Abstract
The case of the Working Time Directive (WTD) is a prime example of a failed attempt by the Member States and the Commission to counter rulings of the European Court of Justice (CJEU) by legislative overrule. Outsourcing the decision making process to the social partners also did not deliver the desired results. After years of trying to reform the WTD, the Commission changed its strategy and issued an interpretive communication instead. However, it is doubtful that this communication will solve all that is wrong with the WTD. What were the obstacles to legislative overrule in this case? What other strategies in avoiding the consequences of CJEU rulings do the Member States apply? What will the future of WTD look like?
Keywords
1. Introduction
The Working Time Directive (WTD) 1 has been subject to an action for annulment, some infringement proceedings, numerous preliminary rulings, a number of unsuccessful social dialogues and successful amendments of minor provisions, but failed attempts to reformulate its most controversial provisions. Frustrated by a decade-long deadlock of the legislative process, the Commission finally gave up its effort to reform the WTD. In May 2017 the Commission published guidelines on how to interpret the directive in a rather unusual, detailed interpretative communication instead. This eventful history of the WTD makes it an excellent case study of how political actors try to rein in the European Court of Justice (CJEU) by reformulating the law so that the CJEU cannot rule the same way in future cases. It is also an example of how difficult this can be. After the CJEU declared several practices of Member States incompatible with EU law concerning the treatment of on-call duty, 2 the Commission, under pressure from the Member States, drafted an amending directive that was meant to make it impossible for the CJEU to continue ruling in the same way. 3 This strategy of legislative overrule failed because the amending directive was rejected by the European Parliament (EP) and no agreement could be reached between the EP and the Council of Ministers in the conciliation committee. 4 The fact that the conciliation committee could not reach an agreement was, until then, unparalleled. 5 Subsequently, the Commission initiated a social dialogue based on Articles 151–156 of the Treaty on the Functioning of the European Union (TFEU) in the hope that the social partners would come to an agreement where Council and EP could not. However, no agreement was reached. The Commission then announced it would carry out what it calls ‘a detailed impact assessment of the directive’ 6 to explore different options of how to proceed. However, the interpretative communication seems to be the final attempt to bring some clarity to how to apply the WTD, at least for the time being. 7 These recent events are a good reason to take a closer look at the rich history of this directive, in the hope that it will increase our understanding of how rulings of the CJEU influence the legislative process of the EU in general.
A short historical overview of the case law and the reactions of the legislators will place the recent developments concerning the WTD into context and give an answer to the question of why a new proposal was introduced at all. At this early stage, on-call duty was the most disputed issue connected to the WTD. The major provisions of this new proposal will be described briefly. However, the focus of the article lies on the events triggered by the proposed amendments. We see an Advocate General (AG) calling upon the CJEU to take a more flexible approach that would reflect the wishes of the Member States, a legislative process that ended with rejection of the proposal, a failed social dialogue, a new attempt by the Commission to revive the process of amending the directive and different strategies employed by Member States to circumvent the contested provisions of the directive.
2. The early years
The WTD prescribes: minimum daily rest periods of 11 hours; breaks when working days are longer than six hours, the details of which are left to collective agreements; a weekly rest period of 24 uninterrupted hours; weekly working hours of not more than 48 hours; annual leave of at least four weeks; and night work to be no longer than an average of eight hours in a 24 hour period. It also addresses, rather superficially, shift work and patterns of work. It defines working time as ‘any period during which the worker is working, at the employer’s disposal and carrying out his activity or duties, in accordance with national law and/or practice’, and rest period as ‘any period which is not working time’. 8 In certain cases the directive allows for derogations from these rules. 9 Additionally, Article 18 of the WTD provides for the possibility that weekly working time could exceed 48 hours with the consent of the individual worker.
The birth of the WTD in 1993 foreshadowed a turbulent life. The United Kingdom (UK) did not agree with the legal basis used by the Commission, namely Article 118a EEC on the health and safety of workers, which allowed for Qualified Majority Voting (QMV) in the Council of Ministers. Instead, the UK wanted the directive to be based on Article 235 EEC on measures necessary to attain the goals of the Treaty but for which no powers are provided in the Treaty, which required unanimity. QMV was nevertheless applied and the UK abstained in the final vote in the Council. The UK then brought an unsuccessful action for annulment against the WTD claiming that the wrong legal basis was used and several other procedural requirements were breached. 10 Four years later, the CJEU found Italy and France in breach of their obligations under the WTD in infringement proceedings. 11 After that, the CJEU delivered a number of preliminary rulings on the WTD, the majority of which concerned the definition of working time found in Article 2 of the directive and its meaning for time spent on-call. 12 In all of the cases dealing with on-call duty, the employers, all of them public health and emergency services, did not calculate time spent on-call as working time. The CJEU consistently declared this practice incompatible with the directive, arguing that inactive time spent at the disposal of the employer has to be counted as working time in its entirety.
Unrelated to the rulings of the CJEU on the WTD, the Council and the Parliament adopted a revised version of the WTD in 2003, Directive 2003/88, which included sectors that had been excluded from its scope and dropped the provision that Sunday should, in principle, be part of the weekly rest period.
To sum up, the rulings of the CJEU on on-call duty caused some concern in the Member States. They argued that defining time spent on-call as working time, as the CJEU did, would lead to unbearable costs and a shortage of personnel. Consequently, they asked the Commission to initiate a reform of the WTD.
3. Attempting to amend
In 2004 a new phase in the life of the WTD began with the Commission sending a proposal to amend the Directive to the EP and the Council after the social partners declined to start a social dialogue. 13 The proposal and the accompanying explanatory memorandum left no doubt that the main purpose of the proposal was to make it impossible for the ECJ to rule in the same way in the future. 14 In contradiction to all CJEU rulings on working time, the proposal defined inactive on-call time as not being working time, ‘[on-call time is the] period during which the worker has the obligation to be available at the workplace in order to intervene, at the employer’s request, to carry out his activity or duties’, and the inactive part of on-call time as the ‘period during which the worker is on call […] but not required by his employer to carry out his activity or duties’. 15 The proposal makes clear that inactive on-call time ‘shall not be regarded as working time’. 16 Under certain conditions the proposal also retained the possibility for individual opt-outs from the maximum weekly working time. 17 Around that time the CJEU decided in Pfeiffer and Others, seven joined cases from Germany, that in order to exceed the allowed weekly working time, consent must be given individually and freely. On-call duty was not discussed in this ruling. 18
The Council of Ministers welcomed the proposal, claiming that after SIMAP and Jaeger it would introduce legal certainty by defining ‘on-call time’ and ‘inactive part of on-call time’. 19 However, the Council of Ministers could not agree yet on the opt-out provisions concerning maximum weekly working time. In 2005, the Committee of the Regions (CoR) supported the proposal, but suggested the removal of several opt-out possibilities and the gradual phasing out of the remaining opt-out possibilities. 20 The opinion of the Economic and Social Committee (ESC) from the same year included something for everyone. The ESC would have liked to leave the regulation of on-call duty to the social partners so that different industries could adopt different rules. At the same time, it criticized the Member States for not complying with the rulings of the ECJ and supported the phasing out of the opt-out possibility. 21 However, the EP wanted a general reference to the case law of the ECJ and to the Court’s definition of working time to be added to the recitals. In stark opposition to the Commission’s proposal, the EP’s amended proposal stated that all on-call time, including inactive on-call time, should be counted as working time. However, under certain circumstances, the EP’s version allowed for the calculation of inactive on-call duty in a specific manner. The report does not specify what ‘specific manner’ exactly means. However, it can be assumed that it includes the possibility to count time spent inactive on-call not fully but to a lesser degree as working time. 22 The social partners reacted to the position of the EP with press releases. The European Trade Union Confederation (ETUC) welcomed the decision, while the Union of Industrial and Employers’ Confederations of Europe (UNICE) deplored it. 23
In its amended proposal, the Commission did not incorporate the most important points of the EP’s position and simply kept its original definition of time spent on-call not being working time. The Commission argued that, otherwise, no agreement could be reached in the Council of Ministers. 24 Nevertheless, subsequent discussions of the proposal in the Council in June and December 2005 did not go smoothly. Member States could not agree on what do to with the opt-out possibility. 25 The Council discussed the WTD again in June 2006 and in November and December 2007 without making a final decision. 26 The proposal basically affirmed the concept of ‘inactive on-call time’ as formulated by the Commission. Again, the press releases leave no doubt concerning the intention of the proposal as far as on-call duty is concerned, namely ‘[…] to avoid any consequences of the European Court of Justice’s case law […]’. 27
Only in September 2008 did the Council agree on a common position on the proposal by QMV. 28 Belgium, Cyprus, Greece, Hungary, Malta, Portugal and Spain did not support the proposal. Some of them stated the reasons for their rejection in a statement to the Council minutes. Of these states, all regret that the possibility for opt-out was still part of the proposal and ‘some of them still insist that the inactive part of on-call time cannot be considered as rest time’. 29 Despite voting in favour of the proposal, France shared the last objection and announced that it would not transpose the option of counting on-call duty as rest period contained in the proposal. 30 The Commission recognized that the common position of the Council and the line taken by the EP were difficult to reconcile. 31 Nevertheless, ‘[t]he Commission has supported the overall agreement [the common position of the Council], having regard to the urgent need to clarify the legal situation and thus to allow more coherent application of the Directive across all Member States’. 32 The list of differences is severe. The EP wanted to make clear that the provisions of the WTD apply per worker, not per contract. The Council was unable to reach an agreement on this point because half of the Member States used the per worker rule and the other half used the per contract rule, and neither half wanted to give up their respective practice. The Commission argued in favour of regulating this issue at a later time. Moreover, the common position left it to the Member States to either count inactive on-call duty as working time or as rest period. The common position of the Council retained the opt-out possibilities, a phasing out of this option was not included. As a reaction to this common position, the EP basically repeated its position from 2005 in the second reading of the proposal and rejected most changes made by the Council. 33 The Commission, in its opinion on this second reading in the EP, rejected the phasing out of the opt-out possibility and the calculation of the weekly work period per worker, but basically accepted the amendments concerning on-call duty. 34 The Council rejected all major amendments proposed by the EP and a conciliation committee under Article 251 TEC (now Article 294 TFEU). Shortly after, the committee ended its work without an agreement. No consensus could be reached on the opt-out possibility, the treatment of on-call duty and how to regulate multiple contracts. 35
In the meantime, the CJEU, unimpressed by this slow and difficult legislative process, continued to rule that inactive time spent on-call is working time. 36 In addition, the Court of Justice decided, in a case brought by the Commission, that the UK had failed to fulfil its obligations by allowing for more derogations than the directive did. 37 The CJEU also delivered a number of preliminary rulings concerning annual leave provisions. 38 In these cases, the CJEU decided predominantly in favour of the employees and against national or company practices.
In 2008, Eurofound, an EU agency tasked with providing information in the field of social policy, published a study on working time in order to provide some data for the discussion on the WTD. 39 Although it came too late to influence the decision-making process, it is the first in a series of studies that run parallel to the legislative process.
In conclusion, this five-year-long attempt to amend the WTD was a direct reaction to the rulings of the CJEU, which kept true, despite the political headwind, to its established line of reasoning: on-call duty spent at the workplace is working time in its entirety. The Commission argued, after the Council had rejected the amendments proposed by the EP, ‘[…] that the present situation regarding on-call time and compensatory rest still urgently requires clarification through legislative change’. 40 It was not long before the Commission started another attempt to amend the WTD.
4. The Commission and the social partners try again
In 2010, a year after the conciliation committee had failed to reach an agreement, the Commission again asked the social partners to initiate a social dialogue. 41 It did so in a rather comprehensive communication from the Commission to the EP, the Council, the EESC and the CoR explaining the need to reform the WTD. In this request, the Commission paints a picture of a new knowledge-based economy with new lifestyles and demand for more working time flexibility. Nevertheless, the Commission acknowledges that there are some sectors, namely those where the worker is under close supervision of the employer, that need safeguards against unhealthy working conditions. The issue of on-call duty is also addressed. The Commission proposes that: ‘For example, inactive periods of on-call time at the workplace could be disregarded when calculating working time. Or inactive periods could be calculated less than 100% as working time, proportionate to the level of attention required (the so-called equivalence system)’. 42 A list of questions concerning proposed changes addressed to the social partners concludes the communication. The Commission received replies from 47 business associations and worker organizations, a majority of which were officially recognized social partners. 43 While the business organizations argued for more flexibility, the worker organizations stressed the importance of collective bargaining. However, most of them were rather vague on the question of whether they wanted to open a social dialogue on the WTD or not. At the same time, the Commission fulfilled its obligation under Article 24 of the WTD and published a report on how the Member States regulate working time. 44 From this report it became clear that compliance with the WTD is, at best, shady. Often, Member States’ laws allow for more than the 48 hours weekly working time limit contained in the directive, and in some states the reference period is (contrary to Article 18 WTD) set to 12 months without collective agreement. Time spent on-call seems to be treated as working time in its entirety in about half the Member States, often with exceptions for certain sectors. In the other Member States, inactive on-call time is not fully counted as working time. The Member States also show low compliance with the rules on the rest period, allowing for all kinds of derogations, excluding specific sectors or delaying the prescribed compensatory rest period. The list of insufficient transposition of the WTD continues: in some states, doctors in training do not fall under the national transposition despite the fact that the WTD explicitly includes them; some states exclude police and armed forces despite the fact that the ECJ ruled that exceptions could only be made in times of crisis; in Italy even employers of libraries, museums and archaeological sites are excluded. The most bizarre feature of the WTD and its application is that it does not say whether the working time limit should be applied per worker or per contract. Although, the Commission recommends application per worker, 11 of the 25 Member States at the time of the report applied it per contract. The report states that while in the 2000 only the UK allowed for workers to opt-out, in 2010 the national legislation of 16 Member States included the possibility of some kind of opt-out. The Commission also identified minor problems concerning annual leave and night work.
Considering all of the shortcomings in transposition found in this 2010 report of the Commission, many of which have at their root the badly formulated WTD (missing definition of working time, no mention of whether it applies per worker or per contract, generous possibilities for opt-outs, etc.), a reform of the WTD seems more pressing than ever if one wants the WTD to be taken seriously. That the Commission continuously praises the efforts of the Member States in the report before it gives its long lists of shortcomings, seems to be more a diplomatic rhetorical device than coming from the heart. Accompanying these first steps towards a new attempt to reform the WTD, a number of studies on the WTD commissioned by the Commission were published. 45 The publication of these reports is a feature distinguishing this new attempt from earlier attempts to reform the WTD. As we will see, it will be in vain.
Following the invitation of the Commission, the social partners then started with a new attempt to amend the WTD in 2011. Foreshadowed by the irreconcilable positions displayed by the social partners in the previous attempts to amend the WTD, the talks were abandoned without result a year later.
Several cases on the WTD were decided by the CJEU between 2010 and 2013, the year the Commission reopened the consultation process. All of these cases were requests for preliminary rulings, except one, which was an infringement proceeding brought by the Commission. In this infringement proceeding, the ECJ found that Spain had failed in its obligations under the WTD by not applying it to civilian personnel in public authorities. 46 Of the preliminary rulings, a few concerned weekly rest and weekly working time, 47 a small number concerned the scope of the directive 48 and one concerned the definition of working time. 49 However, a majority of the cases concerned annual leave and sick leave. 50 Again, the CJEU primarily decided in favour of the employees.
To recapitulate, in this failed attempt to reform the WTD, the Commission changed its argumentation from wanting to counter rulings of the CJEU to overall developments of the economy that would call for an overhaul of the WTD. Although on-call duty was still an issue, it became less prominent as other issues received equal attention. And, indeed, the report on how the Member States regulate working time showed many more application problems than just the handling of time spent on-call. The report makes clear that many states do not even apply the WTD as interpreted by the CJEU to their own public employees. Moreover, whether the Directive is applied per worker or per contract seems to be a very important difference that the WTD does not address. Each Member State is thus free on this point to decide how to apply the directive, despite the Commission’s preference for an application per worker. The fact that the use of the op-out option has spread significantly is arguably a reaction to the strict and consistent interpretation of the WTD by the CJEU. It also does not bode well for demands coming from the EP and the employee organizations to terminate the possibility for opt-outs. A shift of focus can also be seen in the cases reaching the ECJ. Disputes about annual leave replaced disputes on the definition of working time.
5. The latest implementation report and the Commission’s interpretative communication
Shortly after the social partners failed to open a social dialogue, the Commission started again with a review process of the WTD. The Commission conducted a public consultation and assessed options for further action. 51 In the end, the Commission decided to apply a non-legislative strategy in the form of a communicative interpretation, which will be discussed below. During this time, the CJEU decided a number of cases on the WTD, starting with infringement proceedings against Ireland and Greece brought by the Commission. The case against Ireland was dismissed because the Commission did not provide enough evidence. 52 Greece, on the other hand, was found in breach of the directive for excluding doctors in public health services from its scope. 53 The other cases were preliminary rulings on the definition of working time 54 and on possible derogations. 55 In both cases the ECJ decided in favour of the employees. A case on on-call duty is still pending at the time of writing. 56
In April 2017, again following the instructions of Article 24 of the WTD, which calls for an evaluation of the application of the directive every five years, the Commission published another report on the implementation of the WTD. 57 This newest report does not read much differently from the one from 2010 discussed above. Some Member States apply the weekly working time limit per worker, others per contract and still others per worker, if the worker has more than one contract with the same employer, but per contract if the worker has contracts with different employers. Workers in the public sector especially are often still excluded from the scope of the directive by national legislation. Most Member States do not define the status of on-call duty and practices differ between Member States. The report states that Member States make use of the derogations allowed for by Articles 17, 18 and 22 of the directive concerning autonomous workers, compensatory rest and the opt-out possibility. In some cases, the Member States exceed the limits of what is allowed according to these Articles. Asked to give their view on the application of the WTD, the trade unions, represented by ETUC, express their dissatisfaction with the transposition of the WTD concerning on-call, compensatory rest, reference periods, op-out and derogation for autonomous workers. The employers, represented by BusinessEurope, on the other hand, express their satisfaction with how the directive is transposed in the Member States, but at the same time acknowledge the existence of problems of non-compliance with the rules on on-call time and rest periods in certain sectors. The report points to the failure of the Member States, with the exception of the UK, to provide an evaluation of the socioeconomic effects of the directive.
In May 2017, just a month later, the Commission issued a long non-binding interpretative communication on the WTD. 58 In this communication, the Commission stresses the importance of the WTD for European social rights in general and hopes that its communication will contribute to a better application of the WTD by creating legal clarity and certainty. The Commission identifies the many cases on the WTD brought to the CJEU as a source of confusion, ‘[…] since misunderstandings or lack of awareness of the latest developments in case-law may in turn lead to compliance issues and to avoidable complaints or litigation’. 59 The Commission tries to remedy this problem by incorporating the case law of the CJEU into the communication. It argues that the communication will help Member States to anticipate the behaviour of the Commission concerning potential infringement proceedings. 60 The communication deals with the scope of the directive, the definition of working time including on-call duty, the maximum weekly working time, paid annual leave, possible derogations and individual op-outs, or, in other words, with all articles of the WTD. Concerning the scope, the Commission repeats its point of view from earlier publications, namely that the WTD ‘should as far as possible, apply per worker’ and not per contract; 61 a rather weak formulation that, arguably, does not really create legal clarity or certainty as envisaged by the communication. The Commission continuously points out the room of discretion Member States have in applying the WTD, for example in determining when to grant the weekly rest periods, but also by pointing out that the Member States can, if they so wish, adopt rules more favourable for the health and safety of workers than the minimum requirements laid down in the WTD. The communication as a whole is a mix of interpretations of all the articles of the WTD based on rulings of the CJEU and of the Commission’s own view in cases where no CJEU ruling exists. Although the communication addresses all the controversial issues that have accumulated over the years (on-call, opt-outs, derogations, annual leave, etc.), many of them are only superficially resolved by stressing the flexibility inherent in the WTD. The fact that ‘[…] the Commission stresses that the interpretative communication is not binding and does not intend to create new rules’, 62 a feature which is naturally inherent in these kinds of interpretative communications from the Commission, makes clear that this communication is not the last word on the WTD as all the problems that come with the WTD still exist. Nevertheless, this elaborate communication shows the commitment of the Commission to improve the application of the WTD in the Member States.
In the end, the second report on the application made clear that nothing has changed in the seven years since the first report was published. 63 Neither could the conflict between the social partners be resolved, making a social dialogue impossible. In addition, the WTD reliably continues to produce court cases on a yearly basis. The interpretative interpretation is an attempt by the Commission to somehow bring an end to the continuous stream of unsuccessful initiatives to reform the WTD. In other words, after more than ten years of trying without any breakthrough, the Commission gave up.
6. Conclusions
The WTD is either being criticized for being too rigid or too flexible, a dichotomy reflected in the discussion to reform the WTD. 64 It is supposed to fit all kinds of industries, making it too rigid for certain industries. At the same time, full application can be avoided by use of the individual opt-out option and the numerous derogations, making it too flexible. However, this seems to be a conscious choice in order to give quite some room for discretion to the Member States, on the sides of both industry and to individuals. A more serious issue with the WTD is that from the start it left many important and basic issues unaddressed. For example, does the WTD apply per worker or per contract? If guaranteeing the health and safety of workers is the main goal, then it can only apply per worker; to apply it per contract would make no sense from this perspective, as the health and safety of workers with multiple contracts would be endangered. If giving workers flexibility, and at the same time a legal tool in case they feel exploited, then the per contract practice is plausible. The Commission clarified its own view in a number of documents, but only a decision of the CJEU on this point could bring legal certainty. An infringement proceeding against a Member State applying it per contract would give the CJEU the opportunity to clear things up. Moreover, the WTD still does not address time spent on-call, so that the rulings of the CJEU apply as long as employees have not opted-out. In recent history, the annual leave provisions seem to be the dominant issue before the Court of Justice. The CJEU repeatedly emphasized the importance of the (vertically directly effective) paid annual leave provision for EU social law, showing, as in the case of time spent on-call, that it is ready to defend the rights of employees against Member States’ practices. This firm stance and the shortcomings of the WTD give the ECJ quite some influence as it is asked to fill in the gaps, especially because the European legislators are deadlocked and are unable to provide guidance. The Community method proved to be the main obstacle to legislative overrule. 65 As the positions of the two sides of industry are equally irreconcilable, they could not break this political impasse either. 66 However, the other actors successfully avoid the consequences of such rulings as best as they can by making use of the derogations and opt-outs. Throughout its interpretative communication, the Commission points to unresolved issues. The story is, thus, far from over and the number of rulings on the WTD will increase further, maybe with the help of infringement proceedings initiated by the Commission, but for sure by preliminary rulings initiated by national courts.
