The Internal Market at a Social Turn? Social dumping and the protection of workers

1. Introduction

It is the very aim of the free movement rules to ensure a better allocation of workforce, which both benefits the economies of the Member States and the workers. Yet in real life, the picture is more blurred, as it is not always so easy to activate the right of free movement. It might be easier for workers to escape poverty at home through the provision of services enabling their home employer to find them temporary work in other Member States. ¹ Yet, this flexible allocating mechanism known as posting has inherent flaws, as the posted workers are in principle not ensured equal treatment in the host State. Thus, posting can in practice lead to social dumping in specific sectors where posted workers ultimately push salaries to the bottom. Such a race to the bottom can in turn lead to poverty on the part of the workers. There is indeed a strong correlation between decent working conditions, decent wages and social security protection, and the fight against poverty and social exclusion. Thus, a deterioration in such conditions and salaries in the host State might negatively affect the local workers. Social dumping is in fact possible because of the important differences in salary and social protection within the EU, which are still regulated at national/local level. In a free movement context, this leads to tensions between sending and recipient Member States, essentially between East and West but also between North and South, and fuels claims of illegitimacy of the EU and its internal market logic. Social dumping is broadly defined here as unfair competition between undertakings on labour costs potentially leading to a race to the bottom of wages and eventually the replacement of the workers. ² Thus, differences between countries in salary levels, social security protection and personal taxation may imply that the labour costs of someone ‘temporarily’ doing a specific job in a host country can be substantially lower than the labour cost of someone doing the same job in the host country for an employer established there. ³ Moreover, in this contribution social dumping is not limited to a free movement context and includes situations of ‘false’ self-employed. The latter usually enjoy a competitive advantage compared to employees in a similar position through the absence of labour costs. The ‘false’ self-employed thus undercut prices with the risk of pushing to the bottom the social protection of those with employee status.

Competition on labour costs might to a certain extent be legal from an EU law perspective, but it presents a challenge for a ‘highly competitive social market economy aiming at full employment and social progress’, as enshrined in Article 3(3) TEU. This was recently acknowledged by the EU institutions and the Member States, where the interests of businesses to a greater extent have to give way to the interests of the workers. Both the Revised Posted Workers Directive and the EU action plan for the collective bargaining of certain self-employed are evidence thereof. ⁴ This article argues that EU economic law can both create social dumping and remedy it. It calls for a rebalancing of the liberal ethos of the principle of free movement and competition law, on the one hand, and the social objectives of the EU, on the other, addressing key concepts such as equal pay for equal work and unfair competition. It also looks at the issue from an institutional perspective assessing whether the ECJ and the legislator are in tune in this respect. A key question is whether it is possible to redress the balance between the economic and the social from within the principle of free movement and the internal market logic or whether the social push has to come from outside. In other words, can the freedoms and the internal market become more social and outlaw social dumping in its various forms?

The article first briefly discusses the limitations of the free movement of workers in fighting poverty (section 2). It then analyses how social dumping is at all possible in a free movement context (section 3) and discusses some remedies that the EU has put in place (section 4), assessing the impact of the Revised Directive on Posting (section 5). Finally, it addresses situations of social dumping arising outside a free movement context through the phenomenon of ‘bogus’ self-employment, which has been flourishing of late in the new business forms of work platforms (section 6).

2. The limits of the free movement of workers

It is the very aim of the free movement rules to ensure a better allocation of workforce, which both benefits the national economies, by filling vacant positions in the host State; and the workers, by improving their living conditions. Thus, free movement should, at least in theory, enable individuals to fight poverty at home by finding employment in another Member State. Workers are guaranteed equal treatment in the host State as that secures their integration and aims at preventing social dumping situations. As stated in the introduction, free movement is difficult to achieve in practice and does not necessarily lift people out of poverty. In the context of fighting poverty within the EU, the free movement of workers has essentially four main flaws. Firstly, it might be difficult for EU citizens to move in the first place, and thus improve their economic situation. For example, EU citizens might lack financial resources, the required knowledge to fill the jobs, networks to support them in the host country and savings upon which to fall back. More specifically, they might not have sufficient means to provide for a living in the host State during the job seeking period. This is especially the case for migrant workers who have moved from poorer to richer countries where the export of unemployment benefits, if available, might not cover the costs of living. ⁵ Furthermore, it follows from EU law that there is no obligation for the host State to help on an equal treatment basis in this initial period. ⁶ Secondly, despite the inclusive concept of ‘worker’ developed by the ECJ, it might not cover work which is too ‘atypical’, for example, provided over a short period, not regular or task based (on-demand/gig work), with low remuneration or not subordinated to an employer. EU citizens thus might not qualify as workers and therefore will have little protection in the host State. Thus, the most precarious workers might fall within between the two stools of the social security and labour law protection. More generally, there is a risk of exploitation of migrant workers even if they are EU nationals and thus covered by the fundamental principle of equal treatment. This is, for example, the case for workers in a weak position (who lack accommodation, language skills, network, savings, education, etc.) where employers pay them less than the required minimum wage and/or make them work much longer than the law permits. ⁷ Finally, some work-intensive and low-income sectors, such as the meat processing industry and the agricultural sector are populated with migrant workers. It is true that these kinds of jobs, which are easily available to precarious workers from ‘exporting’ countries are precarious jobs in the home countries as well. Thus, this may result in an increase in the absolute level of financial resources for the workers, but still leaves them in a situation in which it is not easy to escape poverty. Furthermore, in some sectors, some of the low-paid work has been displaced from one State to the other. This could be seen as the other side of the same coin of social dumping where jobs vanish in one State to be relocated to States with a more advantageous regulatory framework for the undertakings. For example, within the last 12 years meat production has been moved from Denmark to Germany, where limited union involvement and low minimum salaries facilitated such a displacement of production. ⁸ Furthermore, the Covid-19 crisis has amplified old challenges. Thus, the infection of several meat factory workers with the virus in Germany and elsewhere has confirmed that the vast majority of the workers were migrant workers on temporary contracts. Accommodation in crowded facilities provided by the employers was one of the factors facilitating the propagation of the virus. Yet, the specific nature of the legal status of the workers under EU law is unclear, with most of them being on temporary contracts and a fair proportion of them likely to be posted workers. I now turn to the issue of posting.

3. Posting of workers: How is social dumping at all possible?

When an employer temporarily posts some of its staff to do work on the territory of another Member State, the situation is assessed under the provisions of the free movement of services. As early as 1990, the Court took a liberal stance, looking at the issue from the perspective of businesses where requirements of pay and work conditions imposed by the host State on posted workers were seen as restrictions, which needed to be justified. ⁹ The ECJ ruled that posted workers are not workers under the general provision of Article 45 TFEU, and thus cannot benefit from the principle of equal treatment as they do not intend to integrate into the labour market of the recipient State. ¹⁰ Yet, for the individual, posting has the advantage of being easier and more flexible than ‘settling’ abroad. ¹¹ Indeed, the job is there and the employer takes care of the travel and housing costs, possibly arranging the latter itself. Posting is still a relatively small phenomenon and is in most cases unproblematic. Yet, in some labour intensive sectors of activity, such as the construction and transport sectors, posting might make up a substantial share of the labour market ¹² and such sectors are impacted in most EU countries. ¹³ For example, 42% of all posting in 2014 was concentrated in the construction sector and represented over half of the total number of workers received in countries such as Austria and Belgium, where it amounted to 10% and 20% of the domestic workforce respectively. ¹⁴ In such sectors with an intense mobility from East to West (and North), social dumping might arise. It is argued here that social dumping does not necessarily result from the fraudulent exploitation of legal rules. In contrast, social dumping is, to some extent, built into the EU legal system as it encourages businesses to take advantage of low labour costs (especially social security and salary costs) in their place of establishment. In some labour intensive sectors, it thus risks replacing domestic workers with posted workers as undertakings would be tempted to ‘import’ a cheaper workforce. ¹⁵ While EU law creates situations of risk of social dumping, it can also remedy them. Yet there is no quick fix, as this would involve revisions in various legal fields and require the involvement of all EU institutions and Member States. In the following, I will first focus on the legal framework, which to some extent favours social dumping (sections 3.1 and 3.2), and then on the fraudulent exploitation of rules by the actors (section 4).

4. Selected abuses and remedies

Precarious or exploitative employment relationships concerning posted workers have been reported in the construction and road transport sectors in particular. ³² Thus, the trend at EU level is now on the enforcement and monitoring of the rules, which have an impact on posting. ³³ Posted workers are still envisaged as falling within the provision of services, but the aim is to find a better balance between the free movement of services and the social protection of workers. Both the EU legislator and the ECJ play along those lines and grant more leeway to the Member States in this respect. The enforcement Directive has thus empowered the Member States in this respect. As I will argue below, focusing on social security abuses and the transport sector, the balance now seems to slightly tip in favour of enforcement and protection of the workers, at least on paper, but some challenges remain.

5. Towards equal treatment and a new competition paradigm?

Following Laval and Rüffert, legally ‘the legitimate objective of fighting social dumping could be satisfied by applying the said minimum conditions’ set up in the Directive on posting. ⁵⁰ As argued above, such a rationale creates a legal and factual situation of inequality between posted and domestic workers with the risk of social dumping when differences in labour costs across the Union are important. In the following, I will discuss whether the Revised Directive really achieves a better balance between the economic freedom of movement and the protection of the social rights of workers: is it evidence of a paradigm shift or just a technical adjustment of old rules? Firstly, looking at the legal basis, the Revised Directive is firmly anchored in the free movement of services, which is its sole legal basis. Moreover, the Preamble also mentions the abolition of restrictions to the free movement of services thus aiming at benefiting cross-border businesses. At the same time, the Preamble mentions the social core of the Union and its aim of striking a better balance with the economic rationale of providers. For example, it refers to Article 3 TEU which sets out the Union aims of promoting social justice and protection, and the horizontal clause of Article 9 TFEU which states that the EU should promote i.e. high level of employment and guarantee of adequate social protection when legislating. The Preamble also makes an important reference to the fundamental principle of equal treatment and equal pay, which is central i.e. to the free movement of workers, regardless of the type of work contract, including precarious work such as fixed-term, part-time and temporary agency work. Thus, the Revised Directive has a strong economic foundation, which is yet balanced with an equally strong reference to provisions on social protection. Secondly, this difficult (re)balancing is also reflected in the Directive’s content, which aims at providing more equal treatment between posted workers and domestic workers. The Commission explicitly acknowledged that there was an ‘uneven playing field due to differentiated wage rules for local and posted workers’ and that this has ‘lowered the acceptance of posting as a phenomenon’. ⁵¹ The Revised Directive therefore replaced the words ‘minimum pay’ with ‘remuneration’, which it defined broadly. ⁵² Yet, the political slogan of ‘equal pay for equal work at the same work place’ promoted by the Commission and Member States like France, does not feature in the final version. Furthermore, the Revised Directive does not change the way remuneration is set, as it still refers to national legislation or collective agreements with general application. ⁵³ Thus, a system of ad hoc negotiations like the Swedish one in Laval, is still not possible under the new regime. Yet, unlike in Rüffert, a wage, which is set using one of the mechanisms provided by Article 3(8) of the Directive, can be upheld against a universally applicable collective agreement setting a lower wage. ⁵⁴ In other words, pay set in a collective agreement in accordance with one of the mechanisms envisaged by the Directive will prevail. Thus, in theory there is a real improvement for the workers even though it is still limited and will most likely be felt in countries with a minimum wage where the differences between posted and local workers are greater. Furthermore, the Revised Directive still ‘outlaws’ decentralised systems of collective bargaining such as company-level collective agreements. ⁵⁵ In general, it thus does not provide equal pay for equal work in the same workplace. On the other hand, the EU legislator grants equal treatment with national workers for long-term posting exceeding 12 months, but inserts three important caveats. ⁵⁶ Thus, the host State is required to extend the 12-month period to 18 months upon receipt of a motivated notification from the service provider. Moreover, the applicable national rules are those fixed according to the mechanisms mentioned above, and the situation of the long-term posted worker might thus de facto differ from that of local workers. Finally, the Revised Directive refers to some exceptions to national equal treatment, such as the payment of supplementary retirement pension schemes. On a positive note, in terms of employment protection, the Revised Directive improves the situation of workers posted to another Member State through temporary work agencies, which is a commonly used practice that is prone to abuse. ⁵⁷ Such posted workers are subject to the same work and salary conditions as the local workers of the user undertaking. ⁵⁸ Thus, they enjoy an extended protection, as unlike other posted workers, interim posted workers are subject to collective agreements at company or sectoral level. ⁵⁹

To conclude, the Revised Directive bridges the gap between posted and domestic workers and brings the situation of the former closer to that of local workers. Yet, there is still some way to go to achieve full equal treatment. The legal challenge pending before the ECJ will provide it with the opportunity to rule on the balance between the economic and the social interests, which the Treaty and the EU legislator intend to achieve. Indeed, Hungary has launched an infringement proceeding arguing that the Directive amounts to a breach of the freedom to provide services as it constitutes economic protectionism. ⁶⁰ Thus, the Court is confronted with two conflicting visions, one of a social market economy with an enhanced social protection of workers and discouraging social dumping, and a more liberal vision of the internal market which allows businesses to compete in all areas, including labour costs, thus making labour a commodity. The case thus grants the ECJ a chance to elaborate on the different rationales underpinning the internal market and it will be interesting to see how the Court will reason. Yet, given the highly sensitive nature of the case, it is likely that the Court will – like its Advocate General - opt for a more technical reasoning. I have previously argued that the Court could rather quickly dismiss the case, stating that the EU legislator has a broad discretion in choosing the legal basis and pursuing the objective therein. ⁶¹ Eventually, the Court could add that the free movement of services should be read in the context of the Lisbon Treaty, where the internal market is only one of the many objectives of the EU, which also include the protection of fundamental rights and social protection. ⁶² In this respect, the European Pillar of Social Rights with its principles of ‘fair and equal treatment of workers’ and ‘fair and decent wages’ could be a supportive argument. ⁶³ Another argument could be found within the evolving concept of the internal market, which has become a ‘social market economy’ with the Lisbon Treaty. Thus, a better rebalancing between the economic and the social, or an ‘integration’ of both values, is required within the context of the internal market itself. ⁶⁴ One could even take this argument a step further and claim that such rebalancing or integration process is vital in securing the ‘social acceptability’ of the internal market. ⁶⁵ Time might thus be ripe for the EU to be more specific on its competition parameters. ⁶⁶ It could therefore be argued that regulatory competition, whereby Member States are encouraged to exploit the differences between legal systems to gain an advantage, should be circumscribed. ⁶⁷ In contrast, in line with the Preamble of the Revised Directive, the case could be made for regulatory neutrality where competition should take place on other aspects such as efficiency, quality and training. ⁶⁸ Thus, the internal market might have entered into a new phase, one of ‘a truly integrated and competitive’ market, where undertakings do not compete on labour costs per hour. Finally, such a paradigm shift is a powerful argument towards the full application of the principle of equal treatment of posted workers in situations where there is a risk of social dumping. It could counter-balance the argument that the posted undertaking is not in a similar position as the local undertaking as it encounters additional administrative costs and additional expenses in terms of expenditure on travel and accommodation, and thus should not be subject to the same rules. Advocate General Campos Sanchez-Bordon argued more cautiously that the Revised Directive did not breach Article 56 TFEU as it did not create full equality between posted workers and local workers, thus implying that the contrary would be the case.

It has been demonstrated so far that the EU regulatory framework can negatively affect the position of posted workers and the local labour market. Furthermore, posting, in some labour-intensive sectors, can be an exploitative business model – a situation that the EU legislator is trying to remedy. Yet, social dumping with the risk of replacement of workers can also occur when workers are wrongly categorised as self-employed. This situation often occurs in the context of posting, where posted workers are disguised as self-employed, who are free to set their fees and working conditions. ⁶⁹ Yet, ‘false’ self-employment goes well beyond a free movement context and frequently occurs in the new business model of platform work. This bogus self-employment is not, as such, a problem created by free movement. Yet, as in the case of posted workers, it raises the issue of a fair European labour market and the role that EU law plays in this respect.

6. Self-employment and social dumping: (re)socialising competition law?

‘Disguised’ self-employment might give rise to situations of unfair competition as businesses escape labour costs imposed on the former, and thus undercut prices with the risk of pushing to the bottom the social protection of the workers and eventually replacing them. The question here, is whether EU economic law works as a break or a pull for the protection of the interests of ‘vulnerable’ self-employed, especially individuals providing physical work mediated through an electronic platform (such as cleaning and delivery tasks). Does EU competition law, which is at the core of the internal market, stand in the way of their social protection? Can the inclusive concept of worker developed by the ECJ push, over time, for a fairer European labour market and can it socialise competition law? ⁷⁰ The Court’s broad concept of worker initially developed within the context of the free movement law (Article 45 TFEU) has made inroads into competition law. ⁷¹ For example, agreements between the social partners escape the prohibition of cartels enshrined in Article 110 TFEU if they aim at protecting workers as defined by the ECJ. ⁷² This includes the false self-employed where their independence is notional. In FNV Kunsten, the Court ruled that the independence of the self-employed might be merely notional when they are acting under the direction of another person, do not share that person’s commercial risks, and for the duration of the relationship, form an integral part of that person’s undertaking. ⁷³ Even in a competition law context, the Court adopted a traditional labour law reading of the situation revolving around the concept of worker and the criterion of subordination. ⁷⁴ In contrast, Advocate General Wahl argued that a collective agreement should not be subject to competition rules when it is specifically intended to prevent social dumping. Thus, collective bargaining would be legitimate not only when the self-employed can be assimilated to workers but also when they are genuinely self-employed but can substitute for workers with employee status. The key issue in the case was thus whether the agreement was necessary to protect the members of the union who had an employee status. It is argued here that both the Court’s reading and that of the Advocate General rely on a ‘social dumping’ rationale, as ‘disguised’ self-employment also gives rise to situations of unfair competition where businesses undercut prices with the risk of pushing the social protection of the workers to the bottom.

This competition law discussion highlights the importance of the role played by unions, and the Commission recently launched an initiative addressing the issue of collective bargaining for the self-employed, including platform workers. ⁷⁵ Therefore, there has been a recent political push in favour of empowering self-employed in need of social protection, such as platform workers, to defend their interests through collective agreement. This might work both as a political signal to the Court and result in guidance and/or legislative action from the EU Commission. ⁷⁶ In terms of the substance of such rules, one avenue could be ‘worker-centric’ codifying and expanding the Court’s case law on the concept of worker, including the false self-employed and other social dumping situations. It could eventually create a specific concept of worker for competition law purposes. Another (more limited) avenue departing from such a ‘worker-centric’ approach could be to ensure compulsory social security coverage of all individuals on the labour market. Yet, this road was not followed by the recent Recommendation of the Council on access to social protection. ⁷⁷ Finally, and still moving away from a worker-centric approach, a solution could be to empower all persons, regardless of status, who are in a weak contractual position whose independence is merely notional and who are thus in need of social protection. Some national legislators, such as the French Parliament, have, in this vein, allowed the self-employed performing work mediated through a platform to defend their interests through collective bargaining. ⁷⁸ All approaches fit with a new reading of the internal market provisions along the lines of a highly competitive social market economy aiming i.e. at social progress. Likewise, a reading of the competition law rules in the light of the horizontal clause of Article 9 TFEU on the protection i.e against unemployment and social exclusion could ‘resocialise’ the competition rules. Some have argued that an empowering of platform workers would even be in line with the strict economic rationale of market efficiency, which underpins the competition rules. ⁷⁹ Yet, others have acknowledged that this would only empower the strongest platform workers and overlook those most in need of protection. ⁸⁰ Thus, a new reading of the internal market more firmly anchored in the protection of social values might be needed, and it could be rooted in a human rights approach. Indeed, the Charter of Fundamental Rights of the EU has enhanced the visibility of fundamental rights such as the freedom of association and the right to bargain. ⁸¹ EU institutions are obliged to uphold its provisions when they legislate, as are national authorities when implementing EU law. ⁸² Furthermore, the EU Charter requires that when similar provisions are enshrined in the European Convention of Human Rights, they set a minimum standard that the EU should respect. ⁸³ Support for a broad bargaining power benefiting the dependent self-employed, including platform workers, could thus be found in the case law of the Strasbourg court. ⁸⁴ Yet, the European Social Charter might be a more inclusive source of protection. Its legal position has been strengthened by the Charter, but its potential is still, to a large extent, untapped by the ECJ. ⁸⁵

In any event, some action and guidance from the political arm of governance seem required to enhance the protection of the vulnerable self-employed and outlaw social dumping. Indeed, achievement solely through judicial fiat seems compromised. It is true that the Court has been rather generous in interpreting the concept of worker and core labour law concepts such as that of subordination, adapting it to new and more subtle forms of control. ⁸⁶ Yet, it is uncertain whether the concept can be stretched enough as to encompass business models such as platform work, which are based, to a certain extent, on a shift of responsibility for risks from the business to the performer of the service. ⁸⁷ Some have argued that the Court might not be ready to take that step and ‘recognise that this shifting of risk is an expression of economic dependency on the worker or micro-entrepeneur’. ⁸⁸ This would require the Court’s willingness to move away from a risk approach based on personal dependence to an approach focused on economic dependence. ⁸⁹ In Yodel of April 2020, the Court seemed reluctant to play a key role in this respect. Indeed, the Court found that couriers whose contract allowed them to delegate their tasks to other persons, who could work for other third parties and who could choose their working times rather freely, were most likely not under the control of the contracting company and were thus not covered by the Working Time Directive. ⁹⁰ This is the ECJ’s first case on platform workers from a labour law perspective and the Court’s ruling is surprising and telling in several respects. Firstly, it is puzzling that the Court saw the legal issue as a straightforward one, which could be handled by way of a reasoned order without hearing the parties and not requiring a written opinion of its Advocate General. Secondly, its ruling in Yodel contrasts with its reasoning in the Uber case, where it had adopted a rather subtle criterion of control, adapting it to the realities of the new phenomenon of platform work. ⁹¹ For example, the Court found that Uber was exercising control over its drivers as it determined the price for each delivery. Like in Uber, the courier in Yodel could choose his/her working times, and eventually refuse tasks assigned to her/him. Yet, in Yodel unlike Uber, these features seemed to demonstrate a lack of control of the employer and thus an absence of employee status. Thirdly, the preliminary questions granted the Court a unique opportunity to assess the issue of platform workers more generally thereby also addressing the EU legislator’s contradictory positions in respect of the labour protection of certain self-employed. ⁹² Finally and maybe most importantly, the Court did not answer all the submitted questions. In contrast to the AFMB case, it did not assess whether the contractual framework could lead, in practice, to bogus self-employment or economic dependency of the courier upon the platform, and thus to social dumping. Indeed, the plaintiff carried on his business exclusively for one undertaking and did not delegate his tasks to others. The Court thus refused to look into the economic reality of the courier. This might well be justified from a labour law perspective and the importance of the criterion of personal work, but it can also be interpreted as a clear signal that the Court leaves it to others to assess the status of platform workers. Indeed, the Court might be unwilling to push the boundaries of the concept of worker to include persons, who are de facto dependent on one main contractor from whom they derive their main income.

7. Conclusion

It has been demonstrated that the internal market rules can create situations of social dumping within the context of the posting of workers and might not sufficiently protect vulnerable self-employed. This article has argued that this outcome is both the result of EU case law and legal rules. Yet, both the ECJ and the EU legislator have recently attempted to redress the balance between the social protection of workers and the economic freedom of providers. This rebalancing exercise finds its place within the context of the economic freedoms but within a legal framework, which is imbued with social values. Thus, the set-up of the Lisbon Treaty, which both redirects the internal market in a social direction and enhances the importance of human rights has resocialised the legal landscape. A resocialising of the internal market is both needed and legally possible. Recent initiatives of the EU legislator in the field of posted workers are important steps in this direction. The social protection and empowering of platform workers provides a golden opportunity for the EU Commission to follow on this path. Thus, political action is needed, and platform workers cannot solely count on the ECJ, which might not be prepared to stretch the concept of worker sufficiently so as to include the new economic realities of the dependency of many workers, such as platform workers. Likewise, human rights provisions and practice of other courts and human rights bodies could provide efficient legal ammunition, but it is up to the political governance branch of the EU to give it a first shot.