Top of page ↑


Østre Landsret (Danish Eastern High Court), June 2, 2017

2017/26 What is a collective agreement? (DK)

<p>On 2 June 2017, the Danish Eastern High Court decided that a statutory intervention by government was sufficient to enable derogation from the Working Time Directive (2003/88). The Directive can be derogated from by a collective agreement and although the statutory intervention was not a collective agreement, the High Court found that it was not inconsistent with that requirement.</p>


On 2 June 2017, the Danish Eastern High Court decided that a statutory intervention by government was sufficient to enable derogation from the Working Time Directive (2003/88). The Directive can be derogated from by a collective agreement and although the statutory intervention was not a collective agreement, the High Court found that it was not inconsistent with that requirement.

Legal background

Directive 2003/88/EC lays down certain working time standards for workers. For example, workers covered by the Directive are entitled to a daily 11-hour rest period and a weekly 24-hour rest period.

Some of the provisions in the Directive can be derogated from by law or by collective agreement. If the activity performed is one of the activities mentioned in Article 17 of the Directive, derogation can be made by means of law, regulation, administrative provisions or collective agreement. Also, a more general derogation can be made under Article 18 by collective agreement if the workers concerned are afforded appropriate protection.


In Denmark, almost all primary school teachers’ employment relationships are covered by a collective agreement. Special collective agreements have been entered into with regard to rest periods in general and rest periods during school camps, specifically. In those agreements, the parties have agreed to derogate from the provisions in Directive 2003/88 and the agreements have been in force since the mid-1990s.

However, in 2013 a major dispute erupted between the teachers’ organisations and the employer organisations (i.e. the Danish Ministry of Finance on behalf of the State and Local Government of Denmark, on behalf of the Danish municipalities). Going into the collective bargaining negotiations, the employer organisations had negotiated for fundamental changes to the teachers’ working time. The parties could not agree to renew the collective agreement between them and the employer organisations therefore locked out the teachers.

The lockout went on for four weeks before the Danish Parliament decided to make a statutory intervention. In Denmark, it has been a long-standing tradition that the Danish Parliament can end industrial action if essential social services are at stake. The consequence of a statutory intervention is, broadly, that the collective agreement is renewed and replaced by an Intervention Act. Normally, the legislature will try to strike the fairest balance possible to satisfy the interests of both sides of the dispute. Note however, that an intervention Act as a concept is hard to classify. It is not traditional legislation, nor is it a traditional collective agreement – it is somewhere in between.

The statutory intervention in this dispute broadly aligned teachers’ working time with the working time regime of public servants. The Act adopted in connection with the statutory intervention, replacing the existing collective agreements, contained provisions regarding school camps similar to the provisions in the previous collective agreements. However, with regard to the general daily rest period, it would be possible for the employer to reduce this more often than had been possible under the pre-existing collective agreements.

The teachers’ union sued the Danish Ministry of Employment, being the ministry responsible for the Intervention Act. The union claimed that the Act was inconsistent with Directive 2003/88.

The teachers’ union had two main arguments to that effect: firstly, the nature of school camps was not such as to allow for a derogation under Article 17(3) of Directive 2003/88. Secondly, in relation to the general daily rest period, the union argued that an Act adopted in connection with a statutory intervention replacing a collective agreement does not constitute a collective agreement, yet a collective agreement is required for a derogation under Article 18 of Directive 2003/88. Therefore, the Intervention Act failed to comply with the requirements for derogation as laid down in Directive 2003/88.

The Ministry of Employment responded that school camps constituted activities covered by Article 17(3) (b) and (c) of Directive 2003/88 and specifically relied on the ECJ’s judgment in Union Syndicale Solidaires Isère (C-428/09) in which the ECJ found that some workers employed under ‘educational commitment contracts’ to carry out casual and seasonal activities in holiday and leisure centres did fall within the scope of derogation under Article 17(3) (b) and (c) of Directive 2003/88. Therefore, the Intervention Act fully satisfied the requirements for derogation. In relation to the general daily rest period, the Ministry argued that Article 18 of the Directive provided the basis for derogation by collective agreement and that the Intervention Act was similar to such a collective agreement, since the outcome of a statutory intervention was to renew and replace a collective agreement.

The case was heard by the High Court in the first instance because of the fundamental nature of the case.

During the proceedings, the teachers’ union also asked the High Court to refer questions to the ECJ in the event that the High Court could not rule out the possibility that an Intervention Act could be considered to be of a similar nature as a collective agreement.


The High Court decided in favour of the Ministry of Employment.

The High Court agreed with the Ministry that school camp activities fall within the scope of Article 17(3) (b) and (c) based on the interpretation of this provision by the ECJ. Since activities that fall within the scope of Article 17(3) can be derogated from by means of law and regulation, the Intervention Act on this point was not inconsistent with the Directive.

With regard to the general daily rest periods – in relation to provisions that were not covered by the exemption in Article 17 and enabled employers to reduce the daily rest period more often than under the previous collective agreement – the High Court was faced with the question as to whether an Intervention Act can be put on par with a collective agreement under Article 18 of Directive 2003/88.

The High Court explained that the Danish labour market model allows for a high degree of freedom of contract between the social partners. The Government does not generally interfere in labour relations. It only does so if essential social services are at stake; if the industrial action has been going on for a substantial amount of time; and if there seems to be no hope that the disputing parties will manage to agree a renewal of the collective agreement.

After highlighting these characteristics of the Danish labour market model, the High Court went on to explain the substance of the Intervention Act in dispute. The High Court particularly noted that the purpose of the Act was the renewal and replacement of the pre-existing collective agreement in order to restore the no-strike commitment. The working time rules in the Act were left open to amendment between the parties (i.e. the municipalities and local teachers’ unions) and a majority of the local parties did actually enter into local agreements amending the Act in this respect.

In light of this, the High Court found that the Invention Act satisfied the requirements in Article 18 of Directive 2003/88 in relation to derogation by means of a collective agreement.

Finally, the High Court rejected the teachers’ union’s request for reference of questions to the ECJ, since the High Court did not find there was reasonable doubt as to the interpretation of the relevant EU law provisions at stake.

The High Court’s decision has been appealed to the Danish Supreme Court.


The statutory intervention regime is a rare phenomenon in an international context. It may be that only Norway has any similar practice. As explained above, the function of a statutory intervention is to bring an end to industrial action where important social services are at stake. The regime has developed over time and with the buy-in of the social partners in the Danish labour market. It is inferred from case law and is not rooted in legislation.

Over time, the ILO Freedom of Association Committee has expressed criticism in relation to various Danish statutory interventions, but the right for Parliament to carry out these interventions has not changed.

After the Intervention Act was passed in Denmark, the Danish National Teacher’s Union filed a complaint with the ILO Freedom of Association Committee. This happened in August 2013 and the Committee opened a case, but after hearing the Danish Government, it closed it. Nevertheless, the Committee did express criticism about the process used to carry out the statutory intervention. The Committee generally takes the view that any government interference in labour disputes is inappropriate, and does not accept social interests as a legitimate purpose for intervention. This is in conflict with the Danish approach, and despite the ILO’s criticism, the Danish Parliament has not refrained from interfering when lengthy industrial action has caused a potential threat to essential social services. In Norway, the Supreme Court decided in a judgment of 1997 (RT1997/580) that the Committee’s views on the limits of state interference in labour disputes did not have sufficient basis in the ILO conventions.

In addition, in human rights terms, the concept of statutory intervention is controversial, as the ECtHR has established that the right to take industrial action is autonomously protected under the right of freedom of association in Article 11 in the European Human Rights Convention. This was established in the Yapi-Yol Sen (Case no. 68959/01). The ECtHR has not yet decided a case on whether Danish statutory intervention is consistent with Article 11 of the Convention. However, Danish legal scholars would expect the ECtHR to have a less strict approach than the ILO Freedom of Association Committee, since the ECtHR has previously accepted the maintenance of essential social services as a legitimate reason to restrict a right protected by the Convention.

The case discussed in this case report is not likely to rock the boat with regard to the Danish tradition that the Government can interfere in disputes in the labour market. However, the case does have the potential to change our understanding of the nature of collective agreements and, particularly, whether a collective agreement is an autonomous EU law concept or something for Member States to define. But this question will most likely only be definitively answered if the Supreme Court decides to make a preliminary reference to the ECJ. It is well-known that principles laid down in EU directives can be derogated from by means of collective agreements or be implemented by means of collective agreements, but the ECJ has not yet expressed whether a ‘collective agreement’ is an autonomous concept in EU law.

In Österreichischer Gewerkschaftsbund (C-328/13), Advocate General Cruz Villalón found that a collective agreement was not an autonomous concept of EU law in relation to the Transfer of Undertakings Directive (2001/23), but in the end, the ECJ did not express itself as explicitly as the Advocate General.

At the end of the day, the question the High Court faced in relation to whether the Intervention Act rightfully derogated from Directive 2003/88 in accordance with Article 18 was whether a regulatory intervention can be placed in the same category as a collective agreement – bearing in mind that an intervention Act is neither traditional legislation, nor is it a collective agreement as such. With its judgment, the High Court seems to have taken the view that it shares more common characteristics with a collective agreement than with traditional legislation and it therefore satisfied the requirement of a “collective agreement” as referred to in Article 18 of Directive 2003/88. Since the concept of statutory intervention is not a common European phenomenon and that there seems to be no definition of a collective agreement in EU law, it is notable that the case was not referred to the ECJ.

As mentioned, the case has been appealed to the Supreme Court and it will be interesting to see if the Supreme Court agrees with the views on collective agreements and regulatory interventions laid out in the High Court’s judgment. With the Ajos case in mind (see 2017/14 for a case report about this judgment, published in the EELC 2017-2 issue), and given the Supreme Court basically ignored a decision from the ECJ in the Rasmussen case (C-441/14), it is possible that the Supreme Court will decide simply not to refer a question. Or to disagree with it.

We will update readers with a case report if the Supreme Court reaches a different conclusion from that of the High Court, though no decision can be expected before 2019.

Subject: Collective labour law, collective agreements

Parties: The Danish National Union of Teachers – v – the Ministry of Employment

Court: Østre Landsret (Danish Eastern High Court)

Date: 2 June 2017

Case number: B-2453-14

Hard Copy publication: Not yet available

Internet publication: Available from