers.

Finally, the Commission could clarify the legal status and rights of platform workers. This could be done in two ways. First, the Commission could provide a list of indicia national and European judges and regulators have to take into account when classifying persons who provide services through digital labour platforms. This is, for instance, the approach adopted by the ILO in Recommendation 198. Second, the Commission could propose the adoption of a broader EU ‘worker’ definition that would cover platform workers as well as other persons engaged in casual forms of work.

It has been suggested elsewhere that the EU would benefit from the adoption of an alternative ‘risk’ criterion that is based on the ‘involuntary assumption of business risks’ measured by the ‘inability of a person to spread his risks’ (Georgiou, 2021). The rationale behind this conceptualisation is the following: when persons do not have the ability to spread their business risks (because, for example, they have little or no capital, have made sunk or job-specific investments, have no employees, have no alternative sources of income and/or have little or no control over the business strategy), they are not in a position to make truly free choices. In these instances, the persons are constrained by the lack of alternatives that force them to accept whatever terms are being offered. Since the assumption of business risks on their part in these circumstances cannot be said to be an expression of the persons’ free will, the State has legitimate reason to intervene and re-classify them as subordinate ‘workers’. By contrast, when persons have the ability to spread their business risks (because, for example, they can pass them on to consumers/clients through the mechanism of price, have significant capital, have employees of their own and/or have multiple sources of income), the assumption of such risks on their part is presumed to constitute a genuine choice. In these instances, the persons have taken on the business risks accepting the concomitant ‘risk’ of being classified as ‘self-employed’. Therefore, the State has no legitimate reason to interfere with their choice (as that is expressed in the contract) and re-classify them as ‘workers’. Individuals who can spread their risks can decide to be engaged under contracts for work either as subordinate ‘workers’ or as ‘independent contractors’, making in each case the trade-offs they deem desirable and necessary. While the proffered approach does not make a radical departure from existing case law, it constitutes a refinement tailored to correspond to recent market conditions. It can easily be adopted by EU judges and legislators as it does not stray far from existing jurisprudence: it still relies on the concept of ‘risk’, albeit with a different focus. Acting alongside the criterion of ‘control’, this alternative ‘risk’-based criterion would allow for a broader conceptualisation of the EU notion of ‘worker’. More particularly, it would expand the scope of protection to include platform and other casual workers who are currently excluded from the EU ‘worker’ definition.

Overall, in recent years, the EU has taken several steps to increase the level of protection afforded to workers. After a long period of inactivity, the Commission is finally displaying a more social face, moving away from the hard ‘flexicurity agenda’ (Fredman, 2004) and the strong neo-liberal discourses that dominated its polices over the last few decades. Instead, the EU is finally showing its willingness to modernise its labour and social acquis to ensure that it corresponds to recent market conditions. The adoption of the DTPWC and the launch of the two recent initiatives on collective bargaining and platform work constitute positive steps towards providing more protection to casual workers. It remains to be seen whether the EU will succeed in introducing more worker-protective legislation or whether any such attempts will prove to be nothing more than a Sisyphean task.