Carrying a bigger stick: the EU strengthens its power to suspend visa waivers quickly, in pursuit of immigration policy and external relations objectives

 

Professor Steve Peers, Royal Holloway University of London

Photo credit: Jelger Groeneveld, via Wikimedia Commons

Introduction

What tools does the EU have to get non-EU countries to agree to do something the EU wants them to do – or to stop doing something the EU objects to them doing? One mechanism is foreign policy sanctions, which often limit trade or investment in some way, or freeze assets, but also sometimes impose an entry ban on individuals who are linked to a government, group or policy that the EU objects to. But under the law of the EU, foreign policy sanctions require unanimity, and sometimes that can be hard to reach. As an alternative approach, trade measures as such can be adopted by a qualified majority vote; hence the recent EU law increasing tariffs on some Russian and Belarusian products, which sits alongside the more traditional sanctions on those countries.

Another area for exerting such pressure is visas, where a common visa policy gives the EU leverage to (threaten to) require some or all of a non-EU country’s citizens to obtain visas to visit the EU, or, in the event of a visa requirement being imposed, to make those visas more difficult or expensive to get. Conversely, the EU sometimes offers carrots rather than sticks: it can be willing to drop visa requirements for some or all of a non-EU country’s citizens to visit the EU, or where a visa requirement is maintained, to make visas easier or cheaper to get, in return for some commitments from the non-EU country concerned.

Recently the EU has agreed to amend its visa law to add a new string to its bow in this field – making it easier to reimpose visa requirements on non-EU countries (or categories of their citizens) on a fast-track basis, and to provide for more grounds on which to do so. (The full text of the agreed new law is here) This ‘suspension mechanism’ now includes grounds for reimposing visa requirements going beyond the scope of immigration policy as such. The following blog post analyses the newly agreed law within the broader context.

Background

Current EU visa law generally

All EU countries except Ireland, plus the Schengen associated countries of Norway, Switzerland, Iceland and Liechtenstein, have agreed a common list of countries whose nationals are required (or not required) to have visas for a short-term visit (90 days out of every 180 days).  This is set out in the frequently-amended visa list Regulation, last codified in 2018. The recently-agreed law discussed in this blog post is an amendment to the visa list Regulation.

At present, the countries whose nationals are not subject to EU short-term visa requirements (the ‘whitelist’) comprise most non-EU European states (including the UK), North American and most Latin American countries, some wealthier Asian countries (such as Israel, the UAE, Japan and South Korea), Australia, New Zealand and a number of small middle-income islands in the Caribbean and the Indian and Pacific Oceans. All other countries (the ‘blacklist’) are subject to visa requirements.

While the list of countries on the visa whitelist or blacklist is harmonised, there is some discretion for Member States as regards categories of people from the non-EU countries on these lists. For example, individual Member States can drop visa requirements for holders of official or special passports even though their country is on the visa blacklist; or conversely impose visa requirements for holders of official or special passports even though their country is on the visa whitelist.

The detailed rules on issuing ‘Schengen visas’, ie a visa allowing travel across all the countries participating in Schengen (currently all EU countries minus Cyprus and Ireland, plus the Schengen associated countries), are set out in the EU visa code, adopted in 2009 and amended in 2019.  

The EU also has two types of visa treaty with non-EU countries. First, visa waiver treaties, where the EU and the non-EU country concerned agree reciprocally to waive visa requirements for short-term visits in each direction. Secondly, visa facilitation treaties, where the EU and the non-EU country concerned agree reciprocally to make visas easier to obtain – ie simplifying the documents required, reducing or waiving the fee for an application, and even dropping visa requirements entirely for holders of official or special passports. In practice, the EU usually links visa facilitation treaties to readmission treaties.

Current visa sanctions: broader context   

Since its amendment in 2019, the visa code has provided for the EU to (threaten to) sanction non-EU countries on the visa blacklist, by making visas harder and more expensive to get, if a country (in the EU’s view) does not cooperate sufficiently on readmission of its own citizens. This power has been used (or threatened) several times (see page 5 of the latest Commission annual report).

A further proposal to restrict EU trade preferences for developing countries on the same grounds has not yet been agreed (in part because it has been criticised for potentially breaching WTO law). Negotiations on this proposal broke off last year because the European Parliament objected in principle to this idea. However, they have recently resumed.

The EU has (respectively) wholly or partly suspended three visa facilitation treaties: wholly with Russia (due to the invasion of Ukraine); partly with Belarus (due to Belarus suspending the parallel readmission treaty, plus facilitating migration into neighbouring EU countries); and most recently partly with Georgia, due to concerns about human rights there. In the latter two cases, the EU targeted the removal of visa facilitation upon the holders of official passports; the general public still benefits from visa facilitation (as regards Belarus) and the subsequent visa waiver (as regards Georgia). But because the visa list Regulation (as it stands) still gives Member States discretion whether to drop visa requirements for Georgians with official passports or not, not all Member States have followed up the EU’s decision to partly suspend the treaty with Georgia: at time of writing, only eight Member States have notified the reimposition of visas for holders of official Georgian passports.   

Also the EU has partially then fully suspended the visa waiver treaty with the Pacific island of Vanuatu, due to its investor citizenship policy. This is linked to visa sanctions under the visa list Regulation, discussed further below. 

Current visa sanctions: visa list Regulation

The visa list Regulation provides for two routes to reimpose visa requirements on a fast-track basis. First of all, it’s possible to do so on reciprocity grounds, ie as a reaction to a country on the whitelist not waiving visa requirements in turn for all EU Member States. This issue is contentious in practice: the rules have been amended several times; the Commission frequently reports on them; and the CJEU has ruled on them twice (ruling against the Commission when it challenged the procedure involved, seeking less accountability; but later ruling for the Commission when the European Parliament sued it for ‘failure to act’ when it did not impose sanctions upon the USA in response to a lack of full reciprocity; in the Court’s view, the Commission was not under a legal obligation to sanction the USA). (During negotiations on the amendments discussed in this blog post, the European Parliament also tried to amend the reciprocity rules to overturn the outcome of this judgment going forward; it was unsuccessful).

But our main focus here is the second route to the fast-track reimposition of visa requirements (the suspension mechanism): dropping the visa waiver on immigration and asylum policy grounds, now also to include more general external relations grounds. For this is the only part of the visa list Regulation that will be amended by the newly agreed law. (Note that the amendments have not yet been formally adopted; this will likely take until the autumn).

The visa list Regulation was first amended to provide for a suspension mechanism in 2013, due to concern from Member States about the impact of dropping visa requirements for most of the Western Balkans several years beforehand. Those rules were amended in turn in 2017, this time due to concern from Member States about dropping visa requirements for eastern neighbourhood countries (eventually visas were waived for Moldova, Ukraine and Georgia).

Substantive grounds for using the suspension mechanism

Under the current law before amendment, the suspension mechanism can apply on any of four grounds:

-          there is ‘a substantial increase in the number of nationals of that third country refused entry or found to be staying in the Member State's territory without a right to do so’; or -          ‘a substantial increase in the number of asylum applications from the nationals of that third country for which the recognition rate is low’; or -          ‘a decrease in cooperation on readmission with that third country, substantiated by adequate data, in particular a substantial increase in the refusal rate of readmission applications submitted by the Member State to that third country for its own nationals’ (or nationals of other states who transited through that country, if a readmission treaty with the EU or the complaining Member State obliges the country concerned to readmit them); or -          ‘an increased risk or imminent threat to the public policy or internal security of Member States, in particular a substantial increase in serious criminal offences linked to the nationals of that third country, substantiated by objective, concrete and relevant information and data provided by the competent authorities’. All four grounds refer to a ‘substantial increase’. But what is a ‘substantial increase’? The concept is further defined in the preamble to the Regulation: it means an increase of more than 50% over a two-month period compared to the same period in the previous year (or compared with the last two months prior to dropping the visa requirement for a non-EU country). However, the Commission can use a threshold lower than 50% if it ‘deemed it applicable in the particular case’. As for a ‘low recognition rate’ for asylum applications, this is ‘around 3 or 4%’ according to the preamble, although again the Commission can deem a higher rate applicable in a particular case. For readmission, the ‘decrease in cooperation’ does not necessarily have to take the form of a substantial increase in refusal of readmission requests (as confirmed by ‘in particular’). The main text of the Regulation states that ‘non-cooperation’ on readmission (note the different wording from ‘decrease’) can also, as regards a readmission treaty with the EU, apply ‘for instance’ to ‘refusing or failing to process readmission applications in due time’; or ‘failing to issue travel documents’ or ‘refusing to accept European travel documents’ in accordance with the deadlines in the treaty; or ‘terminating or suspending’ the readmission treaty. Conversely, there is currently no further definition of ‘serious criminal offences’, or any indications of what might otherwise comprise ‘an increased risk or imminent threat to the public policy or internal security of Member States’, given that the ‘substantial increase in serious criminal offences’ ground is only an example of such a risk or threat (‘in particular’).  Procedural rules of the suspension mechanism The mechanism can be triggered either by the Commission or a Member State. In fact, as part of the rules on the suspension mechanism, the Commission has a specific obligation to monitor compliance with commitments given by a non-EU country in the context of a ‘visa liberalisation dialogue’, referring to a process over several years in which non-EU countries make commitments on immigration law and human rights issues, in return for which the EU eventually waives the visa requirement (in practice, this applies to Western Balkans and eastern neighbourhood countries; a dialogue is also currently underway with Armenia). The Commission must report on these countries annually for seven years after the visa waiver, and since 2023 has been reporting on some other whitelist countries too (see the most recent report). In effect, this is a fifth ground of suspension, because the visa list Regulation also refers to triggering the suspension mechanism ‘where a report of the Commission shows that one or more of the specific requirements is no longer complied with as regards a particular third country’. After taking account of evidence required by the Regulation, the Commission can ‘decide’ to sanction the non-EU country; seemingly it is obliged to do so if a simple majority of Member States (so 14 of them) have triggered the process as regards the four grounds for using the mechanism. This would take the form of an implementing act (ie subject to scrutiny by Member States’ representatives) suspending the visa waiver for some groups of that country’s nationals for nine months. If the issues persist, the Commission ‘shall’ adopt a delegated act (ie subject to control from the Council and the European Parliament) suspending the visa waiver for all nationals of that country for 18 months. That can be extended for six months if the Commission makes a legislative proposal (which would need to be agreed by the Council with a qualified majority of Member States, and the European Parliament) to make the move to the blacklist permanent.  The suspension mechanism in practice The EU has only used the rules once so far. As noted above, it punished Vanuatu for its investor citizenship rules, taking the view that the ‘public policy and internal security’ ground of the suspension mechanism justified reimposing visa requirements in these circumstances. (Remember that a substantial increase in criminal offences was only one example of a public policy or internal security threat: ‘in particular’) Applying the procedural rules, this entailed an implementing decision, then a delegated act, later extended while the legislation was amended, as well as the suspension of the visa waiver treaty discussed above. (Note that this process was different from that leading to the CJEU recently finding that a Member State (Malta) had breached EU law by, in effect, selling its nationality as a route to obtaining EU citizenship.) The newly agreed amendments The recently agreed amendments follow the Commission’s proposal in 2023 (and see the Commission's earlier discussion paper), the Council’s (ie Member States’) agreed position in March 2024, and the European Parliament’s position earlier this year. I analysed these competing versions several months ago. The different versions can also be compared in a four-column document, and there is a very useful later four-column document that gives a snapshot of where the negotiations stood in May. With the text of the amendments now agreed, the following sets out what those amendments will change, indicating if it was the Commission, the Council or the European Parliament that had the biggest role in securing the key amendments. I’ll also mention some proposed amendments that weren’t adopted. At first sight, the first two grounds for using the suspension mechanism – a substantial increase in refusals of entry or overstay rates, and a substantial increase in asylum applications from a country with a low recognition rate – remain the same. But in fact, changes to other parts of the Regulation will alter their interpretation: the main text (no longer the preamble) to the Regulation now defines a ‘substantial increase’ as 30%, instead of 50%, and a ‘low recognition rate’ is now defined as 20%, instead of 3-4%, matching the new asylum procedures Regulation. This means that it will be easier to trigger the suspension mechanism on these grounds. Indeed several countries assessed in the Commission's annual report will now potentially fall within the scope of the suspension mechanism (and see page 5 of this Council document). The Commission proposal would not have changed these numbers; the Council insisted on the amendments, with the European Parliament dropping its compromise proposal of 40% to define ‘substantial increase’. However, there is a requirement for the Commission to assess the interpretation of the 30% rate within three years. Unlike the current Regulation, the Commission could decide to change these rates either way, although (at the behest of both the Parliament and Council) it must ‘duly justify’ any decision to do so.    The third ground – a decrease in readmission cooperation – remains, but is reworded more simply, referring now to ‘a decrease on cooperation on readmission…or other cases of non-cooperation on readmission’. However, the text still separately retains a reference to a ‘substantial increase’ in refusal of readmissions, now defining this, as elsewhere, as a 30% increase. The European Parliament’s proposal to require that it must be shown that this reduction in cooperation is the fault of the country concerned was rebuffed. The current definition of other forms of non-cooperation on readmission is also retained, now supplemented by clarifying that it also applies to ‘failing to assist in identifying third-country nationals for whom a Member State has submitted readmission applications or otherwise creating persisting practical obstacles regarding the enforcement of returns’. The Parliament’s attempt to limit the potential use of the suspension mechanism against countries which did not issue travel documents (so the rules would apply only if the non-EU country failed to issue the documents for its own nationals, or for non-nationals who held a right of residence in that country) was rebuffed. The fourth ground – public policy or internal security – now refers to a ‘significant risk’ (rather than an ‘increased risk’, and adds to the list of examples of such risks or threats (this list is still non-exhaustive: ‘in particular’ is retained, at the Council’s behest). The first example remains a ‘substantial increase in serious criminal offences’, which is now defined, as elsewhere, as a 30% increase. On the other hand, ‘serious criminal offences’ is not defined; although the Council wanted the preamble to refer to ‘terrorist offences and activities of organised criminal groups’ in this context, this was not agreed. A new second example is ‘hybrid threats’, referred to in the preamble as including ‘state-sponsored instrumentalisation of migrants’ as defined in the EU’s asylum crisis Regulation; and a new third example is ‘systemic deficiencies in document security legislation or procedures’ (at the behest of the Council, although ‘systemic’ was added to limit its scope; there is no definition of ‘systemic’). A new fifth ground is investor citizenship schemes, defined as schemes ‘whereby citizenship is granted without any genuine link to the third country concerned, in exchange for pre-determined payments or investments’. As we have seen, though, this ground is not genuinely new, as the current ‘public policy or internal security’ ground has been interpreted to cover investor citizenship. The preamble asserts that ‘visa-free third countries should be deterred from using visa-free access to the Union as a tool for leveraging individual investment in return for their citizenship’, and adds a further rationale, at the behest of the European Parliament: ‘a lack of comprehensive security checks, vetting procedures and due diligence by such third countries with regard to investor citizenship schemes pose several serious security risks for Union citizens, such as those stemming from money laundering and corruption’.      Next, a new sixth ground is non-alignment with EU visa policy, where, in particular due to geographic proximity to the EU, this ‘could lead to a substantial increase in the number of third-country nationals, other than nationals of that third country, who enter irregularly the territory of the Member States after having stayed on, or transited through, the territory of that third country’ (italics added). The ‘could lead to’ wording is a compromise: the Commission had proposed ‘there is a risk of’; the Council agreed; but the European Parliament wanted to delete the words, meaning that this ground could only have applied if a ‘substantial increase’ had actually happened. Frankly, the European Parliament has conceded the point here: it’s hard to see any real difference between ‘risk of’ and ‘could lead to’. This brings us to the new seventh ground for using the suspension mechanism: non-compliance with commitments made to the EU during a ‘visa liberalisation dialogue’ which resulted in a country being moved to the whitelist. As we saw above, though, this is already a de facto ground for suspension under the current law. The new eighth ground for using the suspension mechanism is perhaps the most eye-catching, as it refers to external relations issues generally, not necessarily linked to immigration or asylum policy at all: a deterioration in the Union’s external relations with [whitelist country] caused by: (i) serious breaches by that third country of the principles set out in the Charter of the United Nations; (ii) grave violations by that third country of fundamental freedoms or of the obligations deriving from international human rights law or international humanitarian law; (iii) serious breaches of international law and international legal standards; (iv) that third country carrying out hostile acts against the Union or Member States with the aim of destabilising or undermining society or institutions which are key for the public policy and internal security of the Union or the Member States; or (v) non-compliance with international court decisions and rulings. The Commission did not propose this new ground; the European Parliament and the Council each proposed their own version of it, and the final text is based on the Parliament’s version, with some of the Council’s preferred text added in and parts of the Parliament’s version dropped. In particular, the final text uses: the Parliament’s version as regards any deterioration (the Council’s version set a threshold of a ‘significant and abrupt’ deterioration); the Parliament’s version of the necessary link (‘caused by’; the Council’s version was ‘when it relates to’ and ‘is deriving from’); the Parliament’s version as regards the broad scope (the Council version applied ‘in particular’ to human rights); the Parliament’s version as regards ‘serious breaches’ of the UN Charter; largely the Parliament’s version as regards ‘grave violations’ of human rights law et al (although the ‘fundamental freedoms’ point comes from the Council’s text, and the Council version did not differ much: ‘serious human rights violations and abuses’); the Council’s version as regards ‘serious breaches of international law and international legal standards’; the Parliament’s version as regards ‘hostile acts’; and the Council’s version as regards ‘non-compliance with international court decisions and rulings’. The Parliament’s suggested inclusion of ‘violations…of bilateral agreements between it and the Union’ and ‘non-compliance or non-alignment by that third country with relevant Union sanctions’ were dropped – perhaps because the former could have circumvented dispute settlement processes (whereas if the EU goes to dispute settlement and wins, the ‘non-compliance with court rulings’ ground could apply) and the latter attempts to insist that non-EU countries conform to EU foreign policy (although the new Regulation already attempts to insist that they conform to EU visa policy). The notion of ‘hostile acts’ is further defined in the preamble: this ‘could result from foreign interference in political processes, economic coercion, cyber operations, economic espionage or the sabotage of critical infrastructure’.  Finally, a new ninth ground for use of the suspension mechanism is ‘any other ground for suspension set out in a short-stay visa waiver agreement between’ the EU and a whitelist country, ‘limited to the scope of application of such agreement’. Both the Council and the European Parliament pushed for this new ground, although in effect it already existed and was applied already: the EU’s reaction to Vanuatu’s investor citizenship scheme took the form of invoking both the ‘public policy or internal security’ ground in the current Regulation, and the parallel ground to suspend the treaty concerned. The threshold for assessing increases in asylum applications etc compared to the previous year will be more flexible, changing to between two and twelve months. (The Commission proposed to make it even more flexible – ‘at least two months’ – but the European Parliament insisted on at least some limit applying).   At the behest of both the Parliament and the Council, most of the new grounds for suspension could be triggered by Member States too – with the significant exception of the ‘external relations’ exception, which only the Commission can trigger, because (according to the preamble) ‘the external relations of the Union will be affected as a whole’. Interestingly, the new law does not provide for any special rules as regards the evidence to be assessed, or the process to be followed, for applying that exception. In addition to assessing the evidence relating to the ground for using the suspension mechanism, the Commission will also have to assess (at the Parliament’s behest) ‘the necessity, proportionality and consequences of a suspension of the exemption from the visa requirement.’ There’s also a new obligation, if a suspension is adopted, to ‘work together in close cooperation with the third country concerned to find alternative long-term solutions in respect of the relevant ground or grounds…and take into account the political context, the economic matters at stake and the consequences of a suspension of the exemption from the visa requirement for the overall external relations of the Union and the Member States with that third country’. When it comes to applying sanctions in practice, the first level of sanctions (an implementing measure against categories of persons) can now apply for 12 months, rather than 9. There is still an apparent obligation to act here if a simple majority of Member States trigger the rules (although this does not apply to the ‘visa dialogue’ or ‘external relations’ grounds). Moreover, where the suspension mechanism is used, Member States can no longer choose to exempt holders of special or official passports from it, at the behest of the European Parliament. This addresses the issue that arose from the recent partial suspension of the visa facilitation treaty with Georgia, as discussed above (ie partially suspending the treaty as regards visa waivers for holders of official passports did not necessarily require that Member States reintroduce visa requirements for the category of Georgians holding those passports, because the Regulation still left them with discretion on that point). There is a new quicker fast-track possibility for the Commission to use the suspension mechanism in the event of ‘urgency’. The European Parliament attempted to restrict this process to the ‘public policy and internal security’ ground, but this was rebuffed – although the preamble suggests that urgency only applies ‘for example’ to prevent mass influx or ‘serious damage to the public policy or internal security of Member States’. Conversely, the Council successfully insisted that there must still be some advance scrutiny by Member States of draft Commission decisions (albeit quicker than usual) – rather than use the ‘full-fat’ urgency procedure in the comitology Regulation as the Commission had wanted (that procedure allows the Commission to act without prior discussion with Member States at all, although only for a limited period). If the issues persist, as under the current law, the Commission ‘shall’ adopt a delegated act suspending the visa waiver for all nationals of that country. But the new law extends the time those delegated acts can apply from 18 months to 24 months. Furthermore, if the suspension mechanism is invoked on external relations grounds, the new law allows the delegated act to apply still to limited categories of people only, because in such cases (according to the preamble), ‘the Commission should favour a targeted approach’, suspending visa waivers first for those ‘holding positions of responsibility, such as members of that third country’s official delegations, members of local, regional and national governments, members of parliaments or high-ranking public or military officials, minimising adverse consequences to the general population’. As with the implementing measures, if the delegated act applies to the holders of special or official passports, Member States lose their power to exempt them from the visa requirement (ie the recent Georgia issue). In the event of a legislative proposal to make the move to the visa blacklist permanent, the delegated act can be extended for 24 months (rather than six months under the current law). And it can be extended for a further 24 months where it is targeted on holders of special or official passports, on external relations grounds. Finally, the European Parliament had also proposed a change in the law aimed at blacklist countries. It wanted to give the Commission the power to adopt an implementing measure removing Member States’ option to drop the visa requirement for holders of special or official passports from those countries too, in the event of a deterioration in external relations as defined in the new law. However, this proposal was rebuffed.

Comments

All three institutions were broadly in agreement on the main features of the Regulation. Some of them won or lost on particular points: the Parliament lost its attempts to extend its scope to cover some issues relating to reciprocity or blacklist countries, but mostly won as regards a broader scope and lower threshold for the external relations ground, and the insertion of a ‘Georgia clause’ (removing the power of Member States to waive visa requirements for official passport holders when visa requirements are reimposed). The Council won lower thresholds for the definitions of ‘substantial increase’ and low recognition rates for asylum; and the Commission got at least a limited version of fast-track powers to apply the suspension mechanism, albeit still being subject to advance scrutiny by Member States. There was little dispute as regards adding most of the new grounds for using the suspension mechanism, and the lengthened periods for reimposing visa requirements via implementing or delegated acts.

One consistent feature of the new Regulation is the broadening of scope of the suspension mechanism: not only (most obviously) by adding new grounds for its use, but also lowering the thresholds for applying the existing grounds, and by widening the reference period for determining if any of the various ‘substantial increases’ have occurred.

Traditionally Member States have been reluctant to give up powers as regards visa exemption (or requirements) for holders of special or official passports, except when the EU negotiates visa facilitation treaties. But when negotiating the new law they were willing to accept the Parliament’s proposed ‘Georgia clause’, which will constrain their powers as regards holders of such passports from whitelist countries – although they did reject the Parliament’s parallel proposal to restrain their powers as regards holders of such passports from blacklist countries.

Member States have also been remarkably willing to address foreign policy issues – traditionally a closely guarded field where unanimous voting applies – by means of visa sanctions. Usually one might expect the European Parliament to be pushing this, against strong resistance from Member States; but on this occasion, the Council had suggested a new ‘external relations’ ground first, and largely accepted the Parliament’s broader version of the clause. This is so even though the focus on targeting government officials on external relations grounds is remarkably similar to the foreign policy process of naming individuals to be sanctioned.

There’s a key difference, though, despite the overlaps: the visa suspension mechanism can only lead to officials being required to apply for a visa, whereas foreign policy sanctions can entail an entry ban. (The EU has not resorted to Trump-style entry bans for entire countries). So even if (say) Hungary complies with the ‘Georgia clause’, it could be malicious compliance: the (hypothetical) police chief of Tbilisi, fancying a weekend break after a long week of thumping protesters, applies for a Schengen visa from Hungary. Not only is it granted; the Hungarian government lays on a hotel suite full of finely aged wine – and barely dressed women – to welcome him.

The breadth and lack of clarity of the external relations ground to invoke the suspension mechanism leaves the Commission much discretionary power to determine when it might be used. While, as noted above, the EU has always avoided using the reciprocity mechanism in the visa list Regulation, it has used every other form of visa sanctions power it has, at least once. Widening the scope of the suspension mechanism makes it likely that this visa sanctions power will be used more often.

Some political considerations will of course remain: the EU seems determined so far to avoid further confronting Trump (consistent with a longer history of avoiding conflict with the USA in this field); and whatever argument might be made about Israel and international (humanitarian) law, the EU rarely goes beyond mild expressions of concern.

There may be more enthusiasm for using – or at least threatening to use (cf the visa code sanctions) – the enhanced suspension mechanism as regards immigration and asylum policy. On this point the effect of the ‘visa policy alignment’ ground of the new rules – despite the EU’s supposed opposition to smugglers and to unsafe routes for asylum-seekers – is to make it harder for asylum-seekers to get to the EU via a safe, smuggler-free journey to a neighbouring country. Moreover, to apply the ‘visa policy alignment’ ground, there’s no requirement that the asylum recognition rate has to be low. Put another way, the EU wants to end a safe route for would-be asylum-seekers to get close to the EU in order to claim asylum there, regardless of whether their application might be successful. For all the rhetoric about evil gangs of smugglers and unfounded applications, it seems the reality is that the EU just does not want refugees.

Must cases be unfounded to qualify as SLAPPs? What unfoundedness means for GDPR-based SLAPPs

 

 

Léna Perczel, Legal Officer, Political Freedoms Program, Hungarian Civil Liberties Union

 

Photo credit: Dirk Beyer, via Wikimedia commons

 

Countering SLAPPs (Strategic Lawsuits Against Public Participation) has been at the forefront of political, legal and academic discourse over the past two years. The most significant legislative development has been the European Union’s Anti-SLAPP Directive (Directive), backed by a soft law instrument, the Council of Europe’s Anti-SLAPP Recommendation (CoE Recommendation).

But what exactly qualifies as a SLAPP? The Directive, which is limited to cases with cross-border elements, defines the term and treats unfoundedness as a key criterion. In contrast, the CoE Recommendation treats it as just one of several indicators for identifying such lawsuits. While both instruments acknowledge it as a factor, Hungary’s example for General Data Protection Regulation (GDPR) based SLAPP cases suggest it is not necessarily a defining feature. 

In this blog post, I aim to explore—through the lens of this Hungarian case group—whether assessing the SLAPP nature of a case based on unfoundedness could render legal efforts to combat SLAPPs ultimately ineffective. This issue is particularly pressing in Hungary, especially in cases where the press is required to comply with GDPR obligations—yet no benchmark has been established by the European Court of Human Rights (ECtHR, the Court) to date.

 

The GDPR’s burden on the press in Hungary

 

The SLAPP phenomenon gained attention in Hungary when individuals with economic power repeatedly attempted to erase their names and wealth from the media, invoking rights enshrined in the GDPR. This conduct demonstrated that the GDPR can become a powerful tool for SLAPPs when interpreted in a strictly formal manner. By placing the responsibility on data controllers, the GDPR established a rigid procedural framework, obliging them to comply with extensive administrative safeguards. The press becomes a data controller simply by gathering and storing someone’s name, even without publishing it. As a result, a journalist must begin preparing extensive documentation from the moment they start investigating an individual. Unlike commonly used legal remedies against the press, such as press rectification procedures or defamation claims, violations of the GDPR can stand regardless of whether the article is false or reputationally harmful, thus regardless of the journalist’s ethical conduct. Adhering to such duties makes reporting on public matters increasingly difficult. In fact, beyond the administrative burden itself, informing data subjects about articles in preparation can entirely undermine investigative journalism. Data subjects may resort to dismissing evidence or objecting to the data processing, effectively blocking the publication of articles. 

Despite Article 85  of the GDPR, the Hungarian government has not reconciled the GDPR with the freedom of the press, which could have led to exemptions from certain GDPR obligations (such as the strict notification obligations imposed on data controllers, explained below). This lack of reconciliation has created a constitutional loophole: a legal grey zone that reflects the state's failure to fulfil its positive obligation to protect the press. In the absence of clear legal provisions, and due to this unresolved tension, the responsibility has fallen on those applying the law to balance the competing rights of freedom of expression and data protection. 

The Hungarian National Authority for Data Protection and Freedom of Information (DPA) was the first forum in Hungary to detail these obligations, requiring data controllers to inform each data subject preliminarily, proactively, and individually about the data being processed and its legal basis—recognizing only legitimate interest under Article 6(f) of the GDPR as a valid ground for processing. This was despite Forbes’ argument that publishing on public matters falls within the constitutional duty of the press, thus the ground for processing should be public interest (Article 6(e) of the GDPR). This means that journalists, whose work consists primarily of processing personal data, must notify each data subject in advance while conducting their reporting, including during initial research. Compliance is required regardless of whether the data subject has initiated any procedure, making this an even more effective SLAPP tool. In its decision, the DPA entirely failed to consider how such a disproportionate workload could stifle the press. Had public interest been accepted as a legal basis for processing, these notification obligations would not have been imposed on journalists.

 

The manifestation of GDPR-based SLAPPs through legal proceedings against Forbes

 

In 2019, the owners of a Hungarian energy drink company—a family business that gained prominence partly through public funding—initiated proceedings after Forbes included them in its annual wealth rankings. Their inclusion prompted GDPR-based claims.

First, they argued that the press lacked a legal basis for publishing their personal data, and that the data processing therefore constituted a violation of their rights (primary claims). Second, they contended that even if legitimate interest were accepted as the legal basis, the press had failed to meet its procedural obligations—such as informing the data subjects about the legitimate interest assessment (ancillary claims).

Both GDPR-based claims proceeded in parallel before the civil court and the DPA. Initiating multiple proceedings simultaneously by the same claimant is a typical characteristic of SLAPPs, intended to increase pressure on the target. 

In this blog post, I will focus on the DPA case. However, it is important to illustrate the SLAPP nature of these proceedings by noting that, in the civil case, the claimants requested a preliminary injunction—which the court granted (The Metropolitan Court ordered the interim measure in its decision no. 25.Pk.23.297/2019/17-I. The Appellate Court and the Supreme Court upheld the decision in their decisions 2.Pkf.25.030/2020/2. and Pfv.IV.20.395/2020/4 respectively. The decision of the Supreme Court is currently before the EctHR).

As a result, until the court ruled on the merits of the case whether Forbes had a legal basis for processing the data, the magazine was prohibited from publishing any information about the family members—amounting to de facto censorship for more than four years (The interim measure was repealed by the first-instance court’s non-final decision, decision no. 25.P.21.067/2023/21).

It was in the DPA procedure initiated by this claimant that the authority first established a formal interpretation of the GDPR, as explained above. Although the DPA’s decision was challenged in the administrative courts—emphasizing the claimants’ economic position and the press’s constitutional duty—the Supreme Court, while acknowledging that “it is of particular importance to inform the public about the use of public funds for the development of private enterprises,” and that such reporting falls under the press’s watchdog role, nevertheless found no grounds for exempting the press. It affirmed that the press is required to fulfill notification obligations when relying on legitimate interest as a legal basis for data processing.

 

The definitions’ cornerstone: unfoundedness

 

Effectively countering SLAPPs requires clear definitions. This section examines those offered by the Directive and the CoE Recommendation, which both include unfoundedness. Unfoundedness has been central to debates over the Directive’s initial draft. Many still argue that it imposes an unnecessary limitation on what constitutes a SLAPP, potentially hindering the effectiveness of action.

The CoE Recommendation describes unfoundedness as one of several indicators that could help in recognizing SLAPPs, allowing a broader margin of appreciation for legal interpreters. In contrast, the Directive’s scope is limited to unfounded claims.

According to its title, the Directive operates within a dichotomy, providing safeguards against (a) manifestly unfounded claims or (b) abusive court proceedings. While it does not define “manifestly unfounded” or “unfounded”, it expands the definition of “abusive court proceedings against public participation.”

According to the definition, “‘abusive court proceedings against public participation’ mean court proceedings which are not brought to genuinely assert or exercise a right, but have as their main purpose the prevention, restriction or penalisation of public participation, frequently exploiting an imbalance of power between the parties, and which pursue unfounded claims”. Although the title and scope of the Directive suggests (a) and (b) as alternating categories (as indicated by the conjunction “or”), the definition of abusive court proceedings introduces “and,” requiring unfoundedness as part of both categories. This raises the question of whether the two are truly alternatives. (Note: most interpretations suggest that (a) and (b) are indeed alternatives, however, that contradicts the grammatical interpretation.)

One understanding could be that the Directive places “manifestly unfounded” claims and “abusive court proceedings” on a spectrum—with “manifestly unfounded”, as ab ovo unfounded at one end and “abusive” cases, potentially less clearly unfounded, further along that continuum. However, this interpretation creates additional uncertainty for courts in determining where to position a given case on that spectrum.

An interpretation aligned with paragraph 29 of the Directive’s preamble—which provides context for its operative definitions—suggests that only proceedings that are either fully or partially unfounded can be classified as abusive. If this is accepted, the Directive effectively collapses its own dichotomy, making unfoundedness the sole defining element and rendering the distinction between the two categories functionally meaningless. 

This distinction becomes most relevant when determining the appropriate safeguards. Defendants facing manifestly unfounded claims benefit from an early dismissal mechanism, whereas those facing abusive court proceedings—though still partially unfounded—must endure the full process and may only seek reparation after proceedings conclude. The legal uncertainty leaves the court’s decisions subject to accusations of cherry-picking. 

Despite earlier debates over elements of the definition and criticism of the distinction between manifestly unfounded claims and abusive court proceedings in terms of available remedies, this differentiation has persisted, along with the ambiguity surrounding 'unfoundedness.' The lack of a clear definition has left stakeholders in a state of legal uncertainty.

 

Unfoundedness in the context of GDPR-based SLAPPs

 

When examining what unfoundedness means for GDPR-based SLAPPs in Hungary, it is essential to continue distinguishing between the primary claim and the ancillary claims.

As discussed previously, the family members raised two distinct claims: the primary claim, namely the lack of legal basis for processing personal data, and the ancillary claims, concerning the failure to adhere to its procedural obligations.

First, let us examine the primary claim. The family argued that, in the absence of a legal basis, Forbes had no right to publish their personal data. The courts ultimately held that the press had a legitimate interest in reporting on the family members, given their receipt of public funds. However, as the GDPR was a relatively new legal instrument and no relevant precedent existed at the time, the legal question was considered unsettled until a final judgment had been delivered. Consequently, until then, the possibility of classifying such claims as unfounded could not have been seriously contemplated.

And although this decision enabled the press to report on the family in these circumstances, the publication of the family members’ personal data in other contexts will likely continue to be assessed on a case-by-case basis, meaning such claims may not be considered ab ovo (manifestly) unfounded. The case illustrates that unsettled legal questions are inherently difficult to classify as unfounded, allowing SLAPP proceedings to persist and continue imposing a burden on the press.

Second, when examining the ancillary claims, defining "unfoundedness" becomes even more ambiguous. Article 85 of the GDPR states: “Member States shall by law reconcile the right to the protection of personal data pursuant to this Regulation with the right to freedom of expression and information.” From a legal positivist perspective, the absence of implementing legislation under Article 85 of the GDPR has significant consequences. Since no national laws have been adopted to clearly define the boundaries of such reconciliation (for example, by exempting the press from the obligation to preliminarily, individually, and proactively inform data subjects, that is required of other data controllers like big companies), individuals may lawfully invoke GDPR provisions even in ways that restrict journalistic activities or the freedom of the press. As a result, legal claims based on alleged violations of GDPR obligations by the press cannot automatically be deemed unfounded. Therefore, under the Directive’s definition, such claims cannot be classified as SLAPPs.

However, the legal positivist approach is problematic, as it completely disregards context and fundamental rights aspects. From a fundamental rights perspective, it is contrary to freedom of the press to require full compliance with all GDPR-based duties, as it significantly hinders the press’s constitutional role. The lack of reconciliation in Hungary therefore constitutes a constitutional loophole, and exploiting such a loophole should never constitute a well-founded claim: applying the law in a way that contradicts the state’s positive obligations to protect the press and disproportionately hinders its operation is inherently problematic. It also disregards the state’s obligations stemming from the GDPR itself, as it uses mandatory language.

Furthermore, at the European level, the varying degrees of reconciliation between freedom of the press and data protection under Article 85 make it increasingly difficult and uncertain to draw a consistent line around unfoundedness.

 

The European interest

 

While the lawsuits against Forbes will most likely fall outside the Directive’s scope—due to their domestic nature and the fact that the procedure based on the DPA’s decision is administrative—interpreting the Directive’s definition remains relevant, particularly for future cross-border cases that do fall within its scope. Furthermore, the Directive sets only a minimum standard, meaning that national transpositions can expand its scope to include domestic cases, where unfoundedness would still be a determining criterion. Additionally, early-dismissed cases will likely reach the ECtHR, whether brought by the press or the claimant—ultimately forcing the Court to engage with the Directive’s interpretation. The relevance of interpreting the definition of the Directive extends beyond GDPR-based SLAPPs, as other claims that lack precedent or exploit constitutional loopholes can fall outside the scope of the Directive due to the definition. 

As the CoE Recommendation’s scope is not limited to cross-border claims, assessing the current cases from its perspective is highly relevant. In fact, since the ECtHR was established by the Council of Europe, the CoE Recommendation remains an important interpretive source when the Court rules on SLAPP-related cases.

These GDPR-based cases highlighted that the prolonged proceedings and ongoing legal uncertainty drain press resources and have already created a chilling effect. However, within the Directive’s framework, GDPR-based SLAPPs may not even fit the definition of “abusive court proceedings”. Even if they do, it is unlikely they would qualify as “manifestly unfounded,” placing them outside the scope of the early dismissal mechanism. As a result, the Directive might fail to effectively combat SLAPPs, especially the ones emerging in legal grey zones—even when defendants (the press) ultimately win. To put it more bluntly, the narrow definition could completely thwart the objective of the Directive and jeopardize its long-term legitimacy.

While broadening the definition of manifestly unfounded claims carries risks, it is unlikely that the drafters intended early dismissal to apply only in rare cases. The CoE Recommendation’s approach appears to offer a more suitable reference point for identifying SLAPPs. But let us wait and see what the ECtHR has to say. Until then, legal uncertainty continues to shield SLAPPs under the guise of procedural compliance. 

 

Acknowledgements: I would like to sincerely thank Beatrix Vissy and Tivadar Hüttl for their valuable insights and contributions.

National Limits, European Claims: The Ultra Vires Conflict in EU Constitutional Law (part 2)

 

Dr Benedikt Riedl, Mjur (Oxford),  postdoctoral researcher and academic assistant at the Chair of Public Law and Legal Philosophy (Prof Dr Peter M. Huber, former Justice of the Federal Constitutional Court and former Minister of the Interior) at LMU Munich.

Photo credit: Nicola Quarz, via Wikimedia Commons

 

Ultra Vires Review as a Legal Concept within EU Law

When disputes arise over the boundaries of EU authority, a deeper constitutional tension surfaces: Quis iudicabit – who gets to draw the final line? This question shapes one of the central controversies in EU law: the legitimacy and limits of ultra vires review.

The first part of the blogpost argued that ultra vires review is not an isolated or purely national phenomenon, but a widespread and constitutionally grounded legal practice across the EU. This second part goes further, arguing that ultra vires review is not only compatible with Union law but embedded within it. Anchored in the principle of conferral (Article 5(2) TEU) and safeguarded by Article 4(2) TEU, the authority of national constitutional courts to assess whether EU institutions have acted beyond their powers forms an integral part of the EU’s composite legal order. Rather than threatening the primacy of EU law, ultra vires review reinforces its democratic legitimacy by ensuring that integration remains bound by constitutional limits. To that end, this blogpost also sets out four cumulative steps for invoking ultra vires review as a concept of EU law – proposing a structured, treaty-compatible procedure designed to enhance mutual trust, legal certainty, and institutional balance across the European constitutional landscape.

Doctrinal classification of ultra vires review as a general principle of Union law

1 The authority of Member States’ constitutional courts to declare acts of EU institutions ultra vires – and, as a result, to deny them legal effect within the national legal system – constitutes a general legal principle of EU law. This conclusion is grounded in the EU’s foundations in public international law and supported by comparative constitutional analysis. The invalidation of legal acts that manifestly exceed EU competences follows from a systematic interpretation of the fundamental principles underlying EU law. Moreover, the EU Treaties contain no provision that prohibits Member States from conducting ultra vires reviews. Accordingly, ultra vires review should be understood as an established legal concept within the framework of EU law.

Principles of International Law: A Derived Legal System

2 The CJEU asserts the absolute autonomy of EU law without offering a clear normative foundation in the treaties or justification under the law as it stands (de lege lata). In reality, however, it is the national constitutions that provide both the foundational basis and the essential point of reference for EU law. This becomes evident through an analysis of the EU treaties and their historical development. The treaty text, treaty practice, and drafting history all consistently affirm this perspective. The treaty text includes numerous provisions that explicitly recognize the existence and evolution of the Union’s legal order as deriving from the will of the Member States. Treaty practice further reinforces this: several treaty amendments have failed due to the Member States’ refusal to ratify them. This shows that the EU legal order has never been truly autonomous; it has always been contingent on the voluntary political decisions of the Member States. The EU is based on the collective will of its Member States under international law. Its legal order draws its validity and legitimacy from the national acts of assent and the authorisation granted by Member States for the application of EU law. Only on this foundation can the EU legal order maintain the level of democratic legitimacy necessary for a direct connection to the citizens of the Member States.

3 Given the EU’s interconnected legal structure, the position adopted by the Member States is persuasive. It accurately reflects the constitutional reality of Europe: a union of open constitutional states committed – albeit sometimes cautiously – to cooperative judicial solutions. The constitutions of the Member States serve as both the legal foundation and the primary reference point for EU law. As a result, EU law and national constitutional law form a single, integrated legal system. Within this framework, ensuring effective legal protection is a shared responsibility – carried out through a cooperative network of constitutional courts, referred to as the Verfassungsgerichtsverbund in German legal scholarship.

4 Absolute primacy of EU law or the notion of an autonomous legal order cannot be established through subsequent practice or customary law for several reasons. First, the failure of the Constitutional Treaty clearly demonstrates a lack of political consensus among the Member States to codify a clause providing for the EU law’s absolute primacy. Second, due to the interconnected and multi-layered nature of EU law, Member States cannot be reduced to their governments alone. While governments may ratify treaty amendments without reservations, this does not take into account the constitutional courts, which act as additional constitutional authorities. Their jurisprudence must be taken into consideration in areas where legal orders overlap – and these courts have consistently rejected the idea of absolute primacy of EU law. Finally, customary law cannot emerge contra legem – that is, in contradiction to the express provisions of the EU treaties. As will become clear in the following, this may include Articles 2 TEU, 4(2) TEU, 5(2) TEU, 10 TEU, and 19 TEU.

5 Furthermore, the CJEU’s concept of autonomy is incompatible with its own jurisprudence. In Van Gend & Loos and Costa v. E.N.E.L., the CJEU explicitly grounded the primacy of EU law on the principle of conferral. Recently, however, the Court contradicted itself: It has invoked the unconditional ratification of amending treaties to support its position while simultaneously rejecting the justification of primacy based on a delegation of powers. The demands for uniformity and effectiveness of EU law do not justify arbitrary interference with Member States' constitutional identities. In summary, the concept of autonomy lacks both factual support and legal justification.

6 EU law does not derive its legitimacy from a constitutional act or any other foundational norm; instead, its legitimacy is functional rather than constitutional or supra-legal in nature. This limited and contingent basis for legitimacy distinguishes EU law from the constitutional systems of the Member States. As a result, EU law can only evolve into a genuinely autonomous legal order through a process of constitutionalisation by the European peoples – a process that has not yet taken place. The functional and pragmatic rationale underlying EU law, which is focused on practical effectiveness, neither establishes genuine autonomy nor justifies claims to absolute institutional or substantive authority.

7 Since EU law is not autonomous but derived from – and closely linked to – the constitutions and democratic will of the Member States, ultra vires review aligns with the foundational principles of EU law. Due to EU law’s character as a derived legal order, national courts may, in certain cases, review the application of secondary EU law and assess its validity within their domestic legal systems – especially when EU institutions exceed the competences conferred upon them by the Member States.

Comparative Constitutional Analysis: The Invalidation of EU Measures Exceeding Conferred Powers as a Core Legal Principle in Member State Constitutions

8 Comparative constitutional analysis demonstrates that ultra vires review is the procedural and institutional expression of a general constitutional principle: EU legal acts that exceed the competences conferred through the national parliaments’ acts of assent are not valid within the Member States. Here, national constitutional courts step in to conduct an ultra vires review when the CJEU fails to assess the limits of EU competences. By safeguarding the division of powers between the EU and Member States, these reviews uphold a fundamental constitutional principle. While national constitutions refer to this principle in different terms – such as sovereignty, the republican form of government, democracy, or popular sovereignty – they all serve the same functional purpose: to ensure the democratic legitimacy and internal coherence of EU measures within the national legal order.

9 From a formal standpoint, ultra vires review implements the principle of limited conferred powers in a procedural way. It builds on the widely accepted legal principle that actions beyond conferred competences are prohibited. The EU’s lack of competence-competence (Kompetenz-Kompetenz) – i.e. its inability to unilaterally define the scope of its own powers – is explicitly affirmed in Article 5(2) TEU. Since competences are transferred through acts of the Member States, all constitutional courts recognizing ultra vires review consider procedural enforcement of these limits a constitutional necessity. This ensures that the substantive principle of limited conferred powers is not left without procedural safeguards. This logic is also persuasive within the context of the European composite legal order: while no constitutional court disputes the CJEU’s authority to interpret the limits of EU competences, all reject the notion that the CJEU can unilaterally determine the boundaries of its own jurisdiction. The domestic legal consequences of such determinations are governed by each Member State’s acts of assent. In any event, the CJEU has not been granted the ultimate authority to define or transform the constitutional orders of the Member States.

10 In substantive terms, ultra vires review serves to protect both popular and state sovereignty. Its justification lies in the idea that when EU institutions overstep their competences, they undermine democratic accountability and violate the right of citizens in the Member States to self-determination. Political self-determination is understood both individually and collectively, as tied to the state. The core issue remains the same: if EU institutions act beyond their conferred powers, those actions lack a proper legal basis and, consequently, democratic legitimacy. These acts intrude upon the regulatory autonomy of the Member States and the democratic self-determination of their citizens. Ultimately, it is up to the sovereign – the citizens of the Member States – to decide whether and to what extent EU law applies within their national legal systems. This is determined by the respective parliaments through their acts of assent, which define the scope of EU law within the national constitutional framework via the so-called “integration programme.” Under a compelling interpretation, ultra vires review thus acts as a procedural safeguard of the European peoples’ right to self-determination ensuring the democratic responsiveness of EU law to every citizen. At the same time, it enables the very transfer of sovereign powers, without which the democratic principle itself could be placed at risk.

Article 5(2) TEU: The Composite Structure and Constitutional Anchoring of EU Law

11 EU law and national constitutional law form an integrated legal system. The European Union is a union of constitutional states open to legal integration and cooperation. European secondary law and national law are structurally interconnected through the framework of European primary law. Accordingly, the constitutions of the Member States contribute to an interlinked constitutional order. Legal questions that arise at the intersection of different legal systems must therefore be assessed in light of the mutual interdependence between national law and EU law. The principle of conferral has both a European and a national dimension and is therefore of dual nature. At the European level, it is enshrined in Article 5(2) TEU; at the national level, it is implemented through constitutional opening clauses and acts of assent authorising the application of EU law. In EU law, this principle reflects the limited transfer of sovereign powers from the Member States to the Union. As such, it is not solely a principle of EU law; it also incorporates into primary law the constitutional reality of the Member States and the EU’s dependence on their sovereign decisions under international law. Like Article 4(2) TEU, it expresses both the constitutional origin and the composite nature of the Union’s authority.

12 Moreover, the CJEU itself is directly bound by the principle of conferral, as evidenced by Article 19 TEU. The limited scope of its judicial mandate – in both competence and legitimacy – is clearly articulated in the EU Treaties: Article 19(1), first subparagraph, sentence 2 TEU states that the CJEU ensures the observance of the law "in the interpretation and application of the Treaties," while Article 19(1), second subparagraph TEU limits effective legal protection to "the fields covered by Union law." In this way, the European judicial system reflects the EU’s compound structure. The interconnection of judicial systems operates in both directions: the CJEU is integrated into the judicial framework of the Member States, and national courts are functionally embedded in the enforcement of Union law. Legal protection within this network of constitutional courts is established in the EU Treaties as a shared responsibility between the CJEU and the constitutional courts of the Member States.

13 The involvement of national constitutional courts in the ultra vires procedure is the procedural and institutional consequence of the dual nature of the principle of conferral. Since questions of competence arise in both the EU sphere (interpretation of EU law) and the Member State sphere (interpretation of the integration programme and constitutional opening clauses), resolving such conflicts requires cooperation between the CJEU and national constitutional courts in accordance with Article 19 TEU. The procedural treatment of ultra vires conflicts thus reflects the multidimensional nature of the primacy of EU law and must therefore be addressed institutionally at both levels – by the CJEU as well as the national constitutional courts.

14 The transfer of competences to the EU takes place through national legal acts in accordance with constitutionally prescribed procedures and therefore logically precedes any consideration of the primacy of EU law. These constitutional parameters – reflected in the EU Treaties, particularly in Article 5(2) and Article 19 TEU – form the basis for the residual competence of national constitutional courts to review the boundaries of conferred powers. In principle, Member States have entrusted the CJEU with this competence of review, and national constitutional courts can rely on the CJEU to fulfil it in accordance with Article 19(1), subparagraph 2 TEU. However, if the CJEU fails to adequately exercise its control function, national constitutional courts retain a fallback responsibility to safeguard the proper delineation and exercise of competences. Absent this fallback mechanism, they would ultimately be stripped of their original role of protecting the national constitution and identity, reducing their function to a mere absurdity. The ultra vires procedure – while primarily vested in the CJEU and exercisable only as a last resort by national constitutional courts – derives directly from the principle of conferral embedded in the acquis communautaire of EU law.

Democratic principles and the context of legitimacy in the European Union

15 The involvement of national constitutional courts in ultra vires procedures can be justified not only by the principle of conferral but also by invoking Article 2 TEU and the democratic principles underpinning the European Union. This argument is based on the role of national parliaments and the principle of conferred powers. It underscores the fact that when EU institutions act beyond their competences, the essential link of democratic legitimacy between the Union's authority and the citizens of the Member States is compromised. In such instances, the control conducted by constitutional courts serves as a necessary corrective to uphold the democratic foundation of the EU.

16 The European Union derives its democratic legitimacy through two channels: the European Parliament (Article 10(2) subpara. 1 and Article 14 TEU) and national parliaments (Article 10(2) subpara. 2 and Article 12 TEU). The EU's well-documented democratic deficit – characterised by issues such as unequal voting rights, an insulated executive branch, and limited public engagement – is mitigated by the Member States’ respective democracies, particularly through their national parliaments. These parliaments play a crucial role in legitimizing the EU. A distinction can be drawn between the European Parliament’s direct exercise of democratic functions and the national parliaments’ broader responsibility for overseeing the integration process. If EU institutions overstep their competences and thereby undermine the authority of national parliaments, one of the foundational pillars of European democracy is weakened, resulting in a decline of democratic legitimacy for the EU as a whole.

17 The principle of conferral is closely tied to the democratic values enshrined in Article 2 TEU. A core element of the EU’s democratic framework is the concept of legitimacy based on competences. As a derived legal order, the EU is legitimised through the competences conferred upon it. The democratic principles of the Union go beyond a purely formal notion of authority based on the transmission of the citizens’ will into institutional power. Instead, they require a substantive, effective realisation of that will – specifically, the guarantee of genuine popular sovereignty. Grounded in the principle of conferral and reinforced by Article 4(2) TEU, the EU legal order itself respects the legislative authority of national parliaments and the democratic will of the citizens in the Member States.

18 Ultra vires review plays a pivotal role in preserving democratic legitimacy within the European Union – specifically, the legitimacy that flows through the parliaments of the Member States. Its primary function is to uphold democracy within the European network of states, constitutions, legislative bodies, administrations, and judicial systems. By monitoring the limits of conferred competences, ultra vires review serves as a safeguard or compensatory mechanism critical for upholding democracy – in other words, it is democracy-specific. This is particularly significant for preserving individuals’ rights to self-determination. In the context of European supranational integration, ultra vires review exposes the diffusion of political responsibility, thereby promoting greater transparency and accountability in decision-making by EU institutions.

19 In principle, the CJEU is responsible for ensuring compliance with the division of competences within the EU. Even if a decision is incorrect, a competence review has still taken place. However, if a decision by the CJEU no longer meets the required methodological standards, the review is reduced to a merely formal check. In such cases, the substantive core of the European understanding of legitimacy is no longer upheld. At the same time, Article 2 in conjunction with Article 5(1) and (2) TEU allows for a decentralised review of competences by constitutional courts in the Member States. These courts supervise the boundaries of the EU’s derived authority to safeguard the substantive foundations of democracy – namely, meaningful parliamentary influence and effective popular sovereignty.

Rule of Law and Separation of Powers in the Judicial System

20 Several factors have contributed to the CJEU's de facto position as an almost unchecked authority within the EU legal order. These include the constitutionalisation of EU law, its limited embeddedness in a European public sphere, the CJEU’s dual role as both a constitutional and specialised court, the absence of sector-specific competence rules, and its broad, teleological approach to interpretation. This situation is further exacerbated by the fact that the CJEU increasingly positions itself not as the “guardian” of the EU legal order, but as the “engine of integration.” In this role, it applies double standards in its judicial review of competences: When reviewing infringements of EU law by Member States, the CJEU exercises strict scrutiny. By contrast, in cases where EU institutions themselves exceed their competences, it displays significant leniency. To date, the CJEU has found an overreach of competences by EU institutions only in four cases. In some judgments, the level of scrutiny applied to EU bodies is so minimal that it amounts to a superficial or merely evidentiary review. This approach conflicts with the EU’s commitment to representative democracy under Article 10(1) TEU, the right to effective legal protection, and the obligation to examine facts under Article 47 CFR and Articles 6 and 13 ECHR. This jurisprudence has contributed to a gradual shift of competences toward the EU without any formal treaty amendments. As a result, it undermines the legal basis of institutional action and weakens formal legitimacy within the EU. The expanding scope of EU competences, without a corresponding increase in democratic legitimacy, also diminishes the Union’s substantive legitimacy.

21 Within the European judicial system, the CJEU regards itself as the highest authority, positioned above the national constitutional courts. While the Court has repeatedly acknowledged both the EU’s lack of statehood and the limited scope of its competences, it nonetheless expands its jurisdiction into core areas of national constitutional law – thereby infringing upon the jurisdictional authority of the Member States' highest courts. This leads to a growing disconnect between the principles of direct applicability and primacy of EU law, on the one hand, and the foundational principle of conferral, on the other. It also results in EU law penetrating the most protected spheres of national constitutional orders. In doing so, the CJEU disregards the jurisprudence of national constitutional courts – even in areas where EU law and national constitutional law are closely interlinked.

22 In this context, ultra vires review plays a key role in safeguarding the separation of powers and establishing checks and balances within the European network of constitutional courts. A judicial equilibrium between the EU and the national level can be maintained through disagreement as well as ongoing dialogue. Clearly defining the limits of competence enhances legal certainty and builds trust in future steps toward integration. Moreover, ultra vires review offsets the CJEU’s limited capacity for internal control within the European constitutional system – a limitation resulting from the absence of a judicial hierarchy, insufficient public scrutiny, lack of opposition, and constrained legislative oversight. At the same time, it acts as a safeguard, protecting national constitutional systems from an unchecked assertion of EU authority.

Article 4(2) TEU: Protecting Constitutional Identity as a Systemic Exception to the CJEU’s Exclusive Authority to Invalidate EU Acts

23 Article 4(2) TEU directly reflects the considerations outlined above, which stem from the EU’s compound constitutional structure. This provision can be understood as implicitly acknowledging that, in strictly limited and extraordinary situations, Member States may invoke ultra vires review. As will be shown below, such an interpretation does not conflict with the general principles of EU law, but rather takes into account the structural diversity and constitutional identity of the Member States. Article 4(2) TEU may therefore be read as establishing a system-inherent exception to both the primacy of EU law and the CJEU’s exclusive jurisdiction to invalidate Union acts. In rare cases, this exception could be asserted procedurally through an ultra vires review, serving as a last-resort safeguard within the broader framework of European legal integration.

24 Article 4(2) TEU stipulates that “the Union shall respect the equality of Member States before the Treaties as well as their national identities, inherent in their fundamental structures, political and constitutional, inclusive of regional and local self-government.” Fundamental constitutional structures within the meaning of Article 4(2) TEU can encompass a wide range of core constitutional principles, including specific expressions of democracy or the rule of law in Member States. However, the CJEU’s case law on the extent to which Member States may define and invoke their own constitutional identity remains inconsistent and highly context-dependent. For instance, in the M.A.S. and M.B. rulings involving the Italian Constitutional Court, the CJEU showed cautious openness to national constitutional reservations and signalled a willingness to engage in dialogue with national constitutional courts. In contrast, in cases involving the Romanian Constitutional Court’s constitutionally controversial jurisprudence, the CJEU rejected the admissibility of both identity review and ultra vires review under EU law.

25 The interpretation of the wording, structure, history, and purpose of Article 4(2) TEU confirms its character as a reference to national constitutions and the authority of national constitutional courts to define constitutional identity. The wording – which states that national identity is expressed “in their” fundamental constitutional structures – clearly points to domestic constitutions as the relevant legal source. Structurally, the provision complements Articles 4(1) and 4(3) TEU and establishes a binding limit on the Union’s integrative reach. Historically, the travaux préparatoires of the Constitutional Treaty show that responsibility for safeguarding constitutional identity was deliberately left to the Member States, in order to prevent a creeping expansion of EU competences. In terms of purpose, Article 4(2) is meant to protect core constitutional choices from being overridden by Union law. It functions similarly to self-judging clauses in international law, which allow states to make their own determinations, subject only to limited review. In this way, the provision establishes a direct link between Union law and national constitutional law and forms a key element of the EU’s composite constitutional structure. As a counterpart to national identity clauses such as Article 79(3) of the German Basic Law, Article 89(5) of the French Constitution, or Article 9(2) of the Czech Constitution, it opens Union law to core components of national constitutional orders. Defining what constitutes national identity ultimately falls to national constitutional courts, in line with the division of responsibilities in the European constitutional court network, while the CJEU is limited to checking for arbitrariness or abuse.

26 In narrowly defined exceptional cases, Article 4(2) TEU leads to a systemic suspension of the primacy of EU law and constitutes a procedural exception to the CJEU’s monopoly of interpretation. This consequence is inherent in Union law itself, since both the principle of conferral and the protection of national constitutional identity define structural limits to the scope of primacy. In narrowly defined exceptional cases, Article 4(2) TEU leads to a systemic suspension of the primacy of EU law and constitutes a procedural exception to the CJEU’s monopoly of interpretation. This consequence is inherent in Union law itself, since both the principle of conferral and the protection of national constitutional identity define structural limits to the scope of primacy. While the obligation to respect national identity under Article 4(2) TEU is generally relative, it becomes absolute where the core elements of constitutional identity are at stake. This core includes those values, structural principles, and norms that, under a Member State’s own constitutional order, are categorically excluded from being transferred to the Union – either because they are formally entrenched or because they are considered non-transferable in structural terms. Even where national constitutional principles coincide with the fundamental values enshrined in Article 2 TEU, their particular national expression enjoys absolute protection, precisely because those values themselves hold overriding importance within Union law. What matters is not whether the national constitution contains an explicit eternity clause, but whether – according to the domestic constitutional understanding – it defines an inviolable sphere of sovereign self-determination. In such cases, the obligation to respect national identity is not subject to balancing or proportionality review. Accordingly, the Member State’s interest in safeguarding this core outweighs the Union’s interest in legal uniformity. While disregarding such identity would amount to a systemic violation of constitutional self-government, the legal consequence within Union law – namely, the inapplicability of the respective EU measure in that Member State – is a limited and acceptable cost. The resulting impairment of uniformity and effectiveness is consistent with the nature of the EU’s legal order, particularly since the ultra vires review, as an expression of democratic self-determination, also serves the values enshrined in Article 2 TEU. It is therefore grounded in primary law via Article 4(2) TEU and constitutes a necessary procedural exception within the non-hierarchical, pluralistic structure of the European constitutional order.

27 It has long been assumed that only the identity review in the narrow sense corresponds procedurally to this provision. However, as comparative constitutional analysis shows, the ultra vires review – understood as a democracy-protecting form of identity review – is also embedded in the constitutional structures of numerous Member States. This applies particularly where it serves, as in the German and Czech constitutional traditions, to safeguard popular sovereignty or sovereign statehood. A qualified overreach of competences breaks the necessary link of democratic legitimacy between Union action and the national constitutional order, thereby infringing upon the constitutionally protected core of democratic identity. In such cases, the ultra vires review functions as the procedural mechanism for preserving this core, and must therefore be regarded not only as an expression of domestic identity clauses – such as Article 79(3) of the German Basic Law or Article 9(2) of the Czech Constitution – but also as a review mechanism permitted under Union law by virtue of Article 4(2) TEU. The jurisprudence of various national constitutional courts confirms this classification. Ultra vires review thus constitutes a procedurally legitimate expression of the Union’s obligation to respect national identity, reflecting both the multidimensional nature of the primacy principle and the institutional logic of the European constitutional court network.

No exclusion on the basis of conflicting EU law

28 From the CJEU’s perspective, ultra vires review by Member State constitutional courts violates EU law for several reasons: it threatens the uniformity and effectiveness of EU law, challenges the CJEU’s exclusive authority to declare EU acts invalid, and undermines the principle of equality among Member States and their citizens. However, these objections are not convincing from a doctrinal and legal-theoretical standpoint.

29 Ultra vires review does not undermine the CJEU’s exclusive authority to annul EU legal acts. While both the jurisprudence established in the Foto-Frost judgment and Article 267(1)(b) TFEU prohibit national courts from declaring EU acts invalid, national courts are not categorically barred from examining whether such acts exceed the competences conferred on the EU. When a national constitutional court reviews the scope of EU law on the basis of its act of assent, it applies a distinct standard – as confirmed by comparative constitutional law. This is not a parallel review of EU law, but a narrowly limited control, typically confined to manifest excess or arbitrary transgression of conferred competence. On the remedial level, a finding that an EU act is ultra vires does not invalidate it across the European Union; it merely renders the act inapplicable within the respective national legal order. Furthermore, the strict separation between the procedural question of Quis iudicabit? and the substantive principle of conferral appears to be a politically motivated construct – one that effectively grants the CJEU de facto authority over the very scope of its own powers (i.e. the judicial competence over competences).

30 Moreover, isolated ultra vires reviews can, in fact, strengthen the EU legal order in the medium and long term. They help build trust in the European integration process, as Member States are more willing to transfer additional competences when assured that the limits of such transfers will be respected. Conversely, if competences gradually extend beyond their original conferral, thereby weakening the link between direct applicability and the principle of conferral, the effectiveness and credibility of EU law risk being undermined.

31 The Member States have not – and, indeed, could not – conferred upon the CJEU the authority to determine the limits of its own judicial competence, as doing so would inherently restrict their own constitutional identity and spheres of competence. This is because the constitutional core of most Member States, often entrenched or unamendable, impose strict limits on the transfer of sovereign powers. The constitutional clauses authorizing the application of EU law only extend to acts adopted within the limits of conferred competences and consistent with national constitutional identity. This logic is reflected in Article 19(1), second sentence, TEU, which explicitly ties the CJEU’s jurisdiction to the principle of conferral.

32 Despite the CJEU’s recent explicit rejection of ultra vires review – albeit in the context of rulings involving constitutionally compromised courts – its judgment in M.A.S. and M.B. (Taricco II, 5 December 2017) contains indications that the Court acknowledges the legitimacy of constitutional review reservations. This effectively amounts to an exception to its asserted exclusive authority over competence disputes (judicial competence over competences). The ruling suggests a de facto recognition of identity review or, at the very least, a greater deference to national constitutional identity when weighed against the primacy and uniform application of EU law.

33 The CJEU’s new line of reasoning – seeking to justify absolute primacy and a comprehensive monopoly of jurisdiction through the principle of equality in Article 4(2), sentence 1 TEU – suffers from fundamental methodological flaws and is therefore unconvincing from the outset. It misinterprets the wording of Article 4(2) TEU, which imposes obligations solely on the Union, not on the Member States, ignores the internal structure of the provision, which expressly balances the equality of Member States with the respect for their national identities, and disregards its purpose as a protective provision in favour of the Member States rather than as a tool to expand Union competences. Moreover, it rests on a misapprehension of the function of the principle of equality within the EU legal order. Where a situation falls outside the scope of Union law, the principle of equality derived from it is inapplicable from the outset. Accordingly, the CJEU’s argument is circular, because ultra vires review, by definition, concerns precisely those cases where Union law is not applicable.

34 Since ultra vires review is not only compatible with the EU Treaties but also rooted in them, the involvement of national constitutional courts should be recognized as a general principle of EU law. Ultra vires review is grounded in the foundations of international law and the constitutional evolution of the EU legal order. It represents a procedural and institutional expression of the dual nature of the principle of conferral and safeguards the separation of powers within the constitutional court system – particularly, with respect to the legitimacy of European democracy. This review mechanism is justified under Article 4(2) TEU as a treaty-based exception to the CJEU’s exclusive authority to annul EU acts. As a legal concept, the ultra vires procedure reflects the compound structure of the EU, composed of interlinked constitutional systems and courts. Given the exceptional nature of this general legal principle, both the burden and the authority to justify invoking ultra vires review rest with the constitutional and supreme courts of the Member States. This approach creates a balanced legal framework that preserves the effectiveness and uniform application of EU law, while ensuring that ultra vires reviews remain legally contained and subject to clear procedural safeguards.

 

In summary, ultra vires review should be recognised as a structural element of the European legal order. As a procedural expression of the principle of conferral and as a safeguard for constitutional identity under Article 4(2) TEU, it serves a critical democratic and legal function. Rather than contradicting Union law, it preserves the legitimacy of EU integration by ensuring that the division of competences remains constitutionally anchored. The next and final part of this blogpost will examine how ultra vires review can be institutionalised within the existing framework of EU law – not as a threat to legal unity, but as a mechanism for structured constitutional dialogue.

Institutionalising Ultra Vires Review: The Four-Tier Test for Constitutional Courts

The following analysis outlines four cumulative stages that must be observed when invoking ultra vires review within the European constitutional framework. Rather than undermining the primacy of Union law, this structured procedure enhances mutual trust and legal certainty among Europe’s constitutional courts. It clarifies the limited and exceptional conditions under which national courts may act as guardians of their constitutional mandates – while safeguarding the rule of law, democratic legitimacy, and the institutional balance of the EU’s composite legal order.

 

1 The involvement of Member State constitutional courts in the common ultra vires procedure is based not only on their respective national constitutions and acts of assent, but also on Articles 2, 5(2), and 4(2) TEU. Read together with Articles 4(3) and 19 TEU, these treaty provisions give rise to four substantive requirements – or stages – of the ultra vires procedure as a legal concept under EU law. First, there must be a violation of the principle of conferral or of the integration programme as enshrined in Articles 5(2) and 4(2) TEU. At every stage, ultra vires review must comply with the principle of sincere cooperation (Article 4(3) TEU), respect the judicial function of the CJEU pursuant to Article 19 TEU, and fulfil its role as a democratic safeguard mechanism in line with Article 2 TEU, ensuring the legitimacy of EU action.

2 At the first stage, there must be a legal act that exceeds the competences conferred upon the EU. An ultra vires act exists where EU institutions act without a proper legal basis, go beyond the conditions set for the exercise of their powers, or breach prohibitive norms, thereby infringing on the competences of the Member States. It is important to note that neither the obvious nature nor the particular gravity of a competence breach is a requirement under EU law; such thresholds cannot be derived from Article 5(2) TEU. In preliminary ruling proceedings under Article 267 TFEU, it is sufficient for a national constitutional court to have doubts about whether an EU measure falls within the Union’s competences. If the court considers the case to involve an ultra vires act, it refers the question concerning the scope of Union law to the CJEU. The ultra vires procedure thus initially differs from an ordinary preliminary reference only in that it is triggered not by any legal error, but specifically by a suspected overstepping of competences. The decisive difference arises only at the stage of assessing the CJEU’s response – namely, in the standard of review applied in ultra vires control.

3 At the second stage, the constitutional courts of the Member States and the CJEU are required to cooperate and engage in judicial dialogue within the framework of the preliminary ruling procedure. From the perspective of EU law, ultra vires review necessitates the involvement of the CJEU in the decision-making process, as provided by Article 4(3) first subparagraph, Article 19 TEU, and Article 267 TFEU. From the perspective of national constitutional law, this cooperation is equally required due to the constitutions’ openness to European law. As a result, the ultra vires procedure constitutes a matter of shared responsibility, which can only be fulfilled through mutual cooperation and constructive dialogue in the context of the preliminary ruling process.

4 This requires the constitutional courts of the Member States to submit a clear and well-reasoned question to the CJEU, providing all the necessary elements for the Court to reach a conclusive judgment on the ultra vires issue. The submission must include detailed legal arguments indicating a possible overreach of competences. This transparency is essential for resolving a competence dispute through legal reasoning. There must be a realistic opportunity for one court to recognise the other’s legal position as methodologically sound – or even to be persuaded by it. A notable example of this cooperative approach is the GCC’s referral in the OMT case. The GCC comprehensively explained why it believed the OMT decision violated Article 119 TEU, Article 127(1) and (2), and Article 123 TFEU. At the same time, it demonstrated its willingness to cooperate by allowing room for the CJEU to interpret the legal issues in a way that aligned with its conferred competences, thereby upholding the spirit of judicial dialogue within the preliminary ruling procedure.

5 At the third stage, the CJEU is responsible for interpreting EU law, though its decisions remain subject to methodological scrutiny by the constitutional courts of the Member States. If a ruling fails to meet methodological standards, the CJEU exceeds its mandate under Article 19 TEU and may be overruled through ultra vires review. The CJEU’s authority to interpret EU law is derived from the Member States via the act of assent, as codified in Article 19(1) subparagraph 1, sentence 2 TEU. However, this authority is no longer democratically legitimised if the interpretation goes beyond what Member States could reasonably foresee when granting assent; this is particularly the case when the CJEU tends to judicial legislating rather than interpreting the law. The CJEU clearly oversteps its jurisdiction when it fails to respect the methodological boundaries of judicial interpretation. This lack of methodological appropriateness can take several forms: (a) contradictions and inconsistencies in case law that affect the outcome, (b) disregard for standard methods of legal interpretation, (c) failure to engage with arguments raised by the referring court, (d) arbitrary judicial development, and (e) complete neglect of constitutional standards in its review. In such cases, the effective control of jurisdiction is no longer ensured, and the fallback function of national constitutional courts is reactivated as to protect the division of competences and uphold the legitimacy of EU action.

6 The CJEU’s Weiss judgment illustrates an ultra vires ruling that lacks methodological appropriateness due to internal contradictions and inconsistencies (a). Specifically, the Court applied the proportionality test to EU institutions in a far more superficial manner – without justification – than it typically does when reviewing acts by Member State authorities. Furthermore, the CJEU’s complete disregard for the economic policy implications of the PSPP conflicted not only with its reasoning in Gauweiler but also with its general approach in other areas of EU law.

7 A further serious methodological error (b) lies in the CJEU’s failure to apply the established methods of legal interpretation – such as textual analysis, historical context, systematic structure, teleological reasoning, and comparative law. The constitutional traditions of the Member States set boundaries for European legal methodology and require that the CJEU’s interpretative approaches remain consistent with these traditions. CJEU rulings must be sufficiently reasoned to allow for public scrutiny and review by other courts, as mandated by Article 36(1) of the CJEU Statute and Article 87(m) of the CJEU Rules of Procedure. If the CJEU disregards relevant facts or the arguments raised by parties to the case, this may result in legally unintelligible outcomes. A notable example is the Landtová judgment, in which the CJEU ignored both the wording of the European Migrant Workers Regulation and the historical context surrounding the dissolution of Czechoslovakia.

8 Based on its judicial function under Article 19 TEU and the principle of sincere cooperation, the CJEU is required to provide a sufficiently reasoned response to preliminary references, particularly regarding alleged breaches of competence (c.). The third category of judgments that fails to meet methodological standards thus consists of CJEU rulings that do not adequately address the concerns raised by the referring constitutional or supreme court. A key example is the CJEU’s ruling in Weiss. Among other issues, the GCC asked whether the PSPP infringed Article 119, Article 127(1) and (2) TFEU, as well as Articles 17–24 of the Protocol on the Statute of the ESCB and the ECB. The GCC identified the ECB’s failure to properly assess the proportionality of its measures as a central issue in determining whether it had acted within its competence. Specifically, the GCC was concerned about the economic policy effects of the ECB’s measures, which had previously gone unexamined. However, the CJEU did not take these detailed objections seriously and failed to engage with the economic impact in a thorough or coherent way. As a result, a proper proportionality assessment, in line with established legal methodology, was not carried out.

9 Fourth, arbitrary judicial lawmaking constitutes a case of methodological failure (d.). A clear example is the CJEU’s Mangold judgment, where the Court “created” the Union law principle prohibiting age discrimination based on a questionable and insufficient comparative analysis, without identifying a clear textual, structural, or historical basis in the Treaties. This overstepped the permissible bounds of judicial interpretation, amounting instead to an unauthorised amendment of primary law. Legitimate judicial development of the law serves to fill unintended gaps, resolve conflicts of fundamental values, and ensure fair adjudication in individual cases, but it is strictly limited: it must not result in legislative correction, encroach upon areas reserved for political decision-making, or undermine the principle of conferred powers. Furthermore, any judicial development must meet heightened requirements of justification; the comparative analysis must be comprehensive, transparent, and logically coherent, and particular care must be given to respecting the division of competences between the Union and the Member States.

10 The fifth category in which the CJEU manifestly exceeds its jurisdiction arises when it fails to fulfil its institutional judicial role and thereby falls short of the constitutional standards enshrined in Article 19 in conjunction with Article 2 TEU. This occurs, for example, when the Court fully or partially fails to fulfil its duty of judicial review – particularly with respect to executive or independent institutions – or when it applies the intensity of review arbitrarily (e.). A clear example of this is once again the Weiss judgment, where the CJEU effectively abdicated its judicial oversight role by conducting only a cursory review limited to assessing whether the ECB had committed “obvious” overreaches of its mandate. The Court merely asserted the proportionality between monetary policy measures and their economic consequences, rather than actually examining their consequences. This represents a fundamental methodological error: proportionality between ends and means cannot be meaningfully affirmed without first identifying and evaluating the actual effects. Such an examination was not even possible in this case, as the CJEU never clearly identified the economic implications of the ECB’s actions. This fifth category of judgments, which depart from established legal methodology by neglecting rule-of-law standards in judicial review, directly underscores the need for ultra vires control. It serves to prevent gradual, unchecked shifts of power towards the EU level and safeguards Member States’ authority over their own competences as well as the democratic self-determination of their citizens.

11 Although ultra vires review is a tool for safeguarding the legitimacy of EU action, it must be understood within the broader context of EU law. It is rooted in the democratic principles of the European Union and contributes to maintaining the separation of powers and upholding the rule of law. Therefore, its legitimacy and application – especially during the fourth stage of the procedure – are tied to compliance with the rule-of-law requirements set out in Article 2 TEU.

12 Constitutional courts retain a limited degree of discretion in interpretation and judgement. This means that, in exceptional cases, both the activation and rejection of ultra vires review by different Member State constitutional courts may be permissible with respect to the same legal issue. This outcome is also consistent with the EU’s interconnected legal structure, particularly given that the CJEU’s jurisdiction is derivative and, unlike the constitutional courts of the Member States, the CJEU does not possess the competence to determine the scope of its own powers.

13 Article 2 TEU implies a specific prohibition of abuse of rights within Union law, which also limits the scope of ultra vires review. Such review is only permissible where it serves its purpose of safeguarding democratic legitimacy within the Union through constitutional courts of the Member States. It becomes abusive – and therefore contrary to Union law – where it is functionally disconnected from that purpose, for example: (1) where the subject matter is not clearly limited, (2) where the ruling is not embedded in a consistent, EU-friendly line of jurisprudence, (3) where the judiciary is visibly instrumentalized by the executive, or (4) where the legal consequences are left unclear and risk systemic disruption. In these cases, the review no longer safeguards constitutional identity but undermines the legal order. The CJEU may intervene in such situations but is limited – under Article 4(2) and Article 4(3) TEU – to a review for abuse or arbitrariness, following a good faith standard.

14 Ultra vires review must also be understood within the broader context of the horizontal network of constitutional courts. It should not be viewed in isolation but in light of the shared constitutional traditions and judicial decisions of the Member States. This requires mutual consideration of rulings by other constitutional courts. Any departure from these decisions must be appropriately justified and take into account both national and EU legal perspectives. Such an approach ensures coherence in judicial reasoning and alignment with European values and legal principles.

15 In summary, ultra vires review should be understood as a shared responsibility within the European network of constitutional courts. Legally, Member State constitutional courts retain ultimate authority to deliver final judgments concerning their own legal systems. At the same time, the CJEU holds a “right of first access” based on its interpretative monopoly, enabling it to resolve the vast majority of jurisdictional disputes with legal certainty. However, this presupposes that the CJEU exercises a level of judicial scrutiny that genuinely safeguards the principle of conferral and, by extension, the (national) sovereignty of the Member States. This division of roles aligns with the institutional functions of the CJEU and the constitutional courts in the European constitutional court system. The CJEU, as the Union’s court, is responsible for interpreting and ensuring the effectiveness of EU law and is particularly well positioned to do so given its institutional proximity to Union-level issues. In contrast, the constitutional courts of the Member States are not tasked with reinterpreting EU law, but rather with safeguarding national constitutional identity. By observing this task, they also preserve the coherence of the EU’s heterogeneous and pluralistic constitutional value system.

16 The following thesis may serve as a forward-looking conclusion: The GCC’s PSPP judgment may well be remembered as the starting point for establishing effective competence control as a shared responsibility at the level of EU law. Just as the GCC’s Solange I case law was initially met with strong criticism but ultimately played a key role in developing robust European fundamental rights protection, a similar evolution may now be underway. The growing legal conviction among Member States’ constitutional courts – that they may invoke democracy-protecting constitutional reservations such as ultra vires review – is likely to motivate the CJEU to fulfil its institutional judicial role with greater rigour, particularly in complex jurisdictional matters at the intersection of the European composite legal order. In doing so, the CJEU can contribute meaningfully to enhancing legal certainty and legal confidence across the EU – two foundational elements of the rule of law and key drivers of European integration in a legal community. At the same time, it remains evident that if the CJEU consistently and credibly reviews the limits of competences under EU law, the constitutional courts of the Member States will have no cause to resort to ultra vires review.

Ultra vires review is not a relic of constitutional nationalism, but a structured expression of judicial responsibility in a pluralistic Union. When invoked under the four cumulative conditions identified in this part – (1) a manifest breach of conferred powers, (2) sincere judicial dialogue through the preliminary reference procedure, (3) a demonstrable failure of the CJEU to meet the standards of legal methodology, and (4) compliance with the democratic and rule-of-law principles of Article 2 TEU – it strengthens rather than weakens the authority of EU law. These requirements ensure that ultra vires review is not exercised arbitrarily but remains embedded in a framework of institutional cooperation, interpretative discipline, and mutual respect. Properly understood, ultra vires review complements the role of the CJEU by acting as a constitutional backstop: it upholds the foundational balance between Union effectiveness and Member State sovereignty. If the CJEU fulfils its institutional role with methodological clarity and judicial restraint, ultra vires review will remain what it was always meant to be – a principled exception, not a systemic threat.