Monday 4 April 2022

Can a Member State be expelled or suspended from the EU? Updated overview of Article 7 TEU


Professor Steve Peers, University of Essex

With the re-election of the Orban government in Hungary, some of its critics are calling for Hungary to be expelled from the EU. But is that even possible? And if not, what other sanctions can be imposed against a Member State by the EU?

Back in 2019, I wrote two blog posts on this theme: first, on the ‘Article 7’ process for sanctioning Member States for breach of the rule of law or other EU values; second, on the alternative processes (ie other than Article 7) for sanctioning Member States. This is an update of the first of those blog posts; I hope to update the second one at some point too.

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. 

In practice, this process has been triggered both against Poland (by the European Commission) and Hungary (by the European Parliament). So far neither has resulted in any action by the Council, despite holding a number of the hearings referred to in Article 7(1).. So obviously triggering the process does not, as some think, mean that the Council will agree to issue a ‘yellow card’, or has done so already. 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.

Note that 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’. (see 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 was 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.

It’s sometimes suggested that the large bulk of Member States could just leave the EU, forming an “EU 2.0” copy of it among themselves, with only Poland and Hungary left in the original EU. Such a move would be risky for those who support EU membership in the departing Member States, as the critics of the EU would be given an opportunity to prevent their countries signing up to the new EU, or to demand renegotiation of the current terms of membership.

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 second 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 (ie, an issue definitely within the scope of EU law). 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 – for example, in the string of rule of law judgments regarding the Polish courts.

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 (note that the Council's internal rules on Article 7 hearings have been published). 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, the CJEU did address some issues about its scope in 2021, when it ruled on a legal challenge by Hungary to the European Parliament’s decision merely to trigger Article 7. In the Court’s view, the special rules in Article 269 did not exclude a legal challenge to the decision to trigger the process, because those special rules only apply to acts of the Council or European Council. The European Parliament’s resolution had legal effect, because it also triggered the protocol on asylum for EU citizens; therefore it could be challenged (non-binding EU acts, ie with no legal effect, cannot be challenged). Nor did the Court agree with the European Parliament that its resolution was a purely intermediate step.

However, the Court ruled that some aspects of Article 269 did apply to legal actions against the triggering of an Article 7 procedure: only the Member State concerned could bring a challenge, and it could only raise the procedural issues referred to in Article 269, not the substance of whether the body which started the process had wrongly claimed that the Member State was breaching (or risked breaching) EU values to the thresholds set out in Article 7. However, that Member State was entitled to the usual two months to bring the legal action, not the one month referred to in Article 269. Ultimately the Court rejected the Hungarian government’s argument on the merits, finding that European Parliament correctly counted the votes cast (and abstentions) 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 was conceived as a political process par excellence, and it remains supremely political. 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. While recent Hungarian elections have been criticised as undemocratic, Member States seem reluctant to pursue the route of defining what they consider to be acceptable democratic standards and sanctioning other Member States for breaching them.

So is the Article 7 process doomed? In fact, the expansion of EU law in areas with significant relevance to human rights, the willingness of the CJEU to rule on the judicial independence of national courts in general, and the creation of new means to address ‘rule of law’ concerns within the EU budget system 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 EU law system, discussed in the second blog post.


Barnard & Peers: chapter 9

Photo credit: Steve Peers


  1. I’m very glad to see someone giving this some real thought. There is a problem however – it’s decades late! Years ago I got laughed out of a university seminar room when I talked about stuff like this. I said that neofunctionalist analysis was little more than crossed fingers and that nothing at all could be regarded as fixed on the basis that the EU institutions would find change inconvenient. I was told that globalisation was totally irreversible and could only spread democracy – well looking at Russia that went bang fast.

    Indeed Professor Peers, my view is you need to go further and look at the single currency. Having loudly proclaimed to anyone that will listen that the flagship-project-single-currency is irreversible how can the EU then talk about expelling members? Unless the EU wants to say that the single currency is reversible, but I suspect that is not the message they want to give. Similarly is the EU really saying that EU citizenship can reasonably be regarded as withdrawable?

    None of these pressures, questions and developments was unforeseeable. Not one. What happened at the time was that politicians (and academics) simply did not take this stuff seriously and did not think it through. There was never anything at all theoretical about a member leaving. If you do a literature search of around the 1990s you can see how thin the literature was on these questions. The botch job that is the text of A50 speaks to the lack of serious thought ever given to that eventuality. What we are left with is a constitutional deficit and we should not be afraid to say as much for fear of association with a vinegary nationalism. It’s not much use building a fine legal order alongside a gaping constitutional deficit – I don’t even think that statement is controversial.

    I do appreciate that in the UK specifically there has long-been a problem that politicians have wanted to present the EU as a trade bloc when it manifestly is a permanent political construct. That is a UK problem which almost certainly built up the pressures that led to 2016. However this political cowardice does not detract from the failures of the EU establishment and the EU studies academics to give any thought to far-from-theoretical outcomes.

    Politicians and academic should ask themselves some searching questions about why decades ago these very serious and very real questions were just seen as an exercise in whimsy.

    1. As I explain, there is not an expulsion clause, so the example you provide re the single currency is irrelevant. And since suspended countries remain Member States, I don't see how EU citizenship is lost. The loss of EU citizenship on withdrawal seems straightforward to me, frankly; those who don't want to lose EU citizenship have managed to convince people that the issue is more complicated and uncertain than it actually is. But we'll soon find out what the Court thinks of this. Your comments seem motivated by animosity to the EU, which skews your analysis.

    2. Also it's arguable that the EU was not always presented as a trade bloc - see @steveanalyst on Twitter for example. Brexity tropes are not a convincing argument.

    3. Thank you for your reply – appreciated. Animosity? Not really. I voted REMAIN, albeit with some reluctance. I think that the EU has made some almighty and very avoidable errors. I think that there has been too much hopey-changey idealism on the part of many of the EU’s advocates. I think that politicians and academics have been guilty of at least some level of collective group-think on the EU. I don’t even think those views are contentious.

      But animosity – no. My point was, and remains, that all of these pressures you are talking about here were far from unforeseeable and that they could and should have been taken far more seriously at the time.

      Indeed as far as I can tell Bulgaria and Romania, 15 years on are still in CVM. Not exactly a resounding triumph.

      Where I disagree with you I think is on this hair-splitting exercise you talk about in your response to me. There is either a wish to have an expulsion mechanism or there isn’t. Expulsion means that the single currency can be reversed, that citizenship can be revoked and the like as a matter of the political order. Expulsion means expulsion. If you are not prepared to say as much then you have no meaningful view of expulsion. Make whatever value judgment you like about the presence or absence of an expulsion mechanism, the implications of one can’t just be glossed over.

      It is right there in your article. ‘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?’

      These are the questions that some of us were posing decades ago and being told we were idiots for doing so. Indeed I could not believe the reluctance of some to even consider the possibility of there being an Orban-type character or the break up of a member state. These are the what-ifs we needed to resolve and put explicitly to the voters before, not after, signing up to a quasi permanent political construct and stepping into the legal and political unknown. That remains my point.

      What the EU actually did was little more than cross its fingers and hope that the equilibrium would hold. I thought it was a bad idea at the time and I stand by that.

    4. This is all Monday-morning quarterbacking. There are many international treaties or institutions which are vague on these issues. There was possibly a deliberate political choice: the CJEU has noted that there was a discussion of an expulsion clause, but it was rejected. And I don’t see the hair splitting: there isn't an expulsion clause. Whether there should be is theoretical. It's not unreasonable to focus on actual legal issues. Not everything is about you and whether some lecturer was rude ten years ago.

  2. Commenter talks about issues, professor picks and chooses irrelevant parts to attack commenter. What a classy professor

    1. This is clearly untrue and unpleasant. The original comment was already personalised so the reply was going to be personal too. But I also made a number of substantive points in my replies. This blog is not a forum for personal attacks on me, so don't bother making any more of them. - Steve Peers