Tuesday, 16 July 2019

Vorsprung durch Technik? Audi scores victory in trade mark appeal before the EU General Court

Alexandros Antoniou, University of Essex School of Law, a.antoniou@essex.ac.uk

On 12 July 2019, the EU General Court (GC) dismissed an appeal (Audimas v EUIPO - Audi (AUDIMAS)) from a Lithuanian sportswear company, whose trade mark was successfully opposed by the German automobile manufacturer Audi.


In October 2014, the applicant, Audimas AB, obtained through the World Intellectual Property Organisation (WIPO) an international registration for the mark displayed below, designating the European Union (EU) as one of the protection territories.

The mark above represents the applicant company’s name in black font, with stylised open and closing brackets right above the word element. Registration was sought for classes 18, 25 and 35 of the Nice Agreement, covering a wide variety of leather goods, clothing, footwear and headgear as well as advertising and business management services. In June 2015, the international registration designating the EU was notified to the EU Intellectual Property Office (EUIPO), pursuant to the provisions of the Council Regulation 207/2009 on the Community Trade Mark (now replaced with Regulation 2017/1001 on the EU Trade Mark).

In August 2015, the German automotive company AUDI AG opposed the mark applied for on the basis of its previously registered EU word mark ‘AUDI’ for goods and services falling in classes 18, 25 and 35 of the Nice Agreement as well as class 12, which covers vehicles and vehicle components. Audi claimed, in particular, infringement of Article 8(1)(b) of the 2009 Regulation, i.e. invalidity based on ‘relative grounds’ which relate to conflicts with earlier trade mark rights that belong to third parties (these provisions are preserved under the new 2017 Regulation). The Opposition Division of the EUIPO upheld the opposition two years later. Audimas appealed the decision in November 2017 but the Office’s Second Board of Appeal rejected the appeal in May 2018 (‘the contested decision’).

Specifically, the Board of Appeal found that the relevant public in this case consisted of ‘professional customers' and 'end consumers', whose level of attention varied from medium to high. In addition, the signs at issue were broadly similar at least to the extent that Audi’s earlier mark was reproduced in full in the dominant element of the applied-for mark. The Board also considered that a Spanish-speaking consumer would break down the mark applied for in two verbal parts, i.e. ‘audi’ and ‘mas’, because the latter element alluded a meaning to them, namely 'more' or ‘plus’. Moreover, the figurative element of the brackets was found to be ‘purely ornamental’ and its combination with the term ‘mas’ meant that the mark in question was at best only weakly distinctive. The Board of Appeal ultimately concluded that there was a likelihood of confusion for the Spanish-speaking part of the relevant public within the meaning of Articles 8(1)(b) of the 2009 and 2017 Regulations.

The legal framework and applicable principles

Under both Regulations 2009 and 2017, an opposition must be based on rights held by the opponent in an earlier trade mark (or other form of trade sign). The grounds on which an opposition can be based are called ‘relative grounds for refusal’, the relevant provisions of which are found in Article 8 the Regulation. By contrast to ‘absolute grounds for refusal’, which are examined ex officio by the EUIPO, relative grounds for refusal are inter partes proceedings based on the likely conflict with earlier rights. This means that the burden falls on the owner of earlier rights who needs to be vigilant in checking the filing of potentially conflicting EUTM applications and oppose the registration of marks when necessary.

More specifically, under Article 8(1)(b):

upon opposition by the proprietor of an earlier trade mark, the trade mark applied for shall not to be registered if, because of its identity with or similarity to the earlier trade mark and the identity or similarity of the goods or services covered by the trade marks, there exists a likelihood of confusion on the part of the public in the territory in which the earlier trade mark is protected. The likelihood of confusion includes the likelihood of association with the earlier trade mark.

Audimas contested the Board of Appeal’s findings in relation to the comparison of the signs and the existence of a likelihood of confusion. According to the established case law of the Court of Justice on the interpretation of Article 8(1)(b), the risk that the public might believe that the goods or services in question come from the same undertaking or from economically-linked undertakings, constitutes a likelihood of confusion (see Canon [1998] EUECJ C-39/97). The likelihood of confusion on the part of the public must be appreciated globally, according to the perception of the marks in the mind of the average consumer of the goods or services concerned. It is also settled case law that the average consumer normally perceives a mark as a whole and does not proceed to examine its various details (see Sabel [1997] EUECJ C-251/95). Consequently, the visual, aural and conceptual similarities of the marks must be assessed with reference to the overall impression created by the marks bearing in mind their distinctive and dominant components. Account must be taken of all factors relevant to the circumstances of the case and, in particular, the similarity between the marks and the goods or services. A lesser degree of similarity between those goods or services may be offset by a greater degree of similarity between the marks, and vice versa (see Lloyd Schuhfabrik Meyer [1999] EUECJ C-342/97). Where the earlier trade mark has a highly distinctive character, such as Audi’s mark in the present case, either because of its intrinsic qualities or because of the use that has been made of it, the likelihood of confusion is greater.

The level of attention of the relevant public

The GC disagreed with the applicant’s claim that buyers of Audi cars would demonstrate an increased level of attention, which would in turn decrease the likelihood of confusion between the marks at issue. Indeed, the prospect of an expensive purchase, e.g. in the case of luxury products, is a good reason for the average consumer to be more circumspect in relation to the origin and the quality of the articles concerned. A high degree of attention is likely to be displayed during the purchasing process of a specific product which is generally regarded as reflecting its owner’s social status. Nevertheless, both Audi’s earlier mark and Audimas’ applied-for mark extended to clothing and footwear as well. These are mass consumption goods, which are frequently purchased and used by the average consumer who nevertheless still pays a fair degree of attention in choosing them, albeit not above average (see Esprit International v OHMI-Marc O'Polo International [2011] EUECJ T-22/10). Thus, the Board of Appeal was not wrong in assessing the relevant public’s attention in the present case as varying from medium to high.

The goods and services concerned

The goods and services covered by Audimas’ mark were deemed at least in part identical or similar to those covered by Audi’s earlier mark. This conclusion was not disputed by the applicant and was approved by the GC.

The comparison of the signs and the likelihood of confusion

The GC agreed with the Board’s ruling that the dominant element of the applied-for mark was the fanciful term ‘audi’, since the second verbal element of the applied-for mark, i.e. ‘mas’, was already well-known to Spanish speakers and had a lower distinctive character. Moreover, the figurative element of the brackets was not, in the GC’s view, ‘particularly original or elaborate’. As such, it did not add anything ‘striking’ to the overall impression created by it in the perception of the relevant public.

The visual similarity between the marks was reinforced by the fact that both shared the ‘audi’ element which constitutes the earlier mark and is found at the beginning of the applied-for mark. This is consistent with previous case law, according to which the fact that a mark consists exclusively of the earlier mark to which another word is attached is an indication of the similarity between those two marks (see Fon Wireless v EUIPO-Henniger [2016] EUECJ T-777/14). On the phonetic level, the degree of similarity between the marks was found to be essentially ‘greater than average’, notwithstanding the different pronunciation given to the contested mark by the syllable ‘mas’. It is also a settled principle that both verbally and phonetically the average consumer generally pays more attention to the beginning of the mark than its ending, since the first part of a trade mark normally has a greater impact than the final part (see L'Oréal v OHMI-SPA Monopole [2009] EUECJ T-109/07 and Gappol Marzena Porczynska v EUIPO-Gap [2017] EUECJ T-411/15). Finally, the GC noted that the absence of conceptual similarities between the two signs should not deflect attention from their substantial visual and phonetic similarities and concluded that they were ‘broadly similar’.

In the GC’s view, the Board’s conclusion that the Spanish-speaking part of the relevant public would eventually establish a link between the contested mark and Audi could not be called into question. The Board of Appeal had held, and the GC agreed, that it could not be ruled out that the applied-for mark, with its element ‘mas’ attached to the word ‘audi’, could be taken to identify an exclusive series of goods or services of the Audi family of brands. The relevant public could perceive the ‘Audimas’ mark as a ‘particular variant’ of the earlier mark for a specific type of goods with a positive quality. In light of the foregoing, the GC dismissed the applicant’s action.


In sum, the GC upheld the original EUIPO decision, ruling that the applied-for mark by Audimas would likely be confused with Audi’s earlier trademark. The upheld contested decision is undoubtedly a victory for Audi, one of the world leaders in the automotive sector. On the other hand, Audimas has become, since its incorporation in 1931, one of the market leaders in the design and manufacture of sports and active lifestyle apparel in the Baltic States. The brand has been cooperating with the Lithuanian National Olympic Committee as an official supplier of sportswear for the Lithuanian Olympic family for more than 15 years. In 2013, Audimas also began sponsorship of the Belarus National Olympic Committee. This is certainly a displeasing for them outcome, which might be appealed to the Court of Justice of the EU.

Despite Audi’s weak connection to the clothing industry, the GC’s ruling can hardly be a surprise. It is grounded in some well-established principles relating to the comparison of signs in trade mark disputes. When assessing their similarity, the marks in question will be considered as a whole. Although it is not possible to isolate and focus exclusively on one component of a complex mark and compare it with another mark, the assessment of similarity may be made solely on the basis of the dominant component of a complex mark where all the remaining components of the mark – like the brackets and the element ‘mas’ in this case – add very little to, or are negligible in, the overall impression produced by it.

In addition, where a complex mark comprises word and figurative elements, the former would in principle be considered more distinctive than the latter, because the average consumer tends to refer to the goods or services in question by quoting the name of the mark in question, rather than by describing its figurative element (see Coca-Cola v OHMI-Mitico [2014] EUECJ T-480/12). The shared ‘audi’ element was incapable in this instance of lending the applied-for mark a distinctive character and accentuated the likelihood of confusion. Also, the presence of a few different syllables is not always enough to exclude the existence of a phonetic similarity between two signs. In the present case, the phonetic difference between the signs at issue, resulting from the addition of a second syllable ‘mas’ in the applied-for mark, was not sufficient to overcome and preclude the phonetic similarity between ‘Audi’ and ‘Audimas’ taken as a whole.

These opposition proceedings also confirm that, if a likelihood of confusion between two conflicting rights relating to the EU is established in a specific linguistic area, this is enough for the registration of the later mark to be refused. When an opposition is filed pursuant to Article 8(1)(b) and a likelihood of confusion can indeed be found on a substantial part of the public, the reasoning of the decision typically concentrates on that part of the public that is most prone to confusion, making the examination of the perception of the marks in several languages redundant. In addition, where the relevant public consists of both professional and general consumers, the finding of a likelihood of confusion in relation to just one part of the public is enough to uphold an opposition.

Finally, the outcome serves as a useful reminder of the need to conduct comprehensive trademark searches prior to filing a trade mark application, which needs to be carefully tailored to in order to maximise the chances of a successful registration. Opposition hearings can be very costly to defend. If unsuccessful, an applicant will have to pay not only their legal costs but some of the costs of the other party too – and will still not be able to achieve registration of their mark.

Tuesday, 2 July 2019

Should the EU sanction its Member States for breaches of rule of law and human rights? Part 1: The Legal Framework

Professor Steve Peers, University of Essex

I’ve taught EU law and human rights for over twenty years now, and the issue of sanctions against Member States for human rights breaches used to be the easy bit. Why? Because the procedure to enforce such sanctions (set out in Article 7 TEU) had never been used – and there was no apparent prospect that it ever would be. So there was no need to discuss it in any detail. A more theoretical sort of academic might have spent time counting the angels on the head of this constitutional pin, but I was anxious to move on to the real world issues of arrest warrants and asylum seekers.

Everything has since changed. Like Article 50 – which similarly raises fundamental issues about the EU’s relationship with its Member States – Article 7 was apparently dashed off in previous Treaty amendment talks without much thought to its detailed application in practice, perhaps because its authors thought it would never be used. Yet here we are, with both Articles now a live political and legal issue: the Ragnarok of EU law.

There are two recent parallel major developments. First of all, the Article 7 process has been triggered both against Poland (by the European Commission) and Hungary (by the European Parliament). Secondly, there are case law developments raising general questions about Member States’ observance of human rights and the rule of law outside the very specific (and very political) Article 7 process. In this context, last week the CJEU delivered its first judgment that a Member State is infringing judicial independence by means of reforms to its judicial system (see discussion here).

The prospect of the EU sanctioning its Member States for breaches of human rights and the rule of law raises a number of fundamental legal and political issues – and is best understood in a broader historical context. In light of the recent developments (and ongoing disputes), this is an opportune moment to provide an overview and analysis of this issue.

I’ll do this in a series of three blog posts, addressing in turn:

a)      the legal framework for sanctions under Article 7
b)      the overlap of the sanctions rules with other aspects of EU law
c)       the historical context and broader constitutional dynamics.

The legal framework for sanctions

Although many people refer to Article 7 TEU, there are other Treaty provisions which are inextricably linked: Article 2 TEU sets out the values which Article 7 is used to enforce; Article 354 TFEU describes voting rules for the EU institutions; and Article 269 TFEU provides for limited jurisdiction for the CJEU over the sanctions procedure.  All of this must be distinguished from the normal rules of EU law, discussed in the second blog post.

First of all then, what are the values of the EU, legally speaking? Article 2 TEU states:

The Union is founded on the values of respect for human dignity, freedom, democracy, equality, the rule of law and respect for human rights, including the rights of persons belonging to minorities. These values are common to the Member States in a society in which pluralism, non-discrimination, tolerance, justice, solidarity and equality between women and men prevail.

Article 7 then sets out the process of enforcing those values. It begins with Article 7(1), which provides for a kind of ‘yellow card’ – a warning if there is there is ‘a clear risk of a serious breach’ of those EU values:  

1. On a reasoned proposal by one third of the Member States, by the European Parliament or by the European Commission, the Council, acting by a majority of four fifths of its members after obtaining the consent of the European Parliament, may determine that there is a clear risk of a serious breach by a Member State of the values referred to in Article 2. Before making such a determination, the Council shall hear the Member State in question and may address recommendations to it, acting in accordance with the same procedure.

The Council shall regularly verify that the grounds on which such a determination was made continue to apply.

Notice that the ‘yellow card’ process can be triggered by the European Parliament, or a group of Member States, or the Commission. There is no requirement of unanimity of Member States to approve a Council decision to issue a ‘yellow card’ (this is a common misunderstanding), but the threshold of four-fifths of Member States’ governments in the Council is nevertheless fairly high. Triggering the process (as the EP did for Hungary, and the Commission did for Poland), does not, as some think, mean that the Council will agree to issue a ‘yellow card’, or has done so already. Indeed, the Council is still considering the proposals to issue a ‘yellow card’ against both Poland and Hungary, having held several of the hearings referred to in Article 7(1). If the Council ever did issue a ‘yellow card’, note that this does not entail a sanction as such: it is only a finding of a risk to EU values, with possible recommendations. Nevertheless, the issue of a ‘yellow card’ is perceived as extremely politically serious.

This brings us to Article 7(2), which is the ‘red card’ of the process:

2. The European Council, acting by unanimity on a proposal by one third of the Member States or by the Commission and after obtaining the consent of the European Parliament, may determine the existence of a serious and persistent breach by a Member State of the values referred to in Article 2, after inviting the Member State in question to submit its observations.

The procedure here is even tougher: unanimity of the Member States. The European Parliament cannot trigger the process, but could veto it  if the Commission or a group of Member States trigger it. The threshold to be met is higher: not just the risk of a serious breach, but the ‘existence of a serious and persistent breach’ of those values. It’s likely that the EU would get to the ‘red card’ stage after issuing a ‘yellow card’, but that’s not a legal requirement: a ‘straight red’, for (say) a country which had suddenly undergone a military coup, is also conceivable.

What are the consequences of a ‘red card’? Article 7(3) sets them out:

…the Council, acting by a qualified majority, may decide to suspend certain of the rights deriving from the application of the Treaties to the Member State in question, including the voting rights of the representative of the government of that Member State in the Council. In doing so, the Council shall take into account the possible consequences of such a suspension on the rights and obligations of natural and legal persons.

The obligations of the Member State in question under the Treaties shall in any case continue to be binding on that State.

Notice that the Member States don’t have to act unanimously in the Council when deciding exactly what sanctions to apply to the black sheep amongst them. The unanimity threshold only applies when taking the previous step of deciding whether there’s a serious and persistent breach of the EU values.  As for the specific sanctions which might be imposed, the Treaty mentions suspension of voting rights, but that’s just one example. The Council might instead (or additionally) impose other sanctions, such as suspension of MEPs’ voting rights (which raises the awkward question of whether they might also end up sanctioning any opposition MEPs from the Member State in question – whose voices would ideally need to be heard). However, there’s an obligation to consider the rights of individuals and businesses, which suggests that trade sanctions might be problematic. It might also be hard to justify restricting free movement rights, but in any event note that there are specific rules on asylum for EU citizens fleeing from a Member State subject to a ‘red card’. (I discuss them further in the second blog post).

Most significantly, there’s no provision to expel a Member State from the EU as such. Having said that, a Member State subject to suspension might be so outraged to be in that position that it triggers the process of leaving the EU under Article 50. The UK’s withdrawal process has been complicated and controversial enough; now imagine the legal and political complexities of a Member State subject to an Article 7 ‘red card’ triggering Article 50. Should its political authorities’ actions be considered legally and morally valid? What if a group of exiles claim to be the legitimate government of that Member State (a la the USSR-era Baltic States), and that purported government does not wish to leave the EU?  What if a part of that Member State, at odds with the government in power over EU membership and its violation of EU values, attempts to secede?

Of course, the possibility of withdrawal (alongside concerns about sovereignty, and the workings of partisan politics) may also have influenced the pronounced reluctance of the EU to use the Article 7 process. Does the EU really want Michel Barnier’s main task to be crowd control?

Article 7(4) TEU then provides that the Council, again by qualified majority, may ‘vary or revoke’ its sanctions against a Member State ‘in response to changes in the situation which led to their being imposed’. Article 7(5) notes that the rules on voting within the institutions when Article 7 is being applied are set out in Article 354 TFEU. The latter provides that the Member State which is the subject of potential sanctions has no vote at any stage of Article 7, as otherwise this would obviously have made the adoption of any decision on breach of EU values impossible. Abstentions cannot prevent the adoption of a ‘red card’ decision. Where the Council votes to implement a ‘red card’ decision, a higher threshold for adopting EU laws applies (72% of participating Member States in favour, instead of the usual 55%). If a Member State’s voting rights are suspended, the usual rules on Council voting with only some Member States participating apply. For its part, the EP ‘shall act by a two-thirds majority of the votes cast, representing the majority of its component Members’.

Finally, Article 269 TFEU significantly limits the role of the CJEU over the sanctions procedure:

The Court of Justice shall have jurisdiction to decide on the legality of an act adopted by the European Council or by the Council pursuant to Article 7 of the Treaty on European Union solely at the request of the Member State concerned by a determination of the European Council or of the Council and in respect solely of the procedural stipulations contained in that Article.

Such a request must be made within one month from the date of such determination. The Court shall rule within one month from the date of the request.

The legal issues

Given the limit on judicial control of the Article 7 process, it is almost entirely political. So the legal questions arising from it may be largely hypothetical in practice. However, they do exist.

The first important point is the wide scope of issues which can be the subject of the Article 7 process. It is sometimes claimed that the process can only be used to sanction Member States for breaches of EU law, but this is clearly false. There is no reference to EU law breaches in Articles 2 or 7. Indeed, such a limit on the scope of Article 7 would be odd, given that Article 269 TFEU limits the Court’s jurisdiction, yet other provisions of the Treaties (discussed further in the next blog post) give the Court extensive jurisdiction over the enforcement of ordinary EU law.

This claim about the limited scope of Article 7 is also absurd if you consider the broader context. Imagine, for instance, a Member State placing LGBT citizens in concentration camps. A narrow interpretation of Article 7 would mean that the EU could only complain about this to the extent that being locked up in camps would have a discriminatory effect on the detainees’ access to employment. Yes it would, but that would hardly be the most outrageous aspect of detaining LGBT people in camps because of their sexual orientation. (EU law is also relevant to LGBT refugees, but the Article 7 process would have to be triggered first for it to be relevant to refugees who are EU citizens).

So obviously Article 7 is not intended to be limited in this way. Indeed, its broad scope partly explains why the CJEU’s jurisdiction is limited – to avoid giving it jurisdiction to rule on issues which are not normally within the scope of EU law. (Another reason is the intention to keep the Article 7 process in the hands of politicians, not judges).

On the other hand, the Article 7 process and ordinary EU law can overlap. The Court can use its ordinary jurisdiction to rule on an issue being discussed in the Article 7 process, and vice versa. This was confirmed implicitly in last week’s judgment on Poland and the rule of law, given that the issues in that judgment also formed a part of the Commission’s Article 7 case against Poland. In fact, the Advocate-General’s opinion addressed the overlap explicitly (paras 48-50), arguing that ‘[t]here are firm grounds for finding that Article 7 TEU and Article 258 TFEU are separate procedures and may be invoked at the same time’.  As noted already, this alternative option of using ordinary EU law to restrain Member States’ breaches of human rights or the rule of law is discussed in the next blog post in this series.

Exactly how does the Court’s limited jurisdiction over Article 7 work? The wording of Article 269 TFEU definitely covers the decisions on the ‘yellow card’ or the ‘red card’. At first sight, it also applies to the implementation of sanctions, since the text refers to any Council actions pursuant to Article 7 TEU. But on this point, the use of the word ‘determination’ is confusing, as Article 7 doesn’t use that word to refer to the implementation of sanctions,  but only the decisions on whether EU values have been (or might be) breached.

Note also that the only possible challenger is the Member State sanctioned under Article 7 – not any other Member State, an EU institution, or an individual or business. If individuals are barred from challenging the validity of Article 7 implementation decisions, even indirectly via national courts to the CJEU, how else can the Council’s obligation to ‘take into account the possible consequences of such a suspension on the rights and obligations of natural and legal persons’ be enforced? At any rate, there’s no limit on the Court being asked by national courts to interpret the decisions implementing sanctions, which could be significant in working out the impact of sanctions on individuals. In particular, if Council decisions under Article 7 disapply ordinary EU law in some way, there should be no objection to the Court’s ordinary jurisdiction to interpret such ordinary EU law applying.

A Court judgment under Article 269 TFEU can only address procedural issues, not substance. In other words, the Court cannot be asked to rule on the question of whether the Member State concerned has actually breached EU values (or seriously risks breaching them). As we will see in the second blog post, however, the developing case law on the overlap between Article 7 and ‘ordinary’ EU law renders this firewall a little diffuse. Also, one can imagine that a Member State may make arguments about the fairness of the hearings, even where (as in the case of Poland and Hungary) some hearings have been held. Finally, the time limits in Article 269 require significant fast-tracking: the challenge must be made one month after the determination (the usual deadline to bring an action to challenge an EU act is two months after publication) and (uniquely in EU law) one month for the Court to give its ruling.  

Even though Article 7 has not resulted in any sanctions decision yet, some issues about its scope may be addressed in the near future, because Hungary has brought a legal challenge to the European Parliament’s decision merely to trigger Article 7. This case might be inadmissible, because usually it is not possible to challenge the start of an EU legal procedure, but only a legal act once adopted, which may explain why Article 269 TFEU makes no reference to challenging acts of the European Parliament at all (or indeed, to challenging acts of the Commission or the Member States). The substance of the Hungarian government’s argument is that the European Parliament wrongly ignored abstentions when counting votes cast to trigger the Article 7 process.

One key legal and political question is the interpretation of the unanimity requirement to issue a ‘red card’ determination of a serious breach of EU values. Some have suggested that since two Member States are facing Article 7 procedures, and they would have a natural tendency to stick together and vote for each other, unanimity can never be reached. Therefore, for the ‘red card’ procedure to be effective, it must be interpreted to mean that any Member State facing an Article 7 procedure must lose its vote even as regards issuing a ‘red card’ against another Member State.

With respect, this interpretation is untenable. Article 354 TFEU refers to ‘the Member State in question’ not voting in its own case – clearly using the singular, as well as the definite article. There is no way to stretch the canons of interpretation for this to refer to multiple Member States. Such wild leaps of legal fancy are particularly inappropriate when a main point of the process is to ensure protection of the rule of law in the European Union.


Article 7 TEU recently turned 20 years old. It was conceived as a political process par excellence, and it remains supremely political at childhood’s end, even as the first attempts to trigger it are made. Due to its impact on national sovereignty, and the web of transnational partisan politics in which the governments concerned are embedded, Article 7 has long been seen as a ‘nuclear weapon’ – only to be used as a last resort, in a political emergency such as a military coup. Although the attempt to nuance Article 7, by adding a ‘yellow card’ process, dates from 2003, in practice this version of the process is perceived as politically ‘nuclear’ too.

The obvious problem here – which the ‘yellow card’ reform sought but failed to address – is that democracy rarely collapses overnight. In the famous words of Michael Rosen, ‘people think that fascism arrives in fancy dress’, but in fact ‘it arrives as your friend’ – promising to:

…restore your honour, 
make you feel proud, 
protect your house, 
give you a job, 
clean up the neighbourhood, 
remind you of how great you once were, 
clear out the venal and the corrupt, 
remove anything you feel is unlike you...

And to that end, and for those reasons, it often gains a foothold through the democratic process. Yet the values of the EU to be protected also include democracy – and the Article 7 process is in the hands of the governments of fellow Member States. All have some skeletons in their own closet; and all have backs that might need some scratching by the governments of the States being criticised.

So is the Article 7 process doomed? In fact, the expansion of EU law in areas with significant relevance to human rights – and the willingness of the CJEU to rule on the judicial independence of national courts in general – means that recourse to the nuclear option may arguably not be necessary. In effect, the conflict over the protection of human rights and the rule of law in Member States can also be fought by conventional means: the ordinary system for the enforcement of EU law as such, to which we will turn in the second post in this series. As for the broader tension when concerns about the rule of law and human rights stem from a democratic outcome, this will be assessed as part of the broader discussion in the third post.  

Barnard & Peers: chapter 9
Photo credit: euobserver

Sunday, 30 June 2019

The beginning of the end for Poland’s so-called “judicial reforms”? Some thoughts on the ECJ ruling in Commission v Poland (Independence of the Supreme Court case)

Laurent Pech, Professor of European Law, Middlesex University London

Sébastien Platon, Professor of Public Law, University of Bordeaux

Imagine a faraway land where a government and a parliament dominated by the same party decide to retroactively lower the retirement age of the judges working for the country’s supreme court. Imagine that this change is being presented as a “reform” (allegedly) needed to hold to account judges (allegedly) “shamefully involved” in the country’s previous communist regime which however ended about thirty years ago. Imagine that all of this done with authorities claiming “there is nothing going on in [the country] that contravenes the rule of law” as “judges should always be on the side of the state”.

Surely we cannot be talking about a country belonging to the EU. Sadly, you would be wrong to think so. Indeed, the retirement measure described above was at the heart of the infringement action initiated by the European Commission against Poland last July and which resulted, earlier this week, in the European Court of Justice’s first ruling on the compatibility with EU law of one of Poland’s so-called “judicial reforms”. 

This post will explain the extent to which the Court’s ruling may be considered a landmark one, and the Court’s main findings, before assessing the ruling’s immediate and potential impact.

1. A landmark ruling

For the very first time, the Court of Justice has found a national government to have failed to fulfil its obligations under the second paragraph of Article 19(1) TEU which provides that “Member States shall provide remedies sufficient to ensure effective legal protection in the fields covered by Union law.” This is far however from the only “EU law first” one may “credit” to the current Polish authorities:

-          Poland was indeed the first ever EU Member State to be subject to the Commission’s Rule of Law Framework in January 2016;
-          The first EU Member State to be threatened with the payment a fine of at least €100,000 per day in November 2017 by the ECJ should it continue to ignore an interim order adopted by the same Court in July 2017;
-          The first EU Member State to be subject to Article 7(1) TEU proceedings in December 2017;
-          The first EU Member State to have seen its “judicial reforms” provisionally suspended by the ECJ via two interim orders adopted in October and December 2018.

It has now become the first EU Member State to have been found by the ECJ to have failed to fulfil its Treaty obligations by violating both the principles of the irremovability of judges and judicial independence.

As will be shown below, the Court has forcefully and compellingly rejected each one of claims made by the Polish government, including the most recurrent one whereby the Court of Justice would lack jurisdiction to review the multiple, never-ending changes made to the organisation of the Polish judiciary.

2. The Court’s findings

2.1 Organisation of the Polish national justice system as an allegedly exclusive competence immune to EU review

According to the Polish government, supported by the Hungarian government, the organisation of the national justice system constitutes a competence reserved exclusively to the Member States, which would imply that that EU institutions, including the Court of Justice, cannot examine Poland’s “judicial reforms” in light of EU law requirements.

The Court easily explains why this argument cannot survive any serious scrutiny and does so by initially and unusually reminding the Polish government that “as is apparent from Article 49 TEU, which provides the possibility for any European State to apply to become a member” of the EU, Poland “freely and voluntarily committed” itself to respecting and promoting “the common values referred to in Article 2 TEU”, including the rule of law (§ 42). Furthermore, while the Court agrees that “the organisation of justice in the Member States falls within the competence of those Member States”, this obviously cannot be construed as a carte blanche to violate its EU law obligations, not to mention the fact that requiring Poland to comply with its EU law obligations is not akin in any way to exercising “that competence itself” (§ 52). This is merely a reminder of the longstanding difference between the applicability of EU law and the competence of the EU, which explains why EU law can apply in situations where the EU has no competence to legislate.

With respect to the Polish government’s argument that the EU principle of judicial independence can be applicable only in situations governed under EU law, the Court merely reiterates what it previously held in the Portuguese judges ruling. National authorities must respect the principle of judicial independence even in situations where national “judicial reforms” do not implement EU law. Article 19(1) TEU indeed covers any national court which may rule “on questions concerning the application or interpretation of EU law”, in which case any national measure affecting the independence of the said court falls within the fields covered by EU law.

In the present case, it was obvious that Article 19(1) TEU was applicable as it was common ground that Poland’s Supreme Court “may be called upon to rule on questions concerning the application or interpretation of EU law and that, as a ‘court or tribunal’, within the meaning of EU law, it comes within the Polish judicial system in the ‘fields covered by Union law’ … so that that court must meet the requirements of effective judicial protection” (§ 52). It follows that Polish authorities cannot adopt measures which undermine its independence without activating the application of EU law.

2.2 The lowering of retirement age as allegedly required to bring the Supreme Court’s retirement regime in line with the general retirement regime

According the Polish government, it follows from the Court’s own case law, that “that the Member States retain the option to adapt the employment conditions applicable to judges and, thus, their retirement age, in particular in order, as in the present case, to bring that retirement age into line with that provided for in the general retirement scheme, while improving the age structure of officers of the court concerned” (§ 67).

Anyone familiar with the situation in Poland would have immediately found this defence rather surprising. Indeed, the ruling party “has long rallied against what it calls a self-serving “caste” of judges who distort justice for ordinary citizens”. Furthermore, the Polish government’s own “White Paper” of March 2018 indicates that “the reform of judicial retirement age is justified with historical experiences of communism, the failure to account for the past for many years, and pathological [sic] mechanisms of the functioning of courts that have been perpetuated for years” (para 99).

Not unsurprisingly, the ECJ easily came to the conclusion that the forced early retirement of Supreme Court judges is not compatible with the principle of irremovabilily, which is a guarantee of independence. While phrasing this delicately, the Court all but explicitly states that the Polish government has deliberately sought to mislead it when it refers to the information contained in the “explanatory memorandum to the draft New Law on the Supreme Court” and on the basis of which one may have “serious doubts as to whether the reform of the retirement age” was not in fact made “with the aim of side-lining a certain group of judges of that court” (§ 82).

The Court could have stopped there but if only to make it clearer to any future government which might be tempted to follow a similar path, the Court proceeds to perform a proportionality test. In a few words, the Court sees no reason why, for the sake of standardising retirement age, the judges of the Supreme Court should be forced into retirement when all other workers have a right to retire (or not) before holding that the lowering of retirement age with immediate effect, without any transitional measure, is in any event disproportionate.

2.3 Discretionary prerogative granted to the Polish President as (allegedly) required to protect the judiciary

While possibly difficult to believe, the Polish government claimed that “the authorisation conferred on the President of the Republic to decide as to whether to allow” Supreme Court judges “to continue to carry out their duties once they have reached retirement age” constitutes a “prerogative, the specific purpose of which is to protect the judiciary both from interference by the legislative authority and from that by the executive authority” (§ 103).

In other words, it is suggested that the Polish President, itself part of the executive, is the best placed to protect the judiciary from interference by both the executive and legislative branches by deciding alone and in the absence of any substantive conditions, procedural rules and access to judicial review, when to authorise a Supreme Court judge not to be forcibly retired. This is by the way the same office holder who has deliberately ignored court orders, repeatedly attacked Polish judges while also questioning the independence of the Polish ECJ judge and the authority of the ECJ. Even Kafka could not have imagined something more Kafkaesque.

For the Court, the inescapable conclusion is that “by granting the President of the Republic the discretion to extend the period of judicial activity of judges of that court beyond the newly fixed retirement age”, Poland has violated Article 19(1) TEU (§ 124). While explaining why this is so, the Court makes noteworthy observations in relation to the new “National Council of the Judiciary” (NCJ), arguably established in breach of the Polish Constitution and since suspended from the European Networks of Councils for the Judiciary (ENCJ) due to its lack of independence. In a nutshell, the Court explains that the prior involvement of the NCJ cannot “save” the presidential extension regime organised by the Law on the Supreme Court as the NCJ has proved unable to deliver properly reasoned opinions based on objective and relevant criteria to the President for the purposes of authorising Supreme Court judges to continue to carry out their duties.

3. Immediate and potential impact

In a strong editorial, the Financial Times described the Court’s ruling as a landmark one which “will help buttress the rule of law in the EU against authoritarian leaders who have been chipping away at democratic checks and balances with impunity”.

We agree with this assessment.

While the ruling addresses only one of the multiple serious rule of law problems identified by the European Commission in its Article 7 proposal, it does not merely fully confirm the accuracy of the Commission’s diagnosis in the present infringement action but also indirectly its general diagnosis regarding the growing systemic threat to the rule of law in Poland. This legal win is also bound to considerably strengthen the weight of the Commission’s arguments within the framework of ongoing Article 7 proceedings.

The Court’s ruling also establishes a solid de facto precedent with respect to any future attempt in Poland or elsewhere to take control of a court via a retroactive lowering of the retirement age of judges under false pretences. The ruling will similarly add to the growing body of evidence which shows repeated violations by the Polish government of the principle of loyal cooperation in its dealings with the Commission, the Council and now the ECJ.

The Court does not explicitly tackle the question of the NCJ, which was established in 2018 in open violation of what the Commission recommended. The ruling however makes it apparent that the consultation of the NCJ cannot be viewed as an effective safeguard to protect judicial independence. The Court will get a chance to make this crystal clear in Joined Cases C-585/18, C-624/18 and C-625/18. One may expect the Court to follow AG Tanchev and find the new NCJ as lacking the required independence from the legislative and executive authorities. The ramifications of such a finding would be extremely significant as it would essentially mean that every single decision made by the ENCJ-suspended NCJ would be have been made by a compromised body acting in breach of its mandate to safeguard the independence of courts and judges. The potential impact could be extremely significant especially as regards the Polish courts which include the “judges” nominated by the tainted NCJ and appointed by the Polish President.

With respect to the new disciplinary system, the Court could not have more clearly indicated that it shares the Commission’s concerns when it stated that its case-law requires that the rules governing the disciplinary regime “must provide the necessary guarantees in order to prevent any risk of that disciplinary regime being used as a system of political control of the content of judicial decisions” (§ 77). This is virtually the same phrasing used by the Commission when it announced the launch of its latest infringement action regarding Poland’s “reforms” on 3 April 2019. We understand this as an implicit encouragement for the Commission to promptly continue with its action. This means inter alia that it is only a matter of time before the so-called “Disciplinary Chamber” is found to violate the requirements of judicial independence required by EU Law. Again, the ramifications of such a finding would be extremely significant as all of the Disciplinary Chamber’s decisions to date would then have been made by a body masquerading as a court.

Notwithstanding the above, the direct practical consequences of the Court’s ruling will remain modest. Indeed, a significant number of Supreme Court judges had previously refused to subject themselves to the plainly unlawful retirement regime both as a matter of Polish constitutional law and EU law, with all relevant Supreme Court judges requested to return to work following the first interim order by the Chief Justice (the President of the Supreme Administrative Court (SAC) did the same in relation to the SAC judges who were forcibly retired as well). Furthermore, Polish authorities essentially conceded defeat after the final interim order adopted by the ECJ last December (while doing so it however sought to discreetly neutralise some pending preliminary ruling requests in order to prevent the ECJ from examining the NCJ and the Supreme Court’s Disciplinary Chamber in light of the EU requirements of judicial independence).

Most importantly, this ruling does not directly engage with one of the decisive issues raised by the Commission: the decision of the Polish President to increase the number of posts within the Supreme Court, which will eventually enable the ruling party to capture it. However, considering the arguably unlawful nature of the procedure having been used by the Polish President to appoint individuals to the Supreme Court, this issue should eventually reach the ECJ as it was the subject of the most recent preliminary ruling request adopted by a not yet captured chamber of Poland’s Supreme Court on 12 June 2019.

While not addressed by the Court’s ruling, it is to be hoped that within the framework of the infringement action regarding the Polish law on ordinary courts, the ECJ will tackle the forced retirement/dismissal of 61 ordinary court judges. In the absence of any pending actions raising this issue, the ECJ is unlikely to be able to address the dismissal of over 70 court presidents (and 70 vice‑presidents) which took place in 2017-18 on the back of a six-month transitional regime “which gave the Minister of Justice the power to arbitrarily dismiss them without any specific criteria, without justification and without judicial review”. Similarly, to the best of our knowledge, there is no legal action which would enable the ECJ to look into the “the very high number of dismissals and demotions among the Polish prosecutors”.

Lastly, one fundamental issue which is yet to reach the ECJ is the lack of effective constitutional review in Poland ever since the Constitutional Tribunal was unlawfully captured in December 2016, and whose independence and credibility, as recently noted by the Council of Europe Commissioner for Human Rights, “have been seriously compromised by the persisting controversy surrounding the election and the status of its new President and several of its new judges”. This is a key issue which is however bound to arise sooner or later especially if, on the back of successful electoral results, the current ruling party is unable to resit the temptation to further instrumentalise the captured “Constitutional Tribunal”, for instance to justify non-compliance refuse to comply with ECJ rulings on specious constitutional grounds. 

4. Key lesson

The key lesson we draw from this ruling is that any “dialogue” with authorities engaged in rule of law backsliding should be systematically accompanied with the launch of as many infringement actions as possible and as soon as possible.

In the present case, one may not forget how seemingly difficult some within the Commission found it to accept the need for prompt legal action in the first place. To justify legal inaction, we often heard the argument that Article 7 TEU should be considered a lex specialis and therefore exclude the launch of Article 258 infringement actions on issues already highlighted as problematic under any ongoing Article 7 procedure. As observed by AG Tanchev, Article 7 TEU And Article 258 TFEU must however be considered as separate yet complementary procedures which can be invoked in parallel.

To maximise the effectiveness of infringement actions and “prevent the completion of constitutional capture before any eventual ECJ ruling”, accelerated infringement actions ought to be the default position when a Member State openly violates the rule of law. The Commission ought to also systematically request the ECJ to decide these actions under an expedited procedure while also simultaneously request relevant interim measures so as to prevent authorities from changing the facts on the ground before the ECJ is able to issue final rulings.

What the Commission has done and achieved in the present case should be commended. It should also be “considered the new template to follow” whenever judicial independence of national courts is under threat due to autocratically-minded authorities.

Barnard & Peers: chapter 9, chapter 10
Photo credit: France 24

Wednesday, 26 June 2019

More majority voting on EU social policy? Assessing the Commission proposal

Ane Aranguiz, PhD Candidate, University of Antwerp

On 16 April 2019 the Commission launched the discussion on how to render decision-making process at EU level more efficient in the social field by activating the passerelle clauses and moving from unanimity to qualified majority voting (QMV) and from special to ordinary legislative procedure without undergoing an unwieldly process of Treaty reforms – although a unanimous vote of Member States is still necessary to approve this change.

The passerelle clauses are part of a number of ‘flexibility mechanisms’ introduced by the Lisbon Treaty that allow to simplify the decision-making process thereby enabling a more efficient exercise of EU competences where special legislative procedure and unanimity are maintained. The Lisbon Treaty provides for a general passerelle clause enshrined in Article 48(7) TEU that is applicable to all policy areas -with the exception of military or defence-related decisions-, as well as specific passerelle clauses that apply only in certain policy areas, namely, Article 32(3) TEU on Common Foreign and Security Policy, Article 82(3) TFEU on judicial cooperation in civil matters, Article 153(2) TFEU on social policy,  Article 192(2) TFEU on environmental policy and Article 312(2) TFEU on the Multiannual Financial Framework.


This Communication is the last of a series of four aiming at reviewing the passerelle clauses provided for the EU Treaties as envisioned by President Juncker in his 2018 State of the Union address. In September 2018, the Commission presented the first proposal on common foreign and security policy, followed by a communication in January 2019 on taxation. In April 2019, the Commission presented the last two proposals first on energy and climate and later on social policy. (None of these proposals has been followed up by the Member States yet).

In social policy, most areas where the EU has competence to act are already subject to QMV and ordinary legislative procedure, which has allowed for an expansion of the social acquis at the EU level over the years. Yet, a reduced but significant number of areas of social policy still require unanimity among EU Member States and a special legislative procedure. These areas include measures relating to the protection against dismissal, social representation and defence of workers’ and employer’s interests, conditions of employment for third-country legal residents, non-discrimination (based on gender, racial or ethnic origin, religion or belief, disability, age, and sexual orientation) and social security and social protection for workers outside cross-border situations.

The specific passerelle clause under Article 153(2) would allow for the transition of the first three areas, whereas the general passerelle could further be applied to the latter two. Differences remain between the general and specific passerelle clauses regarding the procedural requirements for their activation. In order to activate the general clause, the European Council has to take the initiative and indicate the precise envisaged change in the decision-making procedure and notify national parliaments, which have up to six months to object to the proposal. After that, the European Council may, by unanimity and once consent by the European Parliament has been obtained, adopt the decision authorising the Council to act by QMV or enabling the adoption of the corresponding measures by ordinary legislative procedure. This procedure allows also for the half-way activation of the clause where they move from unanimity to QMV while maintaining the special legislative procedure. The activation of the specific passerelle clause, differently, is ‘only’ subject to unanimous agreement in the Council on the basis of a proposal by the Commission and after consultation with the Parliament.

According to the Commission, other than the fact that these policy areas might have major implications on the financial equilibrium of the national welfare systems, a limitation specifically provided for in Article 153(4)TFEU, there is seemingly no logical reasons that explain why these fields remain subject to unanimity and special legislative procedure. Consequently, in December 2018 the Commission presented its roadmap for the proposal for more efficient law-making in social policy and opened the feedback period that collected 27 opinions from different stakeholders.

The Communication

The Communication opens the debate on the enhanced use of QMV and ordinary legislative procedure with the aim of rendering the decision-making process more timely, flexible and efficient.

The Communication emphasizes that while the activation of the passerelle clause would change the decision-making method, it would not alter the overall EU legal framework and earmarked that EU measures are still subject to the principles of subsidiarity and proportionality, the limitations under the social policy title Article 153 TFEU regarding, inter alia, defining fundamental principles of social security or the specifically excluded areas of the right to association, the right to strike and the right to impose lockouts.

Further in the Communication, the Commission discusses the possibility of activating the passerelle clause in the five areas where unanimity and special legislative procedure is still required. Yet, the Commission concludes that only in two out of the five areas the activation of the would passerelle clause have an added value. Firstly, the Commission argues in favour of the use of the passerelle clause in the field of non-discrimination to facilitate equal protection against discrimination that guarantees an effective redress mechanism for all. Particularly, the Commission states that while there is certain level of protection for gender and racial discrimination in employment, equal treatment on the grounds of belief, disability, age and sexual orientation remains protected only in employment and occupation. The Commission considers necessary to address the inconsistent and incoherent EU legal framework where some individuals are better protected than others. (Note that a Commission proposal in this field from 2008 has not yet been agreed).

The Commission also sees suitable to activate the general passerelle clause with regard to social security and social protection of workers for the adoption of recommendations in the near future. The Commission here recalls the recently politically agreed recommendation on access to social protection for workers and the self-employed which is still pending for final adoption, and considers that a more effective decision-making process is desired to support the process of modernisation and convergence of national social protection systems.

Nevertheless, as for the other three fields where unanimity and special legislative procedure is required, namely, protection against dismissals, employment conditions of third-country nationals and the representation and collective defence of the interests of workers and employers, the Commission does not see fit to activate the passerelle clause due to either the limitations envisioned in the Treaty, the sufficiency of the existing legislation or the strong links and diversity between national social protection systems.


The proposal put forward by the Commission should be given a cautious welcome. From a positive standpoint, the fact that most of social policy fields where the EU has competence require QMV and unanimity is required only in few domains leads to an uneven a fragmented development of the social acquis. Moving from unanimity to QMV in those limited areas allows for a swiftly and effective policy response in all areas of EU law and prevents one single Member State from vetoing social initiatives while still requiring a high degree of consensus. Secondly, transitioning to an ordinary legislative procedure allows for highlighting the role of the European Parliaments as a co-decision making. While the special legislative procedure relegates the role of the European Parliament to the subordinated position of a mere consultant, in ordinary legislative procedures the European Parliament becomes an equal to the Council and allows for a more democratic decision-making process where the direct beneficiaries are being represented. The activation of passerelle clauses in the social field would therefore not only avoid blockage by a single Member State, but also give the European Parliament a real co-legislator role. Considering the obstacles faced in the adoption of social policy legal instruments due to the lack of consensus in the Council, an active involvement of the usually more socially progressive European Parliament, is likely to free the decision-making process in social matters to a certain extent.

Yet, there are a number of points of concern. To begin with, the activation of the passerelle clause is only envisioned for two out of the five social policy areas that still require unanimity and the special legislative procedure. Moreover, these are the exact same two that cannot rely on the special passerelle clause under Article 153(2) but must be based on the general provision under Article 48(7) which, in turn, requires a much stricter procedural formula. On top of this, one of the two fields, namely social security and social protection of workers, is only contemplated with regard to the adoption of recommendations, thus disregarding the possibility to adopt binding instruments. This is particularly striking when considering the challenges faced recently by the Commission in the formulation of a measure for access to social protection of workers and self-employed, where the Commission inclined for a proposal for a recommendation due to the lack of political support to adopt a binding instrument by Member States.

The activation of the passerelle clause is clearly a positive development, yet, the fact that this is such a limited activation is highly regrettable. Continued fragmentation on social policy may moreover lead to the use of enhanced cooperation, where Member States might separately agree on social policy instruments for higher protection of their citizens. Yet, this will unquestionably result in a two-speed Europe between those Members within and out the enhanced cooperation framework.

The dynamism of the Commission in the context of the European Pillar of Social Rights provides the perfect platform to keep adapting, updating and adopting new social legislation at the EU level thus aligning EU law with the social priorities identified by Juncker’s Commission. If, and this is a big if, the discussion opened by the Commission leads to activating the passerelle clause (even if only limitedly), it will in all likelihood lead to new proposals by the Commission tackling non-discrimination in a more comprehensive manner that could be adopted in a more efficient manner. However, this will fundamentally depend on whether or not the next Commission resumes the enthusiastic social activism of the Juncker delegation.

Yet, if the Pillar is indeed the last chance for social Europe that many have claimed, this initiative represents a missed opportunity to render effectiveness in the decision-making process in social policy by closing the door to facilitating measures tackling clear gaps on the current EU legislation, most clearly with regard to protection against dismissals. It is equally regrettable the choice to limit the use of the passerelle clause to adopt a binding unified response to the inadequacies of our current social protection systems. In times of increased Euroscepticism and rising non-standard forms of employment, providing a response to concrete needs of citizens remains an imperative for future-proving the EU, therefore, it is in the best interest of the same to remove any obstructions of the use of Union competences that allow to move closer to an actual social market economy. At the very least, this initiative embodies the intention to partially unclog the ‘way’ when there is certain degree of ‘will’.

Barnard & Peers: chapter 20
Photo credit: The Independent