Tuesday 14 June 2022

Breaching International Law in a General and Far-Reaching Way: The Northern Ireland Protocol Bill


Colin Murray, Reader in Public Law, Newcastle Law School

Photo credit: Dom0803, via Wikimedia Commons


In September 2020 Brandon Lewis set the tone for the UK’s interactions with the Northern Ireland Protocol by admitting that the Internal Market Bill, about to be introduced into Westminster, by admitting that the proposed legislation would breach the UK’s international commitments under the Withdrawal Agreement in a ‘very specific and limited way’. Ahead of the publication of the new Northern Ireland Protocol Bill, Lewis has been determined not to make the same mistake twice and has insisted that it ‘won’t break international law’. The Bill does, and we’re a long way beyond the specific and limited.


The Protocol and Brexit

The UK Government insists that the Protocol cannot continue in its current form as it does not command cross-community consent within Northern Ireland. That Brexit did not command cross community consent and the Protocol operates to mitigate some of the impacts of Brexit on Northern Ireland is written out of this analysis. The precise nature of the breach of the 1998 Agreement is, at best, vague. The UK Government has hitherto actively defended litigation against the Protocol on the basis that it is compliant with the 1998 Agreement. Cross-community consent, moreover, attaches under the Agreement and its implementing legislation to decisions within the Northern Ireland Assembly, not to international treaties concluded by the UK Government.

The UK Government’s other common refrain is that the Protocol it agreed is not the problem, but rather that the EU has been so dogmatic in the application of its trade and goods regulation provisions that it has become unworkable. This position, prominent in the legal position accompanying the new Bill, is unsustainable. The EU has repeatedly agreed reworks to the operation of the Protocol as concluded in 2019, in areas including state aid, medicine supply and in extending grace periods covering certain sectors. If the Protocol is more challenging to operate in practice than some had at first hoped, this is in large part the result to the limited nature of the Trade and Co-operation Agreement (and in particular the UK rejecting maintaining alignment with EU food and agriculture standards). The UK Government’s choices are entirely bound up in the supposed ‘peril that has emerged’ for Northern Ireland.

Brexit has brought with it inevitable dislocations for the Northern Ireland economy, and few dispute that the Protocol could be improved in its operation (a suite of EU reform proposals have been on the table since October 2021, and it has moved unilaterally to address highly contentious issues around medicines). But there is very little hard data in the public domain on the nature of the supposed divergences which have resulted from the Protocol, as opposed to Brexit itself, and even if there was, this would be a reason to trigger the trade protection provision within the Protocol, Article 16, rather than to seek to abandon particular obligations. It is untenable for the UK Government to invoke the doctrine of necessity, based on a legal position that it ‘has no other way of safeguarding the essential interests at stake than through the adoption of the legislative solution’, when its own conduct has contributed to the situation, and when it has made no effort to use the specific mechanisms available within the Protocol for addressing such societal concerns. 

Every time the possibility of triggering Article 16 has come up, since it was prominently raised in the July 2021 Command Paper, the UK Government has drawn back. The Owen Patterson affair, Party-gate and Putin’s invasion of Ukraine all undoubtedly disrupted this agenda. Equally, however, the limits of Article 16 loomed large; responses using this provision would themselves amount to a breach of the Protocol if they were not proportionate to substantive problems. Article 16 was not a magic wand capable of sweeping away inconvenient aspects of the Protocol. For a UK Government committed to aggressively diverging from EU standards the Protocol presents a pressing problem; every one of its own divergences will harden barriers to trade in the Irish sea. It would like to achieve far reaching Protocol reform on its own terms to avoid that outcome. Ministerial talk of ‘safeguarding’ the 1998 Agreement is nothing more than a superficial pretext.

The repeated failures to trigger Article 16 in the last year (involving presenting the data and crafting a targeted response within its terms) nonetheless left the Democratic Unionist Party (DUP) more exposed and incurred a debt to their thwarted expectations. The party, as a result, collapsed the Northern Ireland Executive in February and have been refusing to engage in any form of power-sharing since last month’s Assembly elections. It is now so distrustful of the UK Government’s position that they remain sceptical of reengaging with power-sharing processes in Northern Ireland even with the publication of the new legislation, even as the Foreign Secretary fumes that they should now ‘get on with it’.    


The Protocol Bill

The Bill, on its face, would appear to be everything that the DUP could ask for and more. But it is precisely because it is so far reaching, and so reliant on placing powers into the hands of ministers, that calls into question whether it can make it onto the statute book in its present form. At present, the EU law obligations which remain applicable to Northern Ireland under the Protocol flow directly into domestic law by the ‘conduit pipe’, to use the language of the UK Supreme Court in Miller, of section 7A of the European Union (Withdrawal) Act 2018. Clause 2 of the new Bill excludes a swathe of Protocol provisions from the scope of section 7A, cutting the pipe. This does not, of itself, absolve the UK of its international obligations. The EU can continue to take action against the UK for this breach of the Withdrawal Agreement. But it severs the connection between these Protocol obligations and domestic law.  

The Protocol provisions directly excluded from the operation of section 7A include all of its provisions relating to the movement of goods, including customs, (clause 4) and the regulation of goods (clause 8), state aid rules (clause 12) and the enforcement role of the Court of Justice of the European Union (clause 13). In each of these regards, ministers are given far-reaching powers to make new domestic law, enabling the UK Government to substitute its own scheme in place of the Protocol’s rules. Clause 22 confirms that ministers can make regulations under this Act to make any provision which could be made by an Act of Parliament. The supposed limitation to this power, repeated throughout the Bill, is that the regulations are such that the minister ‘considers appropriate’ in connection with the Protocol, the broader Withdrawal Agreement or this legislation. This is anything but a limited and specific breach of the UK’s international commitments; it utterly traduces the Protocol. Moreover, the powers granted to ministers are remarkably broad. Powers in the same terms faced the most strenuous opposition on grounds of side-lining Parliament when they were included in the Internal Market Bill as proposed, and there is no reason to think that the House of Lords, in particular, will be any more receptive to these powers in this context.

Clause 15 of the Bill purports to protect the operation of the Protocol’s provisions on human rights and equality (Article 2), the Common Travel Area (Article 3) and North-South Co-operation (Article 11). Ministers do not have the power to add these provisions to the Bill’s stated exclusions from the operation of section 7A. The Protocol’s human rights and equality commitments rely upon the operation of Court of Justice jurisprudence which explains how the relevant EU law functions. Thus, when clause 14 of the Bill sets out broad exclusions to domestic courts drawing, within the terms Article 13 of the Protocol, on Court of Justice jurisprudence or general principles of EU law, it does so only with regard to excluded provisions. This would appear to be an attempt to limit the number of fights the Bill starts, and to insulate the UK Government from accusations that it is undermining these high profile (but hitherto uncontroversial) arrangements. But it is irreconcilable with clause 20, which asserts (with no stated protection for Article 2) that courts and tribunals are not bound by ‘any principles laid down, or any decisions made, on or after the day on which this section comes into force by the European Court’. The overriding effect of this general exclusion is remove the courts in Northern Ireland obligation to interpret the provisions of the Protocol ‘in conformity’ with relevant Court of Justice case law. The Bill claims to protect the Protocol’s human rights and equality provisions in one clause, but undermines their substantive operation in another.  

The Bill’s explanatory notes contain a commitment that ‘the UK Government will write to the devolved administrations to seek consent to legislate in the normal manner’. This, of course, is a hollow commitment when Stormont is not functioning. And even if it was functioning, the track record of Brexit legislation establishes that the UK Government is willing to ignore the position of a clear majority of the Northern Ireland Assembly who have expressed their opposition to these terms. Clause 15, moreover, does not protect the operation of the Article 18 “Stormont lock” from being excluded from domestic law by ministers if there is a chance that majority support for the Protocol in the vote due in 2024 would embarrass the Government (not one of the permitted purposes, but they are so broadly drawn as to effectively allow for ministers to pursue that end under another pretext).

Which brings us to the negotiated settlement clause. Clause 19 acknowledges that a new agreement can be made between the UK and EU which modifies, supplements or replaces the Northern Ireland Protocol, in whole or part. In those circumstances ministers can make regulations to give effect to that Agreement. Furthermore, under clause 15, this can include turning back on the conduit pipe between the Protocol and domestic law. This whole spiralling crisis can be disappeared as swiftly as a bad dream in Dallas. But in UK constitutional terms, all of this is stage managed by the Government. Just as the UK Government has finally accepted more extensive trade treaty scrutiny arrangements in recent weeks, this Bill pointedly excepts any reworking of the Protocol from them. Parliament, under this Bill, will be giving ministers the power to immediately make regulations to give effect to such an Agreement without first jumping through processes like public consultations. 


The costs of breaking the Protocol

The EU has a range of options open in reaction to the Bill. It can institute infringement proceedings before the Court of Justice of the European Union in light of the provisions at issue and adopt its own targeted responses in retaliation for the breach. It can also seek to propose further measures to address the problems it has identified with the Protocol. The problem for the EU, however, and ultimately for the UK Government, is that Brussels is unlikely to want to be seen to facilitating this mode of relations by the UK.

The costs of this Bill, however, extend beyond the EU’s immediate reaction. Johnson might command a large Commons majority, but this is less reliable after the scale of last week’s vote of no confidence. The Conservative parliamentary party is likely to become even more fractious if much of its time over the coming months is spent in a fight over the Protocol when there are so many issues of urgent concern. A Lords majority are all but certain to use its powers to delay proposals for a full parliamentary session, especially in light of the consequences for the UK’s international standing of its flouting of such commitments and the possibility that, in seeking to placate Unionism, Northern Ireland’s other parties will become alienated from power sharing.

This generates a considerable presentation problem for Johnson; the Bill must appear sufficiently far reaching to assuage the DUP and sections of his own back benches, but not so extreme as to galvanise parliamentary opposition and to provoke a severe response from the EU. Hence the disconnect between the UK Government’s expressed position that the Protocol ‘is clearly undermining the Belfast Good Friday Agreement’ and the Prime Minister’s insistence that these new proposals amount to ‘a relatively trivial set of adjustments in the grand scheme of things’. The UK Government is breaching its international commitments, but it would rather other people say that than admit this reality.


Threats work?

Back in December 2020, the Internal Market Bill threat to part of the Protocol was withdrawn following an Agreement in the EU-UK Joint Committee on how the Protocol would be implemented. In Brexit lore, only such tough actions yield results from the EU. And yet, the alternative of careful technocratic negotiation and full data exchange on goods movements by the UK remains a counter-factual scenario. Even after months of talks, there isn’t a clear public account of why the UK Government is rejecting the EU’s October 2021 proposals outright.

Picking fights with the EU energises Brexit’s most ardent supporters and bringing about this crisis point might thus be a way for the UK Government to sell an eventual compromise (they pushed as hard as possible to get changes). For the EU, having seen this before, this is not a pattern of engagement that it wants to encourage. Not only will Brussels perceive the risk that the UK Government will keep coming back for more, but why should it bail out Johnson with further compromises when it knows that his position is so weakened and it might soon have to deal with a successor?

The EU therefore appears to be biding its time; this legislation faces a long journey through Parliament (particularly in light of the likelihood that it will not pass the Lords without a year’s delay). It appears ready to recommence infringement proceedings against the UK and even to publish its own blueprint for Protocol reform. Any advance on the “express” channel and food and agriculture approach set out in the October 2021 proposals will take it very much into the “green/red” lane territory that the UK Government is discussing; the fabled landing zone. Johnson’s entire strategy is reliant on the EU not using the full range of reactions open to it in response to this threat on the basis that doing so would further destabilise Northern Ireland.


A bumpy landing, if a landing at all?

The risks in this approach are obvious. Too many actors need to accept contradictory messages at one time. The EU needs to watch the UK tear up an Agreement in front of its nose, but trust that Johnson is only doing what he is doing to bring the DUP back on board (that this legislation is all a bit of theatre before a compromise). The DUP, by contrast, have to accept that legislation so outlandish in its terms that it has no chance of making the statute book in the next 18 months without being significantly watered down, provides a basis for re-entering power sharing. They must be so enthralled by its scope that they ignore its unfeasibility.

Johnson’s weakness, moreover, encourages everyone else to manoeuvre around him. When it came to the “Stormont lock”, the main concession that Johnson gained from EU in 2019, Brussels had held off on offering such an arrangement to Theresa May when her own premiership appeared to be on its last legs. The Commission may have announced that a further package of reform proposals are coming, but they would have to be completely on board with Johnson’s choreography (and very confident in his position) to announce these plans ahead of the forthcoming by-elections.

Maybe some of the tensions which have built up in Northern Ireland around the Protocol will dissipate as a result of this Bill. Perhaps another compromise will be reached without it ever becoming law. But if it isn’t, and this legislation enters force in its current form, the prospects for EU-UK relations, for the stability of Northern Ireland, and for the executive dominance of the UK’s constitutional order will be invidious. And even in the best-case scenario, the current UK Government’s mode of engagement with any issue that involves Europe is set. We’re just waiting for the next crisis. 


Thursday 9 June 2022

It’s the end – but the moment has been prepared for: the CJEU confirms that UK citizens have lost EU citizenship


Professor Steve Peers, University of Essex

In today’s judgment, the CJEU has finally confirmed that UK citizens lost EU citizenship as a consequence of Brexit, following the earlier Advocate-General’s opinion, which I discussed here. (There are other pending cases on this issue – see my compilation of Brexit litigation – but there’s no reason to think that the CJEU would decide them differently).

The judgment is striking for the extent to which it dismisses arguments that British citizens have retained EU citizenship. It’s definitely the end of an era. And yet, it also contains foreshadowing of issues that will be relevant to the post-Brexit future relationship between the UK and the EU.


Summary of the judgment

The case concerned a UK citizen resident in France, who lost the right to vote in local elections following Brexit day (1 February 2020) as a consequence of no longer being a national of a Member State, and therefore no longer having the right to vote in local elections. She challenged her removal from the electoral roll on the grounds that she could no longer vote in the UK (as she had been abroad more than 15 years), and a French court asked the CJEU questions about the interpretation of EU law and the validity of the EU Council decision to conclude the withdrawal agreement.  In the UK citizen’s view, the loss of EU citizenship status ‘cannot be an automatic consequence of the United Kingdom’s withdrawal from the European Union’, because ‘that loss infringes the principles of legal certainty and proportionality and also constitutes discrimination between Union citizens and an infringement of her freedom of movement’.

However, the Court began by observing that ‘citizenship of the Union requires possession of the nationality of a Member State’, by reference to the Treaty provisions which say just that. In the Court’s view, ‘the authors of the Treaties thus established an inseparable and exclusive link between possession of the nationality of a Member State and not only the acquisition, but also the retention, of the status of citizen of the Union’ (emphasis added). The Court placed its prior case law holding that EU citizenship ‘is destined to be the fundamental status of nationals of the Member States’ in that context. It also noted that in the various Treaty provisions on EU citizens’ rights, ‘none of those provisions enshrines that right in favour of nationals of a third State’.

It then agreed with the Commission’s argument that there was no exception from those who moved within the EU: the exercise of free movement rights is ‘consequently, not such as to enable him or her to retain the status of citizen of the Union and all the rights attached thereto by the FEU Treaty if, following the withdrawal of his or her State of origin from the European Union, he or she no longer holds the nationality of a Member State’.

Secondly, the Court reiterated prior case law that the UK chose to leave unilaterally, ‘in accordance with its own constitutional requirements’, and that the leaving Member State is not ‘required to take its decision in concert with the other Member States or with the EU institutions’: ‘[t]he decision to withdraw is for that Member State alone to take, in accordance with its constitutional requirements, and therefore depends solely on its sovereign choice’. Having left the EU in accordance with Article 50, the UK ‘is no longer, as from’ the date of entry into force of the withdrawal agreement, ‘a Member State’. It followed that ‘as from 1 February 2020, United Kingdom nationals no longer hold the nationality of a Member State, but that of a third State’. And ‘possession of the nationality of a Member State is an essential condition for a person to be able to acquire and retain the status of citizen of the Union and to benefit fully from the rights attaching to that status. The loss of nationality of a Member State therefore entails, for the person concerned, the automatic loss of his or her status as a citizen of the Union’ (emphasis added). Being ‘nationals of a third State as from 1 February 2020’, UK citizens ‘lost the status of a citizen of the Union as from that date’, so ‘no longer enjoy’ voting rights in local elections. ‘It is irrelevant, in that regard, that United Kingdom nationals have previously exercised their right to reside in a Member State’.

Was the loss of EU citizenship disproportionate? In the Court’s view, this was ‘an automatic consequence of the sole sovereign decision taken by the United Kingdom to withdraw from the European Union’; and the 15-year rule ‘is a choice of electoral law made by that former Member State, now a third State’. There was no need to carry out an ‘individual examination’ of the loss of citizenship: here the Court distinguished Brexit from its prior case law in which it ruled that loss of EU citizenship resulting from a Member State depriving a person of its nationality in individual cases needed to be justified – as distinct from (again) ‘the automatic result of a sovereign decision made by a former Member State, under Article 50(1) TEU, to withdraw from the European Union and thus to become a third State with respect to the European Union’. So ‘case-law arising from those various judgments cannot therefore be applied to a situation such as that in the main proceedings’.

The Court then dismissed a more technical argument that the withdrawal agreement must be interpreted to mean that voting rights in local elections for British citizens were retained. (In fact, the provisions on the transition period expressly excluded the application of this right). The Court observed that the argument being made ‘would create an asymmetry between the rights conferred by that agreement on United Kingdom nationals and Union citizens’, which it ruled was ‘contrary to the purpose of that agreement’, as the preamble stated the objective of ensuring ‘mutual protection for citizens of the Union and for United Kingdom nationals who exercised their respective rights of free movement before the end of the transition period’.

Similarly, the Court noted that after the end of the transition period, the citizens’ rights Part of the withdrawal agreement (discussed here) ‘lays down…rules designed to protect, on a reciprocal and equal basis, the situation of citizens of the Union and that of United Kingdom nationals…who exercised their rights to freedom of movement before the end of the transition period.’ It stated that ‘[t]he purpose of those rules, which apply…as from the end of the transition period, is…to ensure reciprocal protection for Union citizens and United Kingdom nationals….those rules concern the rights connected with residence, the rights of employed and self-employed persons, professional qualifications and the coordination of social security systems.’ But the citizens’ rights part did not retain voting rights after the end of the transition period either. The prohibition on discrimination on grounds of nationality in the withdrawal agreement only applied to citizens’ rights listed in the agreement – which did not include voting rights. Yet again, the Court noted that the deprivation of this right followed from ‘the United Kingdom’s sovereign decision to withdraw from the European Union’. Various rights in the EU Treaties equally did not help, because (again) UK citizens ‘are no longer nationals of a Member State’.

The Court did, however, note that Member States retain a right ‘to grant, under conditions which they lay down in their national law, a right to vote and to stand as a candidate to nationals of a third State residing in their territory’. (A number of Member States and the UK have indeed to some extent retained voting rights in local elections).

Next, the Court turned to questions about the validity of the Council’s decision to conclude the withdrawal agreement. It confirmed prior case law that it had the jurisdiction to examine whether treaties which the EU has concluded are ‘compatible with the Treaties and with the rules of international law which, in accordance with the Treaties, are binding on the Union’, although in that context it could only rule on whether the EU decision concluding the treaty was valid, not whether the treaty itself was valid. However, the Court ruled that arguments about voting in European Parliament elections, and EU citizens voting in the UK, were irrelevant – focussing on the issue litigated in the national court (UK citizens voting in the EU).

The Court dismissed all the arguments that the decision to conclude the withdrawal agreement was invalid. Again, various Articles of the EU Treaties were not relevant, because as Article 50 says, the Treaties have ‘ceased to apply to the United Kingdom from the date of entry into force of that agreement’, therefore UK nationals ‘no longer hold, as from that date, the nationality of a Member State but that of a third State. It follows that, as from that date, they are no longer citizens of the Union.’

As for the principle of proportionality, the Court took the view that:

there is nothing in the documents before the Court to suggest that the European Union, as a contracting party to the Withdrawal Agreement, exceeded the limits of its discretion in the conduct of external relations, by not requiring that, in that agreement in general or in Article 127 thereof in particular, a right to vote and to stand as a candidate in municipal elections in the Member State of residence be provided for United Kingdom nationals who exercised their right to reside in a Member State before the end of the transition period.

It reiterated prior case law that ‘the EU institutions enjoy broad discretion in policy decisions in the conduct of external relations’ and noted that:

In the exercise of their prerogatives in that area, those institutions may enter into international agreements based, inter alia, on the principle of reciprocity and mutual advantages. Thus, they are not required to grant, unilaterally, third-country nationals rights such as the right to vote and to stand as a candidate in municipal elections in the Member State of residence, which, moreover, is reserved solely to Union citizens….

Therefore the EU Council ‘cannot be criticised for’ concluding the withdrawal agreement without ensuring a continued right to vote for UK citizens in local EU elections.  And again, the loss of a voting right in the UK under the 15-year rule was a decision by the UK, not attributable to the EU.



The judgment is very similar to the Advocate-General’s opinion, although it does not include the point in the opinion that Member States chose not to confer on the EU the power to determine who can become an EU citizen. I don’t think this means it’s likely that the Court would rule, if asked somehow, that the EU does have such power; its judgment implicitly suggests there’s no such power (as the judgment confirms that EU citizenship is exclusively for nationals of Member States). In any event, there’s no sign that the EU institutions and Member States have any interest in unilaterally creating some form of EU consolation citizenship for Brits. The excited suggestion of one MEP to this end is not a source of EU law.

Nor does the judgment repeat the opinion’s observation that the applicant could simply obtain French citizenship, or the explicit conclusion that the applicant had no legitimate expectation to the retention of EU citizenship (although one might conclude that the Court implicitly rejects this when it attributes the responsibility for the consequences of withdrawal to the UK).

The centre-piece of the Court’s judgment is the simple linguistic interpretation of the Treaties: EU citizenship is conferred upon nationals of Member States; the UK is no longer a Member State; therefore UK nationals are no longer citizens of the Union. The Court does not explicitly address the argument that the Treaties only provide rules on the acquisition of EU citizenship, but not its loss; but it implicitly rejects them by ruling that the loss of EU citizenship follows from the withdrawal from the EU. In my view, this line of argument is unfounded because it imports words into the Treaties that don’t exist: there’s no wording in the definition of EU citizenship in the Treaties to suggest that this is solely a rule on how that citizenship is acquired. And it’s reasonable to deduce that there’s no need for an explicit provision on the loss of EU citizenship, because Article 50 covers that ground by necessary implication. Also, a key means of how EU citizenship is acquired in practice is by accession to the EU – which the withdrawal process mirrors. So the more logical comparison is between the collective, not the individual, acquisition and loss of EU citizenship.  

Also, the Court is keen to point out that the consequences of the loss of EU citizenship follow from the UK’s sovereign decision to leave the EU. This is consistent with prior CJEU case law (see Wightman and Shindler, discussed here and here). It again points out (and some Remainers seem unwilling to accept this) that the process of leaving the EU is up to the national law of the Member State choosing to leave, with no approval from the EU required (the notion that the EU ‘shouldn’t have let the UK leave’, or should somehow consider various facts – still less wild allegations – about the process of leaving, is obviously unfounded). In this context, it is worth remembering that some on the Leave side were prone to misleading or false comments about the impact for Brits of leaving the EU: for instance, this infamous tweet by the current Prime Minister (‘Brexit will make no difference’).

The Court does knock on the head the attempt to rely on the previous case law on the loss of EU citizenship (RottmannTjebbes and Wiener Landesregierung – see discussion here), because they all concern the loss of EU citizenship via means of loss of nationality of a Member State. In fact, those judgments reinforce, not contradict, the logic of today’s ruling – that there is an inextricable link between EU citizenship and nationality of a Member State.

Today’s judgment is the end of the assumption – strongly held by many – that EU citizenship is a status conferred upon individuals, not a status linked to a state’s membership of the European Union. There’s certainly an argument that EU citizenship should be based on the former approach, the wording of the Treaties (as the judgment confirms) points inescapably to the latter. Member States simply did not choose to create a status of EU citizenship independently from that of nationality of a Member State.

Finally, though, there are interesting nods to the post-Brexit future in the Court’s judgment. Although it does not mention the possible relevance of EU law on long-term resident non-EU citizens (on which, see here as regards its application to Brits, and here and here as regards proposals for reform), it goes out of its way to mention the citizens’ rights provisions of the withdrawal agreement, which now govern the position of those EU citizens who moved to the UK, and UK citizens who moved to the EU, before the end of the transition period. In this context (and other contexts), it mentions the importance of reciprocity in EU external relations – implicitly echoing the concerns of those who thought that arguments for Brits to retain EU citizenship, like some demands made by Leave supporters, sought to ‘have their cake and eat it too’. The Court also notes that the EU institutions have great political discretion in external relations. Time will tell whether, if UK/EU relations deteriorate still further, this might come to be seen as ‘the gun in the first act’.

Photo: Garry Knight, via Wikimedia commons


Sunday 5 June 2022

Greater Coherence for the Ne Bis in Idem Principle, at Last: The Bpost Judgment (C-117/20)



Victor Davio and Ilaria Gambardella*


*Victor is a PhD Researcher at the Institute for European Law, KU Leuven and Assistant in Constitutional Law at University Saint-Louis Brussels and Ilaria is Teaching Assistant and Research Associate at the Institute for European Law, KU Leuven. They are part of the RESHUFFLE research project hosted by KU Leuven under the direction of Prof. Elise Muir, and supported by the European Research Council (European Union’s Horizon 2020 research and innovation programme, grant agreement No 851621). 


Photo credit: Jean Housen, via Wikimedia commons





On 22 March 2022, the European Court of Justice (‘the ECJ’) delivered in Grand Chamber its judgment in the bpost case (C-117/20), which brings significant developments regarding the principle of ne bis in idem as enshrined in Article 50 of the EU Charter of Fundamental Rights (‘the Charter’). On the same day, the ECJ, also in Grand Chamber, rendered its judgment in the Nordzucker case (C151/20) which concerned the ne bis in idem principle in the field of competition law and in which the ECJ developed a similar approach to the bpost judgment in several respects. We will refer to the Nordzucker judgement to the extent that it is relevant for the analysis.


These judgments, and in particular the bpost judgment which will be discussed below, constitute a step forward towards greater coherence on ne bis in idem, a principle which has been deemed ‘among the most complex legal issues involving fundamental rights’. Also known as the prohibition of double jeopardy, the ne bis in idem principle has a double rationale: on the one hand, it is a rule of criminal procedure which aims to ensure legal certainty and a fair administration of criminal justice; on the other hand, it is an individual right which seeks to protect the individual from abuses of the ius puniendi. This principle is notably relevant in the context of the so-called dual proceedings, namely where a natural or legal person is subject to both criminal and administrative proceedings for the same misconduct and the latter proceedings are criminal within the meaning of the Charter or the European Convention of Human Rights (‘the ECHR’).


In the European landscape, the ne bis in idem principle has been characterised by a phenomenon of double fragmentation. The first is within EU law itself: the principle has been scattered in various legal instruments and has been subject to divergent interpretations by the ECJ depending on the area concerned, including between competition law and other areas of EU law. The second is between EU law and the ECHR: over the past years, the ECJ and the ECtHR have developed different methodologies for the application of the ne bis in idem principle, which places difficulties on national judges who are bound by both sets of rules.


The judgment commented upon provides a welcome, albeit partial, solution to this double fragmentation. Firstly, the ECJ departs from the dichotomy between competition law and other fields of EU law, by holding that the scope of protection of ne bis in idem is the same in all fields of EU law. Secondly, the ECJ aligns its case-law on the ECtHR case-law, and in particular on the Grand Chamber judgment A and B v. Norway of 15 November 2016. There is no perfect coherence between the ECJ and ECtHR case-law, however, since the CJEU confirmed its approach developed in the Menci judgment (C-524/15) which entails that the ne bis in idem principle must be submitted to the proportionality test under Art. 52(1) Charter, as we shall see below.


The facts of the case


Bpost, the incumbent postal services provider in Belgium, offers its services to two main categories of clients: the bulk mailers, which are end consumers, and the mail preparation firms, which are consolidators. As for 2010, bpost established a new tariff system for advertising and administrative mail items based on the ‘per sender’ model. Under the new system, the discounts granted to the consolidators are calculated on the volume of mail items deposited individually by each sender and no longer on the total amount of items from all senders for which the consolidators provided their services.


Due to this new tariff system, bpost was fined twice. First, the Belgian Postal Regulator imposed a fine for infringement of the non-discrimination rule related to tariffs in the postal sector. It considered that the new tariff system created an unjustified difference of treatment between the two main categories of clients. The decision was annulled by the Brussels Court of Appeal in a decision which became final. Second, the Belgian Competition Authority imposed on bpost a fine on the ground of abuse of dominant position. It estimated that the new tariff had an exclusionary effect on consolidators and potential competitors. Once again, the Brussels Court of Appeal annulled the decision on the ground that it was contrary to the ne bis in idem principle. After the decision was set aside by the Court of Cassation, the case came back to the Court of appeal which, therefore, submitted a request for a preliminary ruling to the ECJ.


The referring court stated that the two proceedings conducted by the Belgian Postal Regulator and the Belgian Competition Authority lead both to the imposition of administrative penalties of criminal nature. However, these penalties aimed to punish different offences: in one case, the infringement of non-discrimination in the postal sector and in the other case, the infringement of competition law. In the field of competition law, three criteria are required to verify whether the ‘idem’ condition is fulfilled, namely the same facts, the same offender and the same legal interest protected. However, the latter criterion is not applied in the case law relating to other fields of EU law. Therefore, the referring court asked the ECJ whether it had to rely on the case law on the application of ne bis in idem in the field of competition law (Toshiba (C-17/10)) or rather on the line of cases related to the Menci judgement.


The Opinion of Advocate General Bobek


The starting point for Advocate General Bobek’s Opinion is that the ECJ case-law on the ne bis in idem principle is ‘marked by fragmentation and partial incoherence’ (§3) and can ‘be described as a mosaic of parallel regimes’ (§85). In a pedagogical manner, the Advocate General paints a picture of this mosaic by discussing successively the protected legal interest in EU competition law (§43-52), ne bis in idem in the area of freedom, security and justice (§53-62), the case-law of the ECtHR (§63-74) and the developments in the Menci judgment (§75-84).


In his view, such a situation of fragmentation is untenable for three reasons. First, because the area of competition law cannot, per se and in the abstract, be considered different from other areas of EU law and therefore be subject to a different test (§92). Secondly, because of the unpredictability of the tests since there is no rule to determine which test applies when two of the areas of application of the ne bis in idem principle overlap in a case (§93). Thirdly, because the same provision of primary law, Art. 50 of the Charter, should not have a different content depending on the area of EU law to which it is applied (§95).


Should the test developed by the ECJ in Menci prevail across all EU areas? According to the Advocate General, this question requires a negative answer. In his opinion, the Menci judgment is a problematic decision in that it submits the ne bis in idem principle to the Article 52 Charter test and analyses this principle in the light of proportionality. He submits that: ‘Article 50 of the Charter is a bar. If validly triggered, it prevents the other proceedings from even starting. Such a bar must be defined ex ante and normatively’ (§107). The Advocate General thus favours a procedural version of the ne bis idem principle (which prevents ex ante the introduction of new proceedings) over a substantive version of this principle (which analyses ex post the proportionality of different sanctions under Article 52 of the Charter).


Then, the Advocate General argues in favour of the introduction of a threefold test for the assessment of the idem for the purposes of Article 50 of the Charter. This test would be based on a triple identity: an identity (1) of the offender, (2) of the relevant facts, and (3) of the protected legal interest (§133). What should be understood as a protected legal interest? In his view, it corresponds to ‘the societal good or social value that the given legislative framework or part thereof is intended to protect and uphold’ (§136). Insofar as the ‘bis’ condition (i.e. the existence of a prior final decision) and this triple identity of idem are met, the ne bis in idem principle would preclude the introduction of new proceedings.

In the bpost case, the Advocate General suggests that the triple identity, subject to verification by the referring court, has not been met, in particular in view of a difference in the protected legal interest. While the Belgian postal regulator fined bpost in order to avoid discrimination and ensure the transparency of the internal market for postal services, the Belgian Competition Authority sanctioned bpost’s anti-competitive practices (§160-162). The aim pursued by the two regulators was therefore different.


The judgment of the ECJ


The ECJ starts its reasoning by reminding that the ne bis in idem is a fundamental principle of EU law which is now enshrined in Article 50 of the Charter. The principle, which corresponds to that provided in Article 4 Protocol 7 ECHR, prohibits a duplication of proceedings or penalties of a criminal nature for the same acts and against the same person (§23).


The application of the ne bis in idem principle is subject to a twofold condition: (1) there must be a prior final decision as to the merits of the case (the ‘bis’ condition) and (2) the prior decision or the subsequent proceedings must concern the same facts (the ‘idem’ condition). The identity of material facts is the criterion to assess the existence of the same offence and is defined by the ECJ as ‘a set of concrete circumstances stemming from events which are, in essence, the same, in that they involve the same perpetrator and are inextricably linked together in time and space’ (§37). The ECJ adds that, for the purpose of establishing the existence of the same offence, the legal interest protected is not relevant, ‘in so far as the scope of the protection conferred by Article 50 cannot vary from one field of law to another’ (§34-35). Therefore, the ECJ explicitly rejects the condition of the legal interest protected which was traditionally only required in the field of competition law (Toshiba (C-17/10)). In doing so, the ECJ accepts the suggestion of Advocate General Bobek to harmonise its case law on the application of the ne bis in idem principle. The ECJ had been already invited in the past to depart from the different approach to the application of the ne bis in idem in competition law by both AG Kokott in Toshiba (§ 114-118) and AG Wahl in Powszechny (§45).


Subsequently, the ECJ clarifies that when the two conditions are met, the duplication of proceedings or penalties does not constitute a violation, but a limitation of the fundamental rights guaranteed by Article 50 of the Charter. On that point, the ECJ rejects the suggestion of the Advocate General to consider the ne bis in idem ‘as a bar’. Hence, a limitation to the ne bis in idem principle might be justified according to Article 52(1) of the Charter if the duplication of proceedings is aimed to protect distinct legitimate objectives and does not exceed what is appropriate and necessary in order to obtain these objectives. The ECJ affirms that ‘the the fact that two sets of proceedings are pursuing distinct objectives of general interest which it is legitimate to protect cumulatively can be taken into account, in an analysis of the proportionality of the duplication of proceedings and penalties, as a factor that would justify that duplication, provided that those proceedings are complementary and that the additional burden which that duplication represents can accordingly be justified by the two objectives pursued’ (§49). In this paragraph, as it will be stressed in the comments below, the Court heavily relies on the ECHR case-law (A and B v Norway).


Regarding the circumstances of the case, the ECJ recognises that the two proceedings pursue two distinct legitimate objectives which are, in one case, the liberalisation of the internal market for postal services and, in the other case, the protection of a fair competition on the market. The ECJ leaves to the national authorities to ascertain whether the duplication of proceedings was justified in the specific case. In conclusion, according to the ECJ, Article 50 Charter does not preclude the duplication of proceedings in circumstances such as those of the case, provided that a series of circumstances are respected, which is an evaluation that belongs to the referring Court.




A salient feature of the ECJ’s judgment in the bpost case is that it brings more coherence and consistency to the ne bis in idem principle, which was marked by its high degree of fragmentation. It does so, first and foremost, by harmonising its application within EU law itself, by holding that the scope of protection of the ne bis in idem principle enshrined in Article 50 of the Charter is the same in all areas of EU law, unless otherwise provided by EU law (see also Nordzucker, §40). In doing so, the ECJ set aside the dichotomy between competition law and other areas of EU law, the former having historically been the subject of separate case law as to the application of the ne bis in idem principle (Toshiba (C-17/10) and Slovak Telekom (C-857/19)).


Furthermore, the ECJ further develops a common understanding of the ne bis in idem principle at a pan-European level by aligning its approach on the ECtHR case-law. That the ECJ relies on the latter case-law may not come as a surprise in the light of Article 52(3) of the Charter, according to which the Charter rights shall have the same meaning and scope as those of the corresponding rights guaranteed by the ECHR. Since Article 50 corresponds to Article 4 of Protocol No 7 to the ECHR, its meaning and scope must be similar to those established by the ECHR. The explanations to the Charter indicate that the meaning and the scope of the guaranteed rights are determined not only by the text of the ECHR and its Protocols, but also by the ECtHR case-law.


Yet, what is singular about the bpost case is the extent to which the ECJ draws on the latter case-law. While the ECJ had referred to the ECtHR case-law in its Menci judgment (Menci, §61), the ECJ relies on the latter case-law in a more explicit and clear manner. Notably, the ECJ embraces the concept of ‘coherent whole’ developed in the ECtHR A and B v. Norway judgment. Repeating almost identically the wording of the latter’s judgment (A and B v. Norway, §121), the ECJ considers that: ‘public authorities can legitimately choose complementary legal responses to certain conduct that is harmful to society through different procedures forming a coherent whole so as to address different aspects of the social problem involved, provided that the accumulated legal responses do not represent an excessive burden for the individual concerned’ (bpost, §49). Furthermore, in order to determine whether the duplication of procedures and sanctions pursuing distinct legitimate aims is strictly necessary, the CJEU identifies a series of factors largely inspired by the case-law of the ECtHR, namely that (1) there are clear and precise rules making it possible to predict which acts or omissions are liable to be subject to a duplication of proceedings and penalties and that there will be coordination between the different authorities; that (2) the two sets of proceedings have been conducted in a manner that is sufficiently coordinated and within a proximate timeframe; and that (3) any penalty that may have been imposed in the proceedings that were first in time was taken into account in the assessment of the second penalty burden (bpost, §51; compare with A and B v. Norway, §132-134).


In spite of the increasing convergence between the case-law of the ECJ and the ECHR, some differences still persist regarding the scope of the principle and the methodology adopted by the two European Courts when they apply the ne bis in idem principle.


First, regarding the scope, in the ECHR context the ne bis in idem applies only within the jurisdiction of one Member State, while under EU law it also applies between several Member States. In other words, at EU level, the ne bis in idem principle has developed from a domestic to a transnational level. This is also attested by the Nordzucker case, which concerns two proceedings before two national competition authorities of different Member States.

Second, the two Courts still adopt a slightly different methodological approach to achieve the same result. On the one hand the ECHR only recognises the possibility of a duplication of proceedings when the two proceedings are complementary to each other, so that they can be considered ‘as a whole’, i.e. as one single set of proceedings. In principle, no exceptions and limitations to the principle are possible. On the other hand, in bpost the ECJ explicitly confirms that a duality of proceedings is accepted and refers to the concept of ‘coherent whole’.  However, the duplication of proceedings is considered by the ECJ as a limitation of the principle of ne bis in idem which is subject to the Article 52(1) Charter test and the fulfilment of a series of conditions that are shaped on the ECHR case-law.


Overall, the bpost judgment can be seen as a positive development for the ne bis in idem principle in Europe. It is an important step out of the ne bis idem maze, so to speak. It contributes to greater coherence and clarity regarding this principle, which has long been difficult to grasp for EU (fundamental rights) lawyers. In doing so, this judgment also eases the work of national courts which are at the intersection of at least three fundamental rights legal systems, namely the ECHR, EU fundamental rights and national fundamental rights, and whose role should be facilitated by the ECJ and the ECtHR in view of the increasing complexity of human rights in Europe.