Thursday, 14 December 2017

(Re)constructing the employment law hierarchy of norms: The Charter will not, should not and need not apply?

Niall O’Connor, Lecturer in Law, University of Essex

The Charter and Brexit

During the Brexit referendum campaign, it was Boris Johnson who led the way in deriding the Charter’s influence over British law. In the Telegraph article in which he first threw his support behind Brexit, Johnson wrote ‘[u]nder the 55-clause “Charter of Fundamental Human Rights”, including such peculiar entitlements as the right to found a school, or the right to “pursue a freely chosen occupation” anywhere in the EU, or the right to start a business (…) These are not fundamental rights as we normally understand them, and the mind boggles, as to how they will be enforced’.

For some eurosceptics, such as Johnson, the Charter is an unacceptably powerful weapon in the EU Court’s (CJEU) arsenal, capable of over-riding national sovereignty. Any law student could point out the inaccuracy of Johnson’s perception of the Charter. Indeed, anyone could highlight the inconsistency between raising fears of the Charter’s influence while simultaneously sneering at the difficulties associated with enforcing its admittedly rather nebulous provisions. Nonetheless, the mischaracterisation of the Charter and indeed the role of the CJEU more generally has gained traction and has, to an extent hamstrung the Brexit negotiations from the outset. Particular suspicion has been directed at the Charter’s Solidarity Title (Title IV).

The Solidarity Title

Title IV of the Charter contains a number of fundamental Employment Rights, including article 27 on information and consultation, article 28 which grants workers the right to collective bargaining, article 30 which provides for the right not to be dismissed unfairly and article 31 on the right to fair and just working conditions, Collectively, these rights will be referred to as the ‘Employment Rights’, that is to say those rights that have a particularly close connection to the contract of employment rather than the broader notion of ‘social’ or ‘labour’ rights. The Charter has been praised for its inclusion of social and economic rights alongside more traditional civil and political rights, but it has always been open to question whether the Charter’s Employment Rights are, or should be, considered human rights in the first place. 

Space precludes a more in-depth analysis of the human rights pedigree of social rights but the most obvious answer to this question is that the Charter’s Employment Rights are included in a fundamental human rights document and must therefore be human rights. (See for further detail, Virginia Mantouvalou, ‘Are Labour Rights Human Rights?’ (2012) 3 ELLJ 151.) This assertion is strengthened if we look to the Charter’s Explanations, which act as interpretative guidance. We can see that a number of the Employment Rights derive from earlier fundamental rights texts such as the European Social Charter (ESC) of the Council of Europe, the European Convention on Human Rights (ECHR), International Labour Organization (ILO) Conventions, the International Covenant on Economic, Social and Cultural Rights (ICESCR) and the UN Declaration of Human Rights (UNDHR). It would seem, then, that the Employment Rights have long been considered rights worthy of international protection.

In addition, many of the arguments levelled at the justiciability of social rights generally do not apply to the Charter’s Employment Rights specifically. First, none of the four Employment Rights necessarily entail any state expenditure and do not, therefore, involve the distribution of resources, although there are likely to be costs involved for employers (including in the public sector). Second, the four rights considered have largely been fleshed out in legislation and so in this respect cannot be criticised as being vague, abstract standards. In any event, article 1 of the Charter which protects human dignity is said in the Explanations to constitute ‘the real basis of fundamental rights’. It is clear that human dignity is the value underpinning all of the Charter’s rights, including the Employment Rights.

British antipathy to the Employment Rights has a long pedigree. It has been thought that the UK had achieved an opt-out from the Charter but this notion was dispelled by the courts (NS judgment). Catherine Barnard has suggested that there is only one true UK opt-out from the Charter. (‘The Opt-Out for the UK and Poland from the Charter of Fundamental Rights: Triumph of Rhetoric over Reality?’ in Stefan Griller and Jacques Ziller (eds), The Lisbon Treaty: EU Constitutionalism without a Constitutional Treaty (Springer 2008).) This is to be found in article 1(2) of Protocol 30 which provides that ‘[i]n particular, and for the avoidance of doubt, nothing in Title IV (…) creates justiciable right applicable to (…) the United Kingdom except in so far as (…) the United Kingdom has provided for such rights in its national law’. This provision serves the role of ‘making sure that if any of the provisions of Title IV are in fact classed as rights they are not justiciable in respect of the UK’.

In other words, if any of the provisions in Title IV are found to be ‘rights’ as opposed to ‘principles’ then they will not be directly justiciable in UK courts. To a large extent, even this provision may not have been necessary. Article 52(5) of the Charter already provides that the principles (which the Employment Rights are presumed to be) only lead to rights to the extent that they are implemented in EU or UK law. Article 52(1) further emphasises the rights/principles distinction, providing that rights must be ‘respected’, whereas principles must merely be ‘observed’. (See further Case C-176/12 AMS (discussed here), the opinion in Case C-282/10 Dominguez; and Case C-356/12 Glatzel).

The right has viewed the Employment Rights with suspicion, fearing that they would ‘provide the basis for a judicial assault upon the UK’s (neo-) liberal employment legislation’. (Michael Dougan, ‘The Treaty of Lisbon 2007: Winning Minds, Not Hearts’ (2008) CMLRev 617,666.) Labour lawyers and trade unionists, on the other hand, have tended to be more supportive, indeed the Labour party’s shadow Brexit minister, Keir Starmer continues to insist that the Charter should be preserved post-Brexit, with one of his six red lines for supporting the final deal being the defence of rights and preventing a race to the bottom. A consideration of the effects of the constitutionalisation of Employment Rights in the Charter shows that both views may have been somewhat misconceived.

The Consequences of Constitutionalisation

The most concrete expression of the constitutionalisation of the Employment Rights can be seen in their use as both a standard of review and as a tool of interpretation. Human rights as general principles have been used to interpret EU law from the earliest days of the Union’s existence (see Stauder). It is unsurprising, then, that it is in the interpretation of EU law that the Charter’s impact has been most keenly felt.

First, it should be noted that the CJEU has long adopted a purposive or teleological approach to interpreting EU employment legislation. A good example is the Working Time Directive (WTD). The purpose of that Directive is to lay down minimum requirements intended to improve living and working conditions. The CJEU has consistently held that the Directive must be interpreted broadly and purposively as any other interpretation would frustrate the objectives of the legislation (see Jaeger). Thus, the purposive approach was adopted long before the introduction of the Charter with its recognition in article 31 that limited working time and paid annual leave are fundamental human rights.

Since the Charter’s adoption, the CJEU continues to adopt this purposive approach, referring to the recitals and the objectives of the legislation but now simply adding the Charter into the teleological mix. (See my article, ‘Interpreting Employment Legislation through a Fundamental Rights Lens: What’s the Purpose?’ (2017) 8 ELLJ.) For example, in one recent judgment (ANGED), the CJEU starts by reaffirming that paid leave is a ‘particularly important principle of European social law’. It then notes, almost in passing, that the right is also found in the Charter. It then moves on to look at the objectives of the legislation to find that the Directive must be interpreted broadly. In other words, the effect of the inclusion of paid leave in the Charter is merely confirmatory of the conclusion the CJEU would have arrived at using existing purposive methods of interpretation.

Human rights as general principles of EU law have also long been used as a ground for reviewing not only the legality of EU acts, but also Member State acts falling within the scope of EU law. This role has now been taken up by the Charter. As there is no case law on the matter, the potential for the Employment Rights to act as a standard of review of EU legislation must be largely speculative, although the equality field provides a useful example of the Charter’s potential. In Test-Achats, articles 21 and 23 of the Charter on non-discrimination and equality were relied on to strike down EU legislation permitting difference of treatment between men and women in calculating insurance premiums. It might be that the granting of constitutional status to the Employment Rights may also have the effect of limiting the ability of the Union to adopt legislation derogating from those rights. For example, article 31 may prevent further derogation from the Working Time Directive. (In relation to article 27 see Peter Herzfeld Olsson, ‘Possible Shielding Effects of Article 27 on Worker’s Rights to Information and Consultation in the EU Charter of Fundamental Rights’ (2016) 32 International Journal of Comparative Labour Law and Industrial Relations 251.)

The Employment Rights have already been used as a standard of review of national legislation, although it is somewhat difficult to separate issues of review and interpretation. In many cases the CJEU is first asked to interpret EU legislation or the Charter before then considering the compatibility of national legislation with that interpretation. The CJEU does not have the competence to review national law directly.

Article 30 has been a weak standard of review of national law, with the CJEU usually finding that the issue is outside the scope of EU law and so the Charter is of no application (For instance, see Case C-117/14 Poclava). Article 27 has served litigants little better (see AMS, where the ECJ ruled that this Article had limited legal effect). Article 28 has been confined to considering whether rules set down in collective agreements could be reviewed for compatibility with EU law (Case C-297/10 Hennigs).

Article 31 is the Employment Right that has been most frequently invoked in the review of national legislation, although this may be a reflection of the fact that the WTD (which is stated in the Explanations to be a source of article 31) has been the subject of the most litigation. In King, the question for the CJEU was whether a worker who had been afforded a right to paid leave only part way through the employment relationship (if at all) lost that right if he did not take steps to invoke it. The UK Working Time Regulations stipulated that employees must take their paid annual leave in the relevant year or it is extinguished. AG Tanchev concluded that ‘in the light of the considerable normative weight of the right to paid annual under EU, international and Member State law, requiring a worker rather than an employer, to take steps to create an adequate facility for the exercise of paid annual leave would unlawfully make the existence of the right subject to a pre-condition’.

Overall, the influence of the Charter’s Employment Rights has represented an exercise in continuity. The extent of the Charter’s added value appears to be that the CJEU is more comfortable in relying on a written human rights text. To this extent, the CJEU has been emboldened in that its long-held approach to treating the provisions of employment legislation (most notably the concept of paid annual leave) as important social rights has now essentially been codified in the Charter, although there is a certain irony in a return to textualism to bolster a purposive approach. More democratically legitimate it may be, revolutionary it is not. (The same cannot be said of article 16 which provides for the freedom to conduct a business which has been used to radically disrupt existing approaches to the interpretation of the Transfer of Undertakings Directive. See Case C-426/11 Alemo-Herron.) Does employment law really have anything to lose, then, in the Brexit process?

The Effect on the Employment Law Hierarchy

Traditionally, the relationship between EU law and national law has not strictly been viewed as hierarchical. Rather, the interaction between the CJEU and domestic courts has been seen as one of cooperation rather than confrontation. With the enactment of the Charter, a new constitutional dimension has been added. Most civil law countries are used to conceiving of the employment relationship as consisting of a clear hierarchy of sources. This has not been true of the common law. It really makes no difference in what order the sources of labour law in the UK are discussed and the hierarchy at national level, to the extent that one can be said to exist, is capable of evolution or indeed inversion. It is useful, then, to bear in mind that there are currently a number of confused hierarchies in UK employment law (1) between EU law and domestic law and (2) within domestic law itself.

The EU-domestic hierarchy (1) can rather crudely be characterised as follows: (A) The Charter, as a constitutional human rights document sits at the pinnacle of the hierarchy of norms; (B) general EU law comes next as it must comply with the Charter but can also be used as a standard against which national law must comply; (C) this is followed by domestic law. The hierarchy at domestic level (2) (with a focus here on legislation and the common law) has never been clear and may now be in a double state of flux due to the Charter and Brexit. At domestic level (where the legislation is outside the scope of EU law), the classification of the Employment Rights as fundamental human rights in the Charter appears to have made very little difference.

The relationship between the common law and employment legislation has always been intimate. We need only think of the fact that access to protective legislation usually depends on classification as a worker or an employee, the tests for which derive from the common law. The common law has at times, therefore, had the effect of impeding access to employee-protective measures (Tanton [1999] EWCA Civ 949). By and large, this has been an entirely domestic matter, with little consideration given to the human rights nature of legislation outside the scope of EU law.

Article 30 for example, provides that the right not to be unfairly dismissed is a human right, yet the EU has not adopted comprehensive legislation governing this right. This has meant that in the context of unfair dismissal, the UK courts have been free to adopt a largely employer-friendly approach. Indeed, contrary to the CJEU’s purposive approach in the employment field, common law judges tend to exaggerate contract law tests when applied to employment legislation. (Steven Anderman, ‘The Interpretation of Protective Employment Statutes and Contracts of Employment’ (2000) 29 ILJ 223.)
In certain circumstances, however, the common law has been malleable to legislative intervention, leading to an absorption of social rights standards. A particularly good example is the common law implied term of mutual trust and confidence. It was the need to come to a definition of the constructive dismissal concept that led the courts to develop this implied term in order to modify the notion of repudiatory breach found in commercial contracts (Malik [1997] UKHL 23).

Overall then, the fact that certain pieces of domestic legislation (outside the scope of EU law) have fundamental rights implications has largely been irrelevant to the domestic hierarchy of norms. Sometimes the common law impedes social legislation, sometimes it facilitates it. What, then, are the implications of Brexit for these already unstable hierarchies?

The EU Withdrawal Bill

The UK Government has made it clear that, in its opinion, there can be no real Brexit without removing the UK from the somewhat ambiguously termed ‘direct jurisdiction’ of the CJEU although there are currently ongoing attempts from both the Labour party and a number of Conservative backbenchers to ensure that the Charter is codified in UK law. Just two weeks ago, the Government was forced to postpone consideration of the Charter’s future role. In any event, it will fall on the UK courts to take on the full range of tasks associated with the interpretation and application of (former) EU employment legislation. Clause 2(1) of the Bill provides that EU-derived legislation applicable before Brexit will continue to have effect in UK law. Clause 5(1) makes clear, however, that the principle of supremacy of EU law will no longer strictly apply, although it will, according to clause 5(2) continue to govern the ‘interpretation, disapplication or quashing of any enactment or rule of law passed or made before exit day’.

This may have continued significance for the ability of litigants to enforce their EU-derived employment rights. In Benkharbouche, workers at the embassies of Sudan and Libya were found to be entitled to have their EU-derived employment law (discrimination and working time) claims heard in a UK Employment Tribunal despite the presence of an apparent conflict with the State Immunity Act 1978. Preventing reliance on these EU rights would breach article 47 of the Charter, which guarantees access to justice. (The Charter issues were discussed more fully in the Court of Appeal ruling, discussed here).

This judgment should also remind us of the protection that will continue to be provided by article 6 of the ECHR, the Convention right to a fair trial. Lord Sumption held that ‘a conflict between EU law and English domestic law must be resolved in favour of the former, and the latter must be disapplied; whereas the remedy in the case of inconsistency with article 6 of the Human Rights Convention is a declaration of incompatibility.’ (para 78)

Echoes of this judgment could also be seen in Unison, in which the Supreme Court quashed the order introducing Employment Tribunal Fees. That judgment starts by noting that many of the rights which are enforceable before Employment Tribunals are EU-derived rights which thus engages article 47 of the Charter. That judgment is also significant for the Court’s exploration of alternative avenues to human rights protection, notably the right to access justice at common law.

The effect of clause 5(2) is to create a new category of ‘retained EU law’ that must, somehow, fit into the existing hierarchy of norms. For employment law purposes, this may not an immediate issue. Much of the existing EU employment acquis has already been implemented in domestic legislation. However, there still remain serious doubts as to the precise status of post-Brexit CJEU case law (clause 6 retains that court’s pre-Brexit case law, with more flexibility to amend that case law) and whether this can be applied or departed from by the domestic courts. Lady Hale, the newly appointed President of the Supreme Court has called for clarification in this area. Most importantly, for our purposes, the Bill is clear that the Charter will not apply.

The Charter Will Not, Should Not and Need Not Apply?

Will Not?

Clause 5(4) of the Bill provides that ‘the Charter of Fundamental Rights is not part of domestic law on or after exit day’. This presents some major practical difficulties. First, it is often impossible to pinpoint precisely the influence of the Charter in CJEU decisions. Sometimes, the Charter is front and centre in employment law decisions. In others, it is barely mentioned (if at all). In earlier cases, the CJEU may simply have been reticent in its use of the Charter, recognising the sensitive nature of social rights and grappling (as it continues to do) with the distinction between rights and principles.

 Despite the somewhat sceptical view of the Charter’s influence in this field of law outlined above, it may well be that CJEU is simply not being explicit in its use of the Charter, relying instead on existing interpretative methods, but with an eye to Charter for guidance. In any case, the Charter and employment legislation enjoy a symbiotic relationship. It has been noted that the Explanations act as interpretative guidance. The explanations to the Employment Rights refer to existing EU employment legislation. Therefore, employment legislation must be interpreted in light of the Charter which must itself be interpreted in light of that very legislation. Decoding this cycle may prove an impossible task.

Should Not?

It has been argued that incorporating the Charter into domestic law post-Brexit would be undemocratic. This is because the interpretation of the Charter rights is entirely a matter of the CJEU. Eduardo Gill-Pedro argues in a recent blog that the CJEU interprets the Charter in the light of the objectives of the Union. Once the UK leaves the Union it will no longer share those objectives and so it would be undemocratic to rely on the Charter.

I would argue that the Charter does not merely serve as an overarching political guide to the future direction of the Union. Certainly, in the employment context it has a less ambitious remit, steering as it does, the interpretation and review of employment legislation. If the Government is committed to preserving the existing employment law acquis it should have nothing to fear from retaining the Charter. In any case, stripped of its constitutional status into the future, the Charter may prove little threat to the amendment or repeal of domestic employment legislation save to the extent that the supremacy principle continues to apply on a limited basis. There would, therefore, be no ‘intrusion’ of external values into the legislative process.

Need not?

Finally, it could be argued that whether the Charter applies or not really makes very little practical difference. First, as was noted at the start of this blog, the effect of the Charter’s Employment Rights, largely dismissed as mere ‘principles’ has been somewhat disappointing. Second, and more significantly, the Repeal Bill itself which at clause 5(5) preserves fundamental rights that exist autonomously of the Charter and ‘references to the Charter in any case law are, so far as necessary for this purpose, to be read as if they were references to any corresponding retained fundamental rights or principles’.

It is clear, then, the Charter will continue to play a role in the guise of those provisions that are already reflected in the general principles of EU law (although Schedule 1 to the Withdrawal Bill will limit the legal effect of those general principles). In many areas, but notably the equality field, it was the general principles that opened the way to the application of fundamental rights in employment law, although I accept that the adoption of the Charter had an emboldening effect even prior to its granting of full legal effect. (Compare Case C-144/04 Mangold and Case C-555/07 Kücükdeveci) Although, admittedly the status of the Employment Rights as general principles remains unclear and the Withdrawal Bill makes it clear in any event that the general principles cannot act as a standard of review for retained EU law.

Separating the role of the Charter and the general principles will in any case be difficult. We need only look to Norway to see how the general principles can have an effect on those not formally bound by the Charter. The European Free Trade Association (EFTA) Court has from the outset referred to judgments of the CJEU. The ‘homogeneity’ principle has had a profound impact, leading the EFTA Court to apply CJEU decisions that post-date the EEA Agreement. (Carl Baudenbacher, ‘The Relationship Between the EFTA Court and the Court of Justice of the European Union’ in Carl Baudenbacher (ed), The Handbook on EEA Law (Springer 2016) 179, 184.)

This principle governs the relationship between the European Economic Area (EEA) Agreement and EU law. Article 6 of the EEA Agreement provides that ‘[w]ithout prejudice to future developments of case-law, the provisions of this Agreement, in so far as they are identical in substance to corresponding rules of [the EU Treaties] and to acts adopted in application of [those Treaties], shall, in their implementation and application, be interpreted in conformity with the relevant rulings of the [CJEU] given prior to the date of signature of this Agreement’. The EFTA Court has held that ‘the objective of establishing a dynamic and homogenous European Economic Area can only be achieved if EFTA and EU citizens, as well as economic operators enjoy, relying on EEA law, the same rights in both the EU and EFTA pillars of the EEA’. (Case E-18/11 Irish Bank Resolution Corporation v Kaupping para 122.) The homogeneity principle extends to fundamental rights, with the Court referring to judgments of the ECtHR and CJEU as well as AG Opinions in fundamental rights cases (Case E-2/03 Asgeirsson and Others para 23; Case E-8/97 TV 1000 para 26).

The instincts of both labour lawyers and Charter sceptics may have been wrong. Far from representing a Trojan horse, bringing in its wake alien social rights to disrupt the common law’s traditional flexibility, the Charter’s Solidarity Title has proven to be rather a damp squib, except of course that it has actually had the effect of promoting the development of a countervailing business freedom found in article 16. In AGET Iraklis, for example, the CJEU held that article 16 is closely related to the four economic freedoms contained in the EU Treaties and can be used to defeat competing employment rights. (See also Alemo-Herron)

And yet, it is possible that we labour lawyers simply expected too much from a document that was explicitly adopted to codify the existing, cautious and piecemeal approach to fundamental employment rights. And yet still, it is possible to be too sceptical. The reality is that the Charter’s Employment Rights do have the potential to act as a bulwark against legislation that might undermine the rights of workers – in particular against any attempt to lower the standard of protection for employment rights below the level established by retained EU law.

The Charter’s use as a standard of review is perhaps its most powerful function and yet, at least in the employment field, this function has yet to be fully tested. It may be that we are leaving the Charter party before it has even got started. There is no doubt that the hierarchy of employment law norms is now in a state of flux and we do not know where the pieces may fall. What we do know is that, if the UK government gets its way, the Charter will not be among them. Without the Charter, the human rights landscape in the employment field will be somewhat impoverished. Left at the mercy of the common law, it is likely that UK employment legislation will travel in an altogether more deregulatory direction.

Barnard and Peers: chapter 9, chapter 20

Photo credit: Sky News

The Security of the Status of Long-Term Non-EU Residents in the EU: Some Thoughts on Case C-636/16 López Pastuzano

Dr Diego Acosta, Reader in European and Migration Law at the University of Bristol; author of The Long-Term Residence Directive as Subsidiary Form of EU Citizenship.An Analysis of Directive 2003/109 (Brill, 2011).

When can Member States expel a third-country national (TCN) holding a long-term residence (LTR) permit and having committed a criminal offence? The ECJ engaged with this important issue in its López Pastuzano ruling on 7 December 2017. It must be recalled at the outset that, according Eurostat data, there are more than 7 million LTRs residing in the EU, with that number set to rise after Brexit. Having participated in the drafting of the written observations on behalf of the claimant before the Court, this post is a short explanation of the case and of its implications.


Directive 2003/109 (which was amended in 2011 to extend to refugees and people with subsidiary protection) provides TCNs who, as well as fulfilling other conditions, have regularly and continuously resided for five years in a Member State with a LTR status. (The UK, Ireland and Denmark opted out of the Directive) The advantages of holding such status, as opposed to a mere temporary permit, can be summarised as accessing equal treatment with nationals in a number of areas (Article 11), the conditional right to reside in another Member State (Articles 14 and 15) and the reinforced protection against expulsion (Article 12).

The ECJ has repeatedly held that the main objective of the Directive is the integration of TCNs who are settled on a long-term basis in the Member States (Cases C-502/10, Singh para 45; C-508/10, Commission v Netherlands para 66; C‑571/10, Kamberaj, para 90). Security of residence is an essential condition for that aim (in real life and in the perception of the Union legislator). If that security would be affected, the main purpose of the Directive would be undermined.

The case at hand dealt with Article 12, which fleshes out the reinforced protection against expulsion by establishing a twofold test. Firstly, Member States may expel a long-term resident “solely” when the individual constitutes an actual and sufficiently serious threat to public policy or public security (Article 12(1)); such a decision “cannot be founded on economic considerations” (Article 12(2)). Secondly, before making such decision, Member States “shall have regard to the following factors: (a) the duration of residence in their territory; (b) the age of the person concerned; (c) the consequences for the person concerned and family members; (d) links with the country of residence or the absence of links with the country of origin” (Article 12(3)).

The facts in the case

The case revolved around the implementation of Article 12 of the Directive in Spain. The claimant, a TCN holding a LTR permit, had resided in Spain since 2008, was working and in a relationship with a Spanish national. His mother and two brothers also resided in Spain. In 2014, he was condemned to a sentence of 12 and 3 months, respectively, following two criminal offences. By virtue of the Spanish Organic Law on Foreigners, in particular its Article 57(2), a third-country national can be expelled if he or she has been condemned for an offence sanctioned by prison of more than a year.

In such scenarios, the authorities cannot take into account aspects comparable to those mentioned in Article 12 (3) of the Directive. These safeguards are taken into consideration in other cases of expulsion but not in case of a prison sentence of more than one year. This national rule applies irrespective of whether the individual TCN holds a LTR permit or not.

The Ruling by the ECJ

The ECJ reminds at the outset that the main objective of the Directive is the integration of TCNs holding a LTR permit, for which they enjoy reinforced protection against expulsion (paras 23-24). The Court reiterates that it is solely when the individual constitutes an actual and sufficiently serious threat to public policy or public security that an expulsion measure might be imposed, but only after having considered each of the aspects in Article 12 (3) of the Directive (paras 25-26).

Indeed, the Directive prohibits the automatic application of an expulsion measure following a condemnatory sentence for a criminal offence, but rather it requires a case by case analysis on the elements mentioned, in particular, in Article 12 (3) (para 27). The words “in particular” are crucial here since it means that the list of elements to be assessed in each individual case in Article 12(3) is not exhaustive and that Member States must also consider other elements. This is the type of personalised proportionality assessment that the ECJ has also favoured in other recent rulings (E.g. Case C-579/13 P and S, discussed here, which concerned integration requirements for long-term residents) and that Member States need to conduct in each single case so as to fulfil their obligations under EU law. Moreover, the ECJ makes reference to its previous ruling in Ziebell (para 27) as discussed below. Spain is consequently found to be in breach of the Directive (para 29).


The ECJ did not engage in a deep analysis of which LTRs can be considered to be an actual and sufficiently serious threat to public policy or public security to begin with. However, its reference to the Ziebell case is central (para 27). In Ziebell, the Court interpreted Article 12 of the Directive on the basis of a comprehensive summary of its case law on Article 14 of EEC-Turkey Association Council Decision 1/80 (which concerns the ‘public policy and public security’ exception to the rules on residence of Turkish workers and their family members).

In brief, the same interpretation of the concept of public policy as in the area of EU nationals applies by analogy; since this is a derogation on a right it needs to be interpreted strictly; measures on grounds of public policy may be taken only following a case-by-case assessment by the competent national authorities showing that the personal conduct of the individual concerned constitutes at present a genuine and sufficiently serious threat to a fundamental interest of society.

In addition to that, the principles of proportionality and the respect for the fundamental rights of the individual, in particular, the right to privacy and family life, must be respected, and “such measures cannot be ordered automatically on general preventive grounds following a criminal conviction or as a means of deterring other foreign nationals from committing offences”. Finally, the existence of “previous criminal convictions is, in itself, irrelevant for justifying an expulsion” and “the same must hold all the more true for a justification relating to the duration of any prison terms to which the individual concerned was sentenced.”

In line with this, national courts and national authorities must take into consideration “factual matters which occurred after the final decision of the competent authorities which may point to the cessation or the substantial diminution of the present threat which the conduct of the person concerned constitutes to the requirements of the fundamental interest in question (Case C-371/08, Ziebell paras 81-84).

This protection is very similar to the one that EU nationals enjoy under Article 27 of Directive 2004/38 (the EU citizens’ Directive), if not the same. As the ECJ established in the H.T. case (C‑373/13), relating to refugee law (discussed here), whilst EU Member States “retain the freedom to determine the requirements of public policy and public security in accordance with their national needs, which can vary from one Member State to another and from one era to another … the extent of the protection a society intends to afford to its fundamental interests cannot vary depending on the legal status of the person that undermines those interests” (para 77). This has important implications for all domestic court deciding in cases where a LTR might have committed a criminal offence and when the particular application of the personalised proportionality assessment will be central in deciding the outcome of the residence of the individual or his or her expulsion.

Barnard & Peers: chapter 26
JHA4: chapter I:6

Photo credit:

Saturday, 9 December 2017

The Beginning of the End? Citizens’ rights in the Brexit ‘Sufficient Progress’ deal

Professor Steve Peers

Yesterday’s ‘joint report’ on the progress in the Brexit negotiations (accompanied by a Commission paper and a joint technical note on EU27 and UK citizens’ rights) amounts to an informal deal to proceed to the second phase of Brexit talks. That second phase will focus on a transition period (discussed here; and see the draft EU negotiation guidelines here) and the framework for the post-Brexit relationship between the UK and the EU.

But for now the joint report is the main issue. It’s a political document, not a legal one, but it’s practically very relevant to the formal legal process of drawing up the UK’s withdrawal agreement from the EU, as it sets out many agreed details concerning key parts of that agreement: citizens’ rights, Irish border issues, the financial settlement, and ‘winding up’ rules. Some points still need to be agreed or fleshed out; the formal legal text of the withdrawal agreement must still be drafted; and any agreement on what happens in the transition period will have to be added to the withdrawal agreement. Yet yesterday’s deal is undeniably a milestone, making it rather more likely that a final complete withdrawal agreement will be agreed.

This blog post focusses on the citizens’ rights points in the deal; I hope to return to examine the other issues in more detail soon.

Scope of the agreement

The first point to note is that citizens’ rights are reciprocal: they will cover both EU27 citizens in the UK and UK citizens in the EU27. Much of the public discussion of this issue focusses on the former but overlooks the latter; and it is arguable that as result the joint report overlooks them somewhat too. The UK has come some distance towards the EU27 position on the issues (see discussion of the parties’ earlier positions here) but there are still some elements of compromise.

The joint report indicates that the parties’ agreement does not cover all of their citizens, but only those who have ‘exercised free movement rights by the specified date’. That date is agreed to be Brexit Day (the UK had originally suggested the date of sending the Article 50 notice, but always indicated its willingness to negotiate this point). So the withdrawal agreement will apply to those resident on that date and also (implicitly) those who had previously been resident but departed briefly from the country they were living in, in accordance with EU free movement law.

More precisely, the personal scope of the agreement will be those who reside legally on the territory by Brexit Day (raising the question of how to define ‘legally’), and their family members who are defined by the EU citizens’ Directive. Those working as frontier workers on Brexit Day (as defined in EU law) are also covered. At first sight, this implicitly rules out family members who are instead covered by the free movement of workers Regulation (see, for instance, the Czop and Punakova case), as well as those who returned to a country after leaving it (so-called Surinder Singh cases; see discussion of the relevant case law here), and dual citizens, whose position is (like Surinder Singh cases) defined by analogy with the citizens’ Directive (see discussion of the relevant recent ECJ ruling here). However, the joint technical note refers more generally to those who have moved in accordance with the Treaties – suggesting instead that the final withdrawal agreement will cover such people.

On the other hand, there’s nothing to suggest that non-EU family members of British citizens who live in the UK, who are covered by the ECJ’s Zambrano case law will be covered (for further explanation of this category of people, see the discussion of the most recent case law here).

Several issues are explicitly left to later negotiation, according to the joint technical note: the further free movement rights of UK citizens living in the EU27; the recognition of post-Brexit qualifications; future healthcare arrangements (such as the EHIC card for UK citizens who visit the EU after Brexit, and vice versa); lawyers practicing under home state title; and posted workers.

Family reunion

For those who seek to be joined by family members after Brexit Day, there are limitations compared to the current rules. The current rules on admission of core family members (spouse, registered partner, children, dependent parents et al) will apply to those family members who are related to the relevant EU27 or UK nationals on Brexit Day already. For those who were not related on Brexit Day, national law will apply. This will usually be less generous, either on the UK or the EU27 side (the joint report fails to mention that 25 of the EU27 – all except Ireland and Denmark – are bound by an EU Directive on family reunion, which sets minimum standards for the admission of family members of non-EU citizens). As an exception, however, such national law will not apply to children born or adopted after Brexit Day. In short, anyone whose possible future family member’s status might be affected by the change in rules should probably start shopping for wedding rings – but can keep buying condoms if they want to.   

There’s another change relating to extended family members. Under the EU citizens’ Directive, Member States must facilitate the entry of EU citizens’ non-registered partners, as well as ‘any other family members’ who are dependents, members or the household, or being cared for by the EU national sponsor. As the ECJ clarified in Rahman, this isn’t an obligation to admit all such extended family members, but to consider applications for their admissions and justify any refusal to admit. But the joint report states that even this limited right will be curtailed, applying to partners only, not the other family members referred to. The partnership must also exist and be durable on Brexit Day; again partnerships which only start – or only become durable – after that date won’t be covered.

So the agreement will definitely lower standards that currently apply to family reunion for the persons concerned, in both the UK and the EU27. For instance, the UK has strict income requirements for the sponsors of family members, which will apply to EU27 citizens in the UK after Brexit Day, if they only married their spouse after that date. (Note that these rules will apply regardless of whether the incoming spouse is non-EU or an EU citizen; and they will also make it harder for many UK citizens to bring an EU spouse to this country). On the EU27 side, ECJ case law prohibits a high income threshold for family reunion (see the Chakroun judgment), but waiting periods before admission are possible. Discrimination on grounds of nationality will be banned, but in this context this means migrants will be treated equally badly to nationals.

Residence rights

Next, the joint report sets out agreed rules on residence status. The parties may choose to require UK or EU27 citizens respectively to apply for a new residence status under national law. This is an implicit reference to the UK intention to require all EU citizens to apply for ‘settled status’ after Brexit, although the EU27 may, if they choose, impose a parallel requirement on resident UK citizens. There are detailed rules on simplifying the application process, which in principle apply to both sides but are actually focussed on the UK side – as confirmed by the footnote referring to the recent UK technical paper on applying EU27 citizens’ rights after Brexit. The risk here is that the practical difficulties which UK citizens in the EU27 might face in transferring to national immigration status in those Member States is being ignored. And again, the EU law on non-EU citizens which is relevant for some concerned – the long-term residents’ Directive – goes unmentioned here.

In fact, I recently met a UK civil servant who admitted that the UK side is not interested in negotiating about such details, despite the UK government’s public expressions of concern for UK citizens in the EU27. The awkward fact here is that, due to the inherent reciprocity in this aspect of the talks, the UK government could not be an effective advocate for retaining UK citizens’ rights in the EU27 – because of its primary interest was in curtailing rights of EU27 citizens in the UK. For instance, it was effectively the UK government which pushed to reduce the future family reunion rights of UK citizens in the EU27, because of its desire to apply restrictive UK immigration law to family reunion for EU27 citizens in the UK in future.

The only solution here for UK citizens in the EU27 is to push for the EU side to remember their rights – either by reopening this part of the withdrawal agreement or at least by pushing for soft or hard law to be agreed within the EU27 side which clarifies their position and ensures their rights in more detail in EU Member States after Brexit.

More fundamentally, the EU27 side has conceded to the UK on the very idea of transferring to national status. There’s a grace period of two years to apply for such national status – but what happens to people who don’t apply in time? A genuine notion of ‘acquired rights’ would mean, at the very least, citizens retaining exactly the same status they had on Brexit Day; this might also extend to continuing to acquire status that was in the process of acquisition on Brexit Day.  But instead the joint report allows parties to insist on a transfer to a national status – at the UK’s behest. This (self-)imposes a significant administrative burden upon the UK, and the recent error rate of Home Office officials in dealing with EU citizens and non-EU citizens alike does not inspire confidence. And, as noted already, parallel concerns may arise wherever UK citizens are required to transfer to a national status in EU27 states.

The prospect of transferring status is moreover restricted for some by the concession to the UK of the possibility of removing people deemed to be abusing rights even before they have completed judicial redress procedures, as a derogation from existing protection. It should be noted that the ECJ has already ruled (in the 2014 McCarthy judgment, discussed here) that the ‘abuse of rights’ notion cannot be applied to EU citizens’ family members as easily as the UK government would like. But this still leaves the UK government leeway to argue that others, such as homeless EU27 citizens, are covered by the concept and so have less judicial protection once the withdrawal agreement applies.

Substantively, the joint report says that the current rules in the EU citizens’ directive will apply to residence rights and permanent residence. This leaves open the possibility of refusing permanent residence due to not having ‘comprehensive sickness insurance’ (CSI) – which in the UK government’s view means that many EU citizens who are stay-at-home parents or carers fail to qualify, since NHS coverage is not enough. Although the UK government has promised to waive this requirement unilaterally, the joint report implicitly accepts that this will not be legally binding as part of the withdrawal agreement, instead being simply the exercise of the option to apply more favourable rules than the Directive requires. The future status of these vulnerable people could therefore be changed at the whim of the UK government.

Those who already have a documented form of permanent residence will get the new national status free of charge, subject only to a security and criminality check, and verification of identity and residence. Implicitly those who are entitled to permanent residence but who do not have a document to prove it as of Brexit Day will not be protected by such guarantees. Nor will those whom the UK deems not entitled to permanent residence yet – such as the vulnerable people who don’t have CSI as interpreted by the government. It’s not clear what guarantees will apply to these people instead. 
The systematic checks on criminality would be prohibited under the citizens’ Directive, but will be allowed under the withdrawal agreement. Moreover, the substantive threshold for refusing status will change: conduct after Brexit Day will lead to immigration law consequences in accordance with national law, rather than EU free movement law.  In some cases, this will mean that offences which would not lead to denial of status under free movement law will lead to loss of status under national law.

Permanent residence rights obtained under the withdrawal agreement will, however, be retained for up to five years’ departure – rather than two years’ departure under the EU citizens’ Directive. Of course, this is in the context of the loss of the underlying free movement rights – which were previously available as a backup if the right to permanent residence was lost.

Other rights

The joint report states that UK and EU27 citizens retain rights under EU social security legislation – including the EHIC health insurance card – if they have moved before Brexit Day. There will be an agreed process (yet to be defined) on incorporating future amendments to EU social security law into the withdrawal agreement. (Usually a Joint Committee of some kind, made up of officials of parties to a treaty, is given the power to adopt decisions to amend that treaty in cases like these). 

Equal treatment as regards access to healthcare and social assistance will be guaranteed in accordance with existing EU legislation – so the limits on access to such benefits, as discussed here, will apply too. Qualifications for lawyers and other regulated professions that have been recognised before Brexit Day will still be recognised afterwards, and applications for recognition of qualifications lodged before Brexit Day will still be processed afterwards.

Enforcement of rights

The joint report states that the withdrawal agreement will go into some detail on enforcement of citizens’ rights – far more so than most international treaties. Citizens must be able to ‘rely directly on their rights’ in the agreement and have laws inconsistent with the agreement disapplied: these are implicit references to the EU law concepts of direct effect and supremacy. There’s no limit in time to these provisions.

More precisely, the UK has committed to introduce domestic legislation, which will refer explicitly to the withdrawal agreement and incorporate citizens’ rights directly in national law. This will prevail over ‘inconsistent or incompatible legislation’, unless Parliament expressly repeals the Act giving effect to the withdrawal agreement. This entrenches (but only in the specific field of citizens’ rights) the existing rule of UK constitutional law relating to the effect of EU law in the UK legal system: the European Communities Act gave the courts the (otherwise constitutionally impossible) power to set aside conflicting Acts of Parliament; implied repeal of that Act by later Acts of Parliament is not possible, but express repeal is. 

EU27 citizens may be suspicious of Parliament’s power of express repeal regarding Act giving effect to the withdrawal agreement, but it represents no change from the status quo as an EU Member State. It would be constitutionally (and surely politically) impossible to ask the UK to overrule this power, giving the withdrawal agreement some sort of super-hierarchical legal status. Subject to vague murmurings from the courts, for the UK* the only basic law is that there is no basic law.  (*except Scotland).

But what happens if the UK parliament did expressly repeal EU27 citizens’ rights? That would then be a matter for the dispute settlement provisions of the withdrawal agreement, which are likely to provide for some sort of sanction at international level in the event that, following some form of arbitration process, it is found that either side has breached its obligations under the agreement. The dispute settlement issue will be part of the ‘governance’ aspects of the withdrawal agreement, which have yet to be agreed; but the joint report explicitly notes that the aspects of citizens’ rights agreed so far is ‘without prejudice’ to the further discussion of such issues.  Ideally there will be some sort of at least indirect access by EU27 and UK citizens to this dispute settlement process, for instance by means of a complaint and request to trigger that system that has to be considered and answered by the UK or EU authorities.

As for the EU27 side, the joint report states simply that the withdrawal agreement will bind the EU and its Member States in accordance with the Treaties. This does not as such give effect to the withdrawal agreement in the domestic law of the EU and its Member States, but it is probably assumed that the withdrawal agreement will have direct effect in EU law. ECJ case law suggests that some international treaties have direct effect in EU law, but some do not, depending on the nature and purpose of each agreement.

By comparison with Demirel, where the ECJ accepted direct effect for the EU/Turkey association agreement that conferred fewer rights and did not guarantee reciprocal direct effect on the Turkish side, the withdrawal agreement is very likely to satisfy the test for direct effect in EU law. But for the avoidance of any doubt – and to ensure reciprocity from the outset – the withdrawal agreement should explicitly set out both parties’ intention to secure direct effect and supremacy of the citizens’ rights rules in their respective legal orders (on the relevance of the parties’ intentions, see Kupferberg).

Jurisdiction of the ECJ

It would be possible to leave it to the courts of each side to guarantee rights established by the withdrawal agreement, particularly in light of the strong provisions on enforcement of those rights in domestic legal systems. However, the EU27 side was particularly keen to ensure some continuing role for the ECJ.

There are several aspects to the ECJ’s role. First of all, where the withdrawal agreement refers to concepts of EU law – and the joint report indicates that it often will – those concepts will be interpreted in accordance with ECJ case law delivered before Brexit Day.  This is consistent with the EU Withdrawal Bill before the UK Parliament (discussed here), although that Bill also provides that the UK Parliament, executive or Supreme Court might decide to depart from such ‘retained’ case law. As discussed further above, the parties have agreed to limit such departures as regards the status of EU27 citizens in the UK.

There’s no limit in time to the obligation to rely on this pre-existing case law, and doing so will ensure greater legal certainty. For instance, there will be no need, to litigate from scratch how to define an EU27 or UK ‘worker’ or when a relevant family member is ‘dependent’, since there is ECJ case law to rely upon (for instance, see here and here respectively).

Secondly, the joint report states that the UK courts will have to have ‘due regard’ to ECJ judgments issued after Brexit Day. This goes further than the UK’s Withdrawal Bill, which would only give an option to courts in the UK to take account of the ECJ’s post-Brexit case law; although (as discussed here) the UK government had previously signalled its willingness to agree to such an approach as regards civil law. Again, there’s no limit in time to this obligation.

Thirdly, the withdrawal agreement should specify that courts or tribunals in the UK could ask the ECJ to rule on a provision of the withdrawal agreement concerning citizens’ rights, if there is no clear case law on the issue, if the litigation brought (presumably before a court in the UK) within a period of eight years after the ‘date of application’ (a concept not further defined) of the citizens’ rights part of the agreement. Note that this differs from the current rules in that the UK Supreme Court will not be obliged to send questions to the ECJ; although it retains the current rule that any other national court or tribunal may do so. The joint report does not state that the ECJ’s rulings in such cases will bind the national court, although the ECJ has made clear that whenever it has jurisdiction, even as regards non-EU countries, its rulings must be binding (see Opinion 1/00, for instance).   

Fourthly, in common with some other international treaties which the EU has signed, there will be an exchange of case law between the two sides, and the right of intervention of the UK before the ECJ. Unusually, there will be a parallel right of the Commission to intervene before UK courts and tribunals.

Finally, the UK will set up an ‘independent national authority’ to monitor the implementation of this part of the withdrawal agreement, with the details to be discussed further in the next phase. It remains to be seen whether it will have the same power as the Commission has on the EU27 side to consider complaints from individuals and to bring proceedings in the courts in the UK to enforce EU27 citizens’ rights. There’s no time limit on the final two points.

Taken as a whole, these provisions are remarkably similar to the rules set out in the Treaty establishing the European Economic Area, which links Norway, Iceland and Liechtenstein to EU internal market law and some other EU policies. Most notably, the rules on previous and subsequent ECJ case law are identical, as are the rules on exchange of case law and judicial intervention (if we substitute the courts in the UK for the EFTA Court). It’s not yet clear if the UK ‘independent authority’ will be as similar to the Commission as the EFTA Surveillance Authority is, but its very existence is a step in the direction of the EEA model.  And the joint report goes further than the EEA in requiring that UK courts must be able to ask the ECJ questions (the EEA only gives EEA states an option to allow this – but then it provides for an EFTA Court instead of the ECJ). While this ‘one-country EEA’ model will only (for now) apply to the specific field of EU27 citizens’ rights, it might end up as a template also for the transitional rules and future relationship which the UK and EU27 will discuss next.


Some have suggested that any ‘special’ rights for EU27 citizens in the UK are objectionable, comparing them to the status of colonial occupiers or to the application of US gun laws in the UK. These comparisons are frankly absurd. The joint report refers only to retaining some aspects of a pre-existing immigration status. To compare keeping legally acquired status (again, reciprocally for UK citizens in the EU27) to the position of an invading power’s citizens is beyond offensive; to compare it to a foreign country’s violent constitutional quirks is simply random. And citizens of the UK’s former colonies may recall that Britain exported to them not only railways and parliaments, but also famines and massacres.

Moreover, to the extent that the joint report states that the withdrawal agreement will guarantee the acquired rights of EU27 and UK citizens’ rights, it only gives effect to what many Leave supporters purported to advocate during the referendum. In particular, a Daily Telegraph article by Leave campaigners (widely disseminated during the referendum) asserted that international law would automatically guarantee full acquired rights for UK citizens living in the EU27 states. The official Leave campaign likewise promised to guarantee ‘no less favourable rights’ for EU27 citizens in the UK. These outrageous guarantees had been promised by the very people now outraged by them. 

From the opposite perspective, many EU27 and UK citizens are disappointed by the joint report. For those UK citizens who haven’t moved within the EU and who are dismayed by the thought of losing free movement rights, their complaint lies with the UK government, which assumed that the Leave vote was a vote to end the free movement of persons. For those who complain that people born in Northern Ireland will have EU free movement rights (due to their Irish citizenship) whereas most people born in the rest of the UK will not, the situation is created by Irish citizenship law; the withdrawal agreement will only recognise the existence of that rule, not create it. Of course, the distinction between UK and Irish citizenship will matter more after Brexit; but that simply brings us back to the UK government’s intention to end free movement.

As for those who have moved, there are parts of the joint report that should be welcomed, and parts where they have good grounds for concern. While the joint report does not itself create rights, that was inevitable given that the parties have decided not to ‘ring fence’ the citizens’ rights issue in a separate treaty. UK and EU27 citizens should keep lobbying for this to take place, as it would especially be necessary if the Article 50 talks subsequently collapse, as they still might (though this now seems rather less likely).

As to the substance of rights, a number of key guarantees ensuring many aspects of acquired rights will be retained are set out in the joint report. There are many important provisions on the administrative process and enforcement of rights too.

However, there are grounds for concern, as I discussed above, about the most vulnerable: the homeless who will lose effective appeal rights and the carers whose rights will depend on the fragile goodwill of the UK government – a whim which could change overnight in response to some angry vomit spewed by a tabloid newspaper. UK citizens in the EU27 are in limbo as regards future free movement rights; and because the UK government values an irrational migration target more than the family lives of working-class British citizens who fall in love with foreigners, the future family lives of lower income EU27 citizens must equally suffer.

While many EU27 citizens in the UK might prefer to keep the role of the ECJ indefinitely, there are a number of other enforcement guarantees for them in the withdrawal agreement that are not subject to any time limit. Any ECJ link with the courts of a non-EU country for any period of time is already exceptional. While the behaviour of the Home Office sometimes gives rise to understandable doubts, by definition a rule of law problem cannot be solved by demanding another court – especially a foreign court whose rulings will provoke greater opposition from nationalists than a domestic court. It needs to be solved by making the case for the rule of law – and the substantive case for EU citizens’ acquired rights – at the domestic level, coupled with an effective dispute settlement system if necessary at the international level.

Overall, the value of yesterday’s agreement also lies in the increased prospect that there will be a final deal on these issues at all – since ‘no deal’ could well leave them worse off than this planned compromise.  But it is not too late to advocate for improvements that would more fully ensure that the millions of UK and EU27 citizens who moved before Brexit Day will not have their lives ruined as a result of the Brexit process.

Barnard & Peers: chapter 13; chapter 27

Photo credit: vice sports

* *This blog post was supported by an ESRC Priority Brexit Grant on 'Brexit and UK and EU Immigration Policy'

Thursday, 7 December 2017

Criminal law, human rights and constitutional pluralism: two views on Taricco II

Jarndyce and Jarndyce drones on. This scarecrow of a suit has, in course of time, become so complicated that no man alive knows what it means. The parties to it understand it least, but it has been observed that no two Chancery lawyers can talk about it for five minutes without coming to a total disagreement as to all the premises. Innumerable children have been born into the cause; innumerable young people have married into it; innumerable old people have died out of it. Scores of persons have deliriously found themselves made parties in Jarndyce and Jarndyce without knowing how or why; whole families have inherited legendary hatreds with the suit. The little plaintiff or defendant who was promised a new rocking-horse when Jarndyce and Jarndyce should be settled has grown up, possessed himself of a real horse, and trotted away into the other world. Fair wards of court have faded into mothers and grandmothers; a long procession of Chancellors has come in and gone out; the legion of bills in the suit have been transformed into mere bills of mortality; there are not three Jarndyces left upon the earth perhaps since old Tom Jarndyce in despair blew his brains out at a coffee-house in Chancery Lane; but Jarndyce and Jarndyce still drags its dreary length before the court, perennially hopeless.

-          Charles Dickens, Bleak House

Interminable legal proceedings have long outlived the voluminous fog of Dickensian London. In a bid to cut them short, some legal systems impose strict deadlines requiring their conclusion. But while it is often argued that justice delayed is justice denied, justice curtailed denies justice too.

In the field of EU law, curtailed justice can cause a particular problem if it results in the inability to enforce criminal sanctions against those who have allegedly defrauded EU funds. Two years ago, in its judgment in Taricco I (discussed here), the ECJ attempted a solution: national courts had to disapply short limitation rules if they prevented the effective prosecution of such frauds. But this ruling was soon embroiled in further litigation, raising questions of national constitutional identity, within the framework of the relationship between EU and national courts and legal systems, in the particular context of criminal justice.

Earlier this week, the ECJ revisited its ruling, now suggesting a compromise with the concerns of the Italian constitutional court. This post comprises two contributions examining the new judgment from different angles: the broader debate about constitutional pluralism, and the ECJ’s relations with the Italian court.

1) Judicial dialogue after Taricco II: who has the last word, in the end?

Barbora Budinska, post-graduate researcher, Leiden University; and Zuzana Vikarska, DPhil candidate at the University of Oxford

On Tuesday (5 December), the second judgment in the Taricco saga added yet another piece of puzzle to the ongoing judicial dialogues between the Court of Justice and the national constitutional courts. The judgment will, most likely, provide food for thought on a variety of themes. In this post, we would like to focus on and shortly sketch the evolvement of the judicial cooperation between the Court of Justice and the Italian Constitutional Court in the Taricco saga and compare it with the progressing dialogue between the Court of Justice and the German Constitutional Court in and after the Gauweiler saga.

There are a number of interesting features that Gauweiler (discussed here) and Taricco II have in common: not only have they both originated at the “barking but not (yet) biting” constitutional courts in Karlsruhe and Rome (with their famous doctrines on constitutional limits to EU integration), but they also share the puzzling motive of constitutional identity, invoked by both constitutional courts, extensively examined by the Advocates-General, yet, formally ignored by the Court of Justice. Against this background, our analysis aims to show that although the Court refrained from explicitly clarifying the concepts of constitutional identity and national identity in its judgments, it still showed a great deal of understanding for national specificities and identities, in line with the principle of respect for national identities embedded in Article 4(2) TEU.

Many landmark judgments of the Court of Justice seem to involve alcohol. After Scotch whisky, Crème de cassis, pure German beer, or the Picon liqueur, this time the Court was dealing with bottles of champagne sold in Italy below the market price, allegedly due to VAT fraud. As a result, an Italian district court launched criminal proceeding againsts Ivo Taricco (and many others) for so-called “fraudulent ‘VAT carousel’ legal arrangements” (Taricco I, para 18).

This scheme involves shell companies, non-existent transactions, fake invoices, and fraudulent annual VAT returns, all of which could put the perpetrators to prison for up to seven years, if they are found guilty. Yet, according to Italian criminal law, the offences are subject to rather short limitation periods and will soon be time-barred (in February 2018 at the latest), despite the pending judicial proceedings. The Italian district court hearing this matter in the first instance pointed out that it is quite usual that criminal proceedings concerning tax evasion are complex and take a considerable amount of time. Therefore, the combination of time-consuming criminal proceedings and (rather short) limitation periods leads to a situation where “de facto impunity is a normal, rather than exceptional, occurrence” (Taricco I, para 24) and where the EU’s financial interests may be put at risk (cf. Article 325 TFEU). The national court therefore turned to the Court of Justice, asking for advice.

The Court opened its judgment in Taricco I by outlining the general principles on effective collection of VAT revenue and the fight against tax evasion, referring repeatedly to the Fransson case (which confirmed the effective collection of VAT revenue is an EU law issue) and to the well-established mantra of “effective, proportionate and dissuasive” sanctions. However, what concerned the referring court was not the sanction as such, but the situation where the offence is commonly time-barred before the criminal proceedings come to an end. In response to this concern, the Court of Justice suggested that if the national court believes that cases of serious fraud regularly escape criminal punishment, then the national provisions are not effective and dissuasive and should be disapplied, “without having to request or await the prior repeal of those articles by way of legislation or any other constitutional procedure” (para 49).

This is nothing new for EU lawyers, as the logic of disapplying national provisions has been clear (and constitutionally controversial) ever since Simmenthal. Yet, a criminal lawyer could just as well point out that such disapplication amounts to “a retroactive application of a rule of substantive criminal law to the detriment of the defendants” which “should be prohibited under the Italian Constitution,” as has recently been claimed by Fabio Guiffrida.

Importantly, the Court did not disregard the fact that fundamental rights of the accused could be at stake: clearly, disapplication of a limitation clause in a pending criminal case puts the accused in a worse position. Yet, having briefly considered Article 49 of the Charter and Article 7 ECHR, which both prohibit retroactivity in criminal law, the Court concluded that the rights guaranteed by these provisions would not be infringed, since “the acts which the accused are alleged to have committed constituted, at the time when they were committed, the same offence and were punishable by the same criminal penalties as those applicable at present” (Taricco I, para 56). Notably, the Court concluded that its legal assessment is still “subject to verification” by the referring court, therefore seemingly handing the responsibility over to the national actors.

Considering the very EU-enthusiastic tone of the reference, the Italian district court was probably pleased by the answer given by the Court of Justice. Not just the referring court, but also other criminal courts in Italy dealing with similar cases understood Taricco I as a green light allowing them to continue prosecuting suspects after the lapse of the limitation periods. This, however, put the convicted individuals in a worse position, for the sake of compliance with EU law. Unsurprisingly, at some point, the Corte Costituzionale (also referred to as “ICC”) was approached with questions of constitutionality of these practices.

In turn, the ICC considered the situation from the perspective of the Italian constitutional order, realising a potential clash between the reasoning in Taricco I on the one hand, and the principle of legality as understood under the Italian constitution on the other hand. Importantly, the ICC decided not to apply its controlimiti doctrine, first established in Frontini in 1973; but instead, it referred its questions to the Court of Justice, seeking what seems to be a “revision” of Taricco I.
First, the ICC reiterated that the limitation periods in Italy form part of substantive criminal law and therefore fall within the scope of the constitutional principle of legality. The ICC voiced its concern about the compatibility of Taricco I with such a “supreme principle of the Italian legal system.” Gently concealing its criticism as “doubts” about the correct interpretation of the Taricco I judgment, the ICC felt compelled to turn to the Court of Justice and present its own views on the situation.

In its referral, the ICC put forward that “the primacy of EU law does not express a mere technical configuration of the system of national and supranational sources of law. It rather reflects the conviction that the objective of unity, [...] justifies the renunciation of areas of sovereignty.” And then, in the same breath, the ICC stated that “the very force of unity within a legal order characterised by pluralism (Article 2 TEU) result from its capacity to embrace the minimum level of diversity that is necessary in order to preserve the national identity inherent within the fundamental structure of the Member State (Article 4(2) TEU). Otherwise, the European Treaties would seek, in a contradictory fashion, to undermine the very constitutional foundation out of which they were born by the wishes of the Member States.” The ICC thus accepts the Court of Justice´s authority on “establishing the meaning of EU law” but expects in return that the Court will leave “to the national authorities the ultimate assessment concerning compliance with the supreme principles of the national order.

The ICC therefore submitted three questions to the Court of Justice. The first one focused on the lack of sufficiently precise legal basis for disapplication of the national provisions, the second one pointed out that limitation periods are part of substantive criminal law in Italy, and the third one asked rather directly whether Italian courts should disapply the limitation periods even if that brings them “at variance with the overriding principles of the constitution … or with the inalienable rights of the individual.

The Court’s judgment in Taricco II: don’t feel obliged to fulfil the obligation

Using the accelerated procedure, the Court of Justice prioritised the Taricco II case and dealt with it in less than a year. It understood very well that the reference from the Corte Costituzionale was an invitation to talk, and opened its reasoning with observations on judicial dialogue and the preliminary ruling mechanism being an instrument of cooperation (paras 22-23). In response to the reference, the Court of Justice provided the ICC with three messages, not necessarily matching the three questions posed.

1. We have asked you to take care of something and we don’t care how you do it.

The Court first reiterated that Article 325 TFEU imposes on the Member States an obligation of results; i.e. to punish serious VAT fraud effectively (para 1) and in the same way as they punish fraud at national level (para 2). Just like in case of directives, Member States are free to choose the means to achieve these results (Taricco II, paras 32-41) while compliance is mainly in the hands of national legislatures, who must “ensure that the national rules … do not lead to impunity in a significant number of cases of serious VAT fraud” (para 41), and national courts, who must “give full effect to the obligations under Art 325(1) and (2) TFEU and to disapply national provisions” if necessary (para 39).

2. We acknowledge the principle of legality and we still don’t think it’s in danger.

Referring to a number of cases from both Strasbourg and Luxembourg, the Court of Justice split the principle of legality into three parts: foreseeability, precision, and non-retroactivity. Foreseeability requires that both the offence and the penalty be clearly defined (para 55), precision focuses on the individuals, checking whether they are in a position to know which acts or omissions will make them criminally liable (para 56), whereas non-retroactivity prohibits the courts to punish conduct which was not prohibited at the time of that conduct, or to aggravate the rules in pending proceedings (para 57). The Court showed understanding for all these aspects of legality, indicating that they are not specific to the Italian legal order, but rather shared by all the Member States, both as common constitutional traditions and as values embedded in the ECHR (paras 53-54).

3. We imposed on you an obligation, but you are not obliged to comply with it.

This is where the Court of Justice got cold feet and downplayed the requirements laid down in Taricco I, this time giving the national courts much softer instructions. Does disapplication of national rules lead to uncertainty and imprecision? Well, in that case you are not obliged to disapply. Does disapplication put suspects in pending cases in a worse position? (Of course, it does, and we knew it already in Taricco I!) Well, then you are not obliged to disapply either.

In fact, if you think that the obligation to disapply conflicts with the principle of legality, you are not obliged to comply with that obligation (isn’t it a magical formulation?). But remember our first message: we have asked you to take care of something, so please do. If your courts cannot do it, your parliament should (para 61).

In his opinion in Taricco II, Advocate-General Bot mentioned that the order for reference from the ICC reminded him of the questions submitted in 2014 by the Federal Constitutional Court of Germany (the Bundesverfassungsgericht or the BVerfG), which gave rise to the Court’s notorious Gauweiler judgment – arguably one of the most heavyweight cases the Court of Justice has ever had to deal with. There, in its very first referral for a preliminary ruling, the BVerfG not only submitted its questions regarding the European Central Bank’s (ECB) Outright Monetary Transactions (OMT) programme but also strongly advised the Court on what the answers to these questions should be. The judges of the BVerfG’s Second Senate were convinced that the OMT programme (based on a press release announcing the same) would constitute an ultra-vires act inasmuch as it would, first, exceed the monetary policy mandate of the ECB pursuant to Articles 119 and 127 TFEU, and second, circumvent the prohibition of monetary financing according to Article 123(1) TFEU.

1. The references from the constitutional courts

Advocate-General Bot is right insofar as pointing out some similarities between the two orders issued by, on the one hand, the German constitutional court in Gauweiler, and on the other hand, the Italian constitutional court in Taricco II: both courts posed a series of preliminary questions while simultaneously suggesting to the Court what they believed to be the “correct” answers. What is more, the two apex courts indicated (more or less aggressively) that in case of an unsatisfactory response of the Court, they, as guardians of their particular constitutions (and of their constitutional identities), would be forced to make use of the limits developed in their respective jurisprudence against the Court’s claim of absolute primacy of EU law, i.e. the well-established BVerfG’s doctrine of threefold limits (fundamental rights, ultra vires acts and constitutional identity), and the controlimiti jurisprudence developed by the Corte Costituzionale.

Yet, from a perspective of judicial dialogue, there is a remarkable difference between Gauweiler and Taricco II. In the former, the BVerfG’s order for preliminary ruling opted for a very harsh tone and manoeuvred the Court of Justice deliberately between a rock and a hard place. On the one side, the announcement of the OMT programme by the ECB’s President Mario Draghi brought the desirable effect of calming the crises-stricken markets; yet, on the other side, the BVerfG left little doubt that it considered the same OMT programme as manifestly exceeding the ECB mandate, and thereby constituting an ultra vires act and violating German constitutional identity. The Court of Justice was left with a daunting reconciliation task of avoiding an open constitutional conflict with the BVerfG and simultaneously preventing the rise of new tensions on the bond markets.

On the other hand, the “tone” chosen by the Corte Costituzionale in Taricco II is considered much friendlier than the one adopted by the BVerfG in Gauweiler. The message from Rome was just as ultimate as the message from Karlsruhe; and the ICC made it clear that it expected the Court of Justice to reinterpret its Taricco I judgment in a way compatible with (the ICC’s understanding of) Italian constitutional identity. Yet, unlike the BVerfG, the ICC chose a strategy combining flattery with diplomatically enunciated ultimata, in other words, the method of carrot and stick, as Davide Paris and Daniel Sarmiento have pointedly noted.

2. The answers from Luxembourg

Advocate-General Cruz Villalón was the first to try and soften the harshness with which the German referral arrived at the Court of Justice. His opinion, issued in a spirit of conciliation between the two apex courts, opened with an elaboration on what he called the “functional difficulty” of the BVerfG’s request: the tension between the purpose of the preliminary ruling procedure and its binding nature on the one hand, and the right to have “the last word” on the validity of an EU act which the BVerfG reserved for itself (para 35).

In an attempt to square the circle and turn the Karlsruhe dictate into a dialogue, the Advocate-General reminded the BVerfG of its own Honeywell judgment, in which the latter made a commitment to genuinely seek the guidance of the Court of Justice before making a final decision (para 44). Moreover, the Advocate-General referred to Article 4(3) TEU enshrining the principle of sincere cooperation, arguing that this principle can constitute a basis for a “cooperative relationship” between the courts. The principle entails obligations for both, the BVerfG as well as the Court of Justice. The former should show its willingness to accept the Court’s preliminary ruling as a sufficient and binding basis for its final decision. The Court, on the other hand, should, first, respond to the BVerfG “in the greatest spirit of cooperation possible”; and second, it should make the effort and address the substance of the referred questions, putting aside any “functional difficulties” arising from the mutual relationship (paras 47-49, 65-68). In response to the BVerfG’s claims of constitutional identity, the Advocate-General invited both courts to strive for “basic convergence between the constitutional identity of the Union and that of each of the Member States” (para 61).

The Court in Gauweiler, seemingly following the advice of the Advocate-General, adopted a pragmatic, yet strict and uncompromising, approach and engaged in a substantive dialogue with the BVerfG on the legality of the OMT programme. It purposefully avoided any elaboration on the tension between the ECB’s actions and the German constitutional identity and contended itself with referring to the settled case law on the binding nature of its preliminary rulings. In response, the BVerfG refrained from its extremely critical tone used in the referral and, overall, accepted the Court’s analysis.

Advocate-General Bot, on the other hand, opted for a much more confrontational tone in his opinion in Taricco II. He recycled his line of argumentation from Melloni (which already smells like trouble), first acknowledging the principle of respect for national identities but then rejecting its applicability to the case at hand, claiming that he was “not convinced” that the Italian claim would constitute a genuine identity claim: “The present case does indeed concern a fundamental right protected by the Italian Constitution, the importance of which should not be underestimated, but that does not mean that the application of Article 4(2) TEU must be envisaged here” (para 179). Marco Bassini and Oreste Pollicino labelled Advocate-General Bot’s opinion in Taricco II as “probably the most conservative and radical view on the supremacy of the EU legal order.”

If the Advocate-General’s harsh stand on national identity were at least supported by a white flag raised by the Italian government (as was the case with the Spanish government in Melloni, as follows from para 141 of Advocate-General’s opinion), it could be regarded as acceptable. Yet, in Taricco II, the identity claim was invoked by no one else than the country’s constitutional court, and nothing suggests that Italy gave up the identity claim in the proceedings in Luxembourg. The Court’s handling of the situation (summarised above) is therefore praiseworthy; unlike the Advocate-General, the court managed “to reconcile the imperative of ensuring the effectiveness of EU law with the good health of its dialogue with national constitutional courts,” as was optimistically foreseen by Alessandra Silveira and Sophie Perez Fernandes earlier this year.

What is more, after the Gauweiler judgment, it seems that even the BVerfG has become more open to an amicable judicial cooperation: in July 2017, it issued its second referral, addressing yet another policy tool of the ECB, the so-called Quantitative Easing (QE) programme. In the order, the BVerfG questions whether the QE programme exceeds the mandate of the ECB (cf. Articles 119 and 127 TFEU) and amounts to a circumvention of the prohibition of monetary financing (Article 123(1) TFEU). Moreover, it invites (perhaps almost challenges) the Court of Justice to apply its own standards developed for the OMT programme in the Gauweiler case. Yet, as some commentators have observed, the language of the second referral is more cordial (see, for example here and here), acknowledging the purpose of the judicial dialogue between the two apex courts, as well as the obligation of the BVerfG under Article 267(3) TFEU. Depending on the response from the Court of Justice regarding the QE programme, this might be a start of a “true judicial dialogue”.

There is, indeed, an inherent functional difficulty, in a dialogue between the Court of Justice and any national constitutional court. They represent different interests, work with different conceptual frameworks, and push for different solutions. A constant fight for the last word, supplemented with (allegedly unconditional) weapons of primacy on the one hand and constitutional identity on the other, leads to no satisfactory solution. The interaction between the Court of Justice and its national counterparts is a diplomatic dialogue of a sort, and it therefore requires diplomatic language. We are certainly not the first ones to emphasise the need for a conciliatory tone in judicial dialogues; still, we are happy to be among the first ones to rejoice about the diplomatic accomplishment of the Court of Justice in its Taricco II ruling - maybe with a glass of champagne? J

2) Taricco II: the Italian Constitutional Court raises its voice and the Court of Justice listens

Massimo Fichera, Adjunct Professor of EU Law. Academy of Finland Research Fellow, University of Helsinki


“The Calm After The Storm” is a poem by the famous 19th century Italian poet Giacomo Leopardi. In a way, this image could represent the climate after the Taricco II case has been finally decided. On 5 December 2017, the CJEU delivered its long-awaited decision in C-42/17 Criminal Proceedings against M.A.S. and M.B. (Taricco II). It was the first time the Italian Constitutional Court threatened to use its “counter-limits” weapon - and the third time it resorted to the preliminary ruling procedure.
To be sure, the relationship between Constitutional Courts/Supreme Courts and the Court of Justice of the European Union (CJEU) has increasingly intensified in recent years. Several rulings, from Ajos (a Danish case discussed here) to the Hungarian asylum-seeker quota case (discussed here), Melloni and the Slovak pensions case, have tested the extent to which EU law is allowed to penetrate into national legal systems. The Taricco saga is just another example of this trend – albeit particularly significant on at least two counts: (a) the interaction between EU criminal law and human rights and  (b) European constitutionalism.

It will be remembered that the question in Taricco was whether Italian rules on prescription periods violated EU law, in particular a number of provisions in the area of state aid, economic and monetary union, competition and the main VAT Directive (although only the last one was considered by the CJEU to have been infringed by national law). As time, quite simply, runs out during the proceedings, this makes fighting fraud against the EU budget and the EU financial interests, as prescribed by Article 325 TFEU, harder. Indeed, Member States are obliged to take effective and dissuasive measures (whatever their nature) to fight fraud, and must do so by adopting the same measures they would adopt to protect their own financial interests.

What is more, there exists a specific obligation to punish fraud against the EU financial interests in an effective and dissuasive manner, in “a serious number of cases” of VAT evasion: this must happen through penal measures. On the face of it, the problem looks easy to solve: in addition to the provisions of the VAT Directive, Article 325 is directly effective, therefore the national judge ought to disapply the Italian provisions on prescription periods, as they run counter an EU law obligation. It is a pure and straightforward application of the classic principles of primacy and direct effect – a case study for first year EU law students.

In fact, this is what the CJEU decided in Taricco and is one of the most important aspects of the ruling. Yet, things are not as smooth as they look. The Italian Constitutional Court, requested by a couple of ordinary courts to trigger its “counterlimits” (i.e. to grind its teeth), chose the middle ground. Facing the choice of either obeying the CJEU or shooting at it, it merely raised its voice. It turned back to the CJEU and pointed out that the actual enforcement of the judgment would breach the fundamental principles of the Italian constitutional system, notably the principle of legality as interpreted by the latter- i.e. as a principle applicable also to prescription periods, because they are considered part of substantive criminal law, not of procedural law, as is the case with other EU countries. This is how we get to Taricco II.


The ruling of the CJEU was preceded by the Opinion of Advocate-General Bot, who admittedly took a hard position. He was rather adamant that, in such circumstances, EU law needs to be complied with. Full stop. Bot argued that progressive harmonisation of criminal law, in particular through a common definition of fraud against the financial interests of the EU (see the recent Directive of the European Parliament and of the Council 2017/1371) and the creation of a European Public Prosecutor (see the recent Council Regulation 2017/1939)  can only be successful if such harmonisation is associated with effective measures, including a uniform discipline of prescription periods.

True, Bot conceded that national courts do not have clear and objective criteria, which would allow them to identify unequivocally those circumstances in which, following serious harm to the EU financial interests, the obligation to disapply national law would emerge. According to Bot, as the existence of a systemic risk of impunity, as suggested in Taricco, is too vague, it would be preferable to refer to the nature of the offence. Yet Bot stops short of providing a remedy to this situation. From his point of view, compliance with EU law obligations cannot be doubted or challenged.

The CJEU, instead, is much more cautious. It first shows the “stick”: national judges are in principle obliged to fully respect the obligation enshrined in Article 325 TFEU – with the consequence, as mentioned above, that national provisions preventing compliance with such obligation ought to be disapplied. However, soon comes the "carrot". Protection of the financial interests of the EU through criminal law belongs to the shared competences of the EU and Member States.

In this context, it cannot be ignored, says the Court, that, at the time of the commission of the crime, there was no harmonised legislation on fraud against the financial interests of the EU. As a result, Italy had a large room of manoeuvre and was free to regulate the field as it wished – including the application of the principle of legality to prescription periods. It is undeniable that the alleged offenders were not able to foresee the circumstances in which Article 325 TFEU applies. The main problem deriving from following the Melloni/Fransson doctrine here is that the Italian system does not allow the judge to replace the legislator by providing the missing criteria. Hence, not only the principles of foreseeability, clarity and non-retroactivity, but also the principle of separation of powers would be compromised.

In such situation, it is understandable why the Italian Constitutional Court indicated its readiness to trigger the “counter-limits”. And the CJEU could not help recognising the peculiarities of the Italian system, despite all the criticism levelled (not only by scholars, but also by the ECtHR) at the discipline of the prescription periods, which, as a matter of fact, ensure a high degree of impunity. Importantly, the CJEU, while discussing the principle of legality, considers its relevance both for the EU legal order and national legal orders. Moreover, the Court includes it in the Member States’ “common constitutional traditions”, by mentioning several provisions, including Article 7 (1) of the ECHR.


The Taricco saga proves the importance of EU criminal law, and the extent to which it has penetrated and still is penetrating into the national sphere. The approach followed by the EU has been that of allowing some degree of discretion to Member States, whilst setting the guidelines for their action.
Moreover, one could discern three fundamental mindframes in the Taricco saga. They all contribute, willingly or unwillingly, to a rich discussion not only on issues of constitutional identity, but primarily on how we should interpret the principles of primacy and direct effect – and, ultimately, on the nature of the EU legal order/system.

First and foremost, we have the “supremacy” mindframe, which can be observed in the first Taricco judgement and in Advocate General Bot’s Opinion. Essentially, Bot follows his own Opinion in Melloni, which was developed on the basis of the interpretation of Article 53 CFR, shared by the CJEU on that occasion. Whenever the EU provides an exhaustive harmonisation of fundamental rights in a certain area, Member States may not require higher standard of protection, even when these are ensured by their national constitution. In other words, it is the EU that decides on the adequate level of protection of fundamental rights by performing its own balance between rights and the effectiveness of EU law (see also Fransson).

Of course, to the extent that the subject matter has not been completely regulated by EU law, Member States are still allowed to go beyond EU standards. This observation should not be downplayed. Yet, and interestingly, while acknowledging this state of affairs, Bot still supported the Taricco judgement. Bot’s interpretation of the relationship between EU law and domestic law is, in a sense, hierarchical. It is not by chance that, in his view, mutual trust and mutual recognition (as in Aranyosi and Căldăraru) are to be viewed as “absolute”: there is no space in the Framework Decision on the European Arrest Warrant for any grounds for refusal of surrender by the national judge of the requested State based on the infringement of the fundamental rights of the surrendered person in the issuing country. (Although in those cases, as discussed here, the ECJ found a compromise – the postponement of surrender until human rights could be ensured).

The CJEU in Taricco II instead follows a “primacy” mindframe. It does not view the relationship between EU law and national in hierarchical terms and, rather, seeks cooperation. In this light, its revirement can be considered as part of a strategy of rapprochement and reconciliation with the Italian Constitutional Court. In other words, the CJEU has realised that, behind the conflict, there emerges a crucial legal-cultural difference.

The Italian legal system does not accept an interpretation of the principle of separation of powers in such a way that a judge is free to establish legal criteria and categories, which would normally be entrusted to the legislator. True, the Italian legislator has been inactive, and could have filled the gap earlier. Yet, this does not justify an intrusion of EU law to the extent of altering the constitutional balance of a Member State. The effort of the CJEU (which employs in its reasoning the notion of “common constitutional traditions”) is certainly laudable. While confirming the importance of the national identity clause (Article 4(2) TEU), the CJEU attempts to build up common principles with the aim of showing how all provisions on fundamental rights and fundamental principles can be read together systematically.

The third mindframe can be defined “statalist”. Such mindframe can be observed in all those legal systems which resist EU law. In Italy, for example, part of the doctrine interprets “counter-limits” as widely as possible, as embracing not only substantive constitutional norms, but also the principle of conferral and “identity review”. From such perspective, the power to decide to which extent legislative competence should be exercised by the EU belongs to the Italian Constitutional Court.

For the same reason, it has been argued that, instead of referring the questions to the CJEU, the Constitutional Court should have triggered its  “counter-limits” immediately, because the EU is acting ultra vires and has interpreted the notion of direct effect too extensively. In other words, there would simply exist no conditions allowing Article 325 TFEU to be directly effective. Needless to say, any such interpretation would make EU law pointless, because it would have as a consequence that of permitting any national court to challenge the applicability of direct effect (and primacy) at any time. This is an old debate (as the German Bundesverfassungsgericht and other constitutional courts can prove), but it is important to bear in mind that the “statalist” view is still very much present and does not need to be expressed in its more radical forms, as can be observed in Hungary or Poland.

Both courts – the Italian Constitutional Court and the CJEU after its revirement – ought to be praised for their attempt at building up a systemic view of fundamental rights and fundamental principles. They could have adopted a more intransigent stance, and yet they have showed willingness to cooperate, from two separate yet cooperating systems. However, Taricco II also shows the importance of conflict in EU law. It is by raising its voice that the Constitutional Court has showed how delicate matters relating to national diversity may be addressed. Whether this method will work in other contexts is another issue.

Photo credit: BBC News
JHA4: chapter II:5

Barnard & Peers: chapter 6, chapter 8, chapter 25