Tuesday, 17 January 2017

Brexit: the Prime Minister sets the wrong course

Steve Peers

Today’s speech by Prime Minister Theresa May gave a number of indications as to the government’s intentions as regards implementing Brexit. Overall, while the speech contained some welcome parts, it made fundamentally the wrong decision about the country’s future.

Welcome parts of the speech

The welcome parts of the speech include the argument that it ‘remains overwhelmingly and compellingly in Britain’s national interest that the EU should succeed’. Indeed, any ‘unravelling’ of the Union between these neighbouring states is not in the economic or security interests of the UK. Her suggestions about what the EU should do next contain much common sense – although it is doubtful that the remaining EU is very interesting in listening to the leader of a country that is intending to leave. In particularly, her complaints about too much ‘uniformity’ and not enough ‘diversity’ will strike some as bizarre – coming from a country with opt-outs on the single currency, Schengen, justice and home affairs (and previously social policy) plus a budget rebate.

The commitment to retain status for EU citizens in the UK is not new, but still welcome. It is disappointing however that there was no commitment to entrench their rights unilaterally, as recently proposed by a group of Leave and Remain supporters in a British Future report. There could be compromise ways to address this: publishing a draft Bill to this effect, or entrenching the rights in law conditional on EU reciprocity. One can only hope that the issue will be addressed at an early stage of the negotiations.

The interest in continued collaboration on research, police cooperation and foreign policy is also welcome, since the UK still has joint interests with other Member States in these fields. But it is content-free: what exactly would the UK like to participate in? How does this square with her assertion that the UK will not be involved with ‘bits’ of the EU?

Single market and customs union

The Prime Minister declared her opposition to ‘partial membership of the European Union, associate membership of the European Union, or anything that leaves us half-in, half-out.’ But there is no such thing as ‘partial’ or ‘associate’ membership of the EU. May is slaying straw dragons in her own imagination here.

She goes on to confirm her opposition to single market membership (as distinct from single market access) for the UK, for several reasons. It is striking that she makes no assertion that the UK will be better off out of the single market economically. Indeed, the IFS has estimated that the UK will lose 4% of GDP if it leaves the single market without a free trade deal, due to the loss of market access that this entails. While May goes on to say that she seeks a free trade deal, this is bound to entail less trade between the UK and the EU than single market membership, as free trade deals do not remove as many non-tariff barriers as the single market rules.

So what are her reasons? One is control of immigration – and free movement of persons is a non-negotiable condition of the EU for participation in the single market.  Here she fails to consider that the European Economic Area (EEA) treaty includes a safeguard on free movement which could be invoked in order to control it. May’s description of free movement includes overstated claims about its effect on public services, ignoring the impact of limited government funding of health and education in recent years – while she cannot bring herself to mention the overall economic benefit derived from EU migrants.

Another is budget contributions. She rules out any budget contributions except for participation in individual programmes. There is no consideration of whether the EEA option – giving money directly to poorer EU countries, with more control over the spending by the contributor – would be desirable in return for increased market access.

Next, there is the role of the ECJ. May states that single market membership ‘would mean accepting a role for the European Court of Justice that would see it still having direct legal authority in our country.’ Let’s not mince words: this is not true. The EEA states are not subject to the ECJ at all, but to the separate EFTA Court. That court has less jurisdiction than the ECJ, and a large number of its rulings are not binding at all. It is only obliged to follow ECJ rulings delivered before 1991.

More broadly, May states that this ‘would to all intents and purposes mean not leaving the EU at all.’ Again, this is not really true. EEA members are not subject to EU rules on agriculture, foreign policy, fisheries, justice and home affairs (except via separate treaties, in part) or trade with non-EU countries – the very issue which May devotes a large part of her speech to.

This brings me to a false dichotomy on which her speech rests: that there is some sort of choice to be made between EU membership and ‘Global Britain’. In fact, barriers to trade with non-EU countries have been coming down, both due to EU membership of the WTO and due to bilateral trade deals between the EU and non-EU countries. The share of UK trade with non-EU countries has therefore been rising – as Leavers are often quick to point out. Many other EU countries trade more that the UK does with non-EU states – as May herself pointed out last year. So it’s not EU membership that significantly holds back trade with non-EU states.

It is true that inside the EU’s customs union, the UK cannot sign its own trade deals with non-EU states. But the UK could seek to remain in the single market (like Norway) but leave the customs union. Indeed, Norway and other EEA countries have a number of their own trade agreements. In effect, this would be the best of both worlds – maintaining the maximum possible access to the EU’s internal market via means of full participation, while simultaneously having the freedom to sign additional trade deals with non-EU countries.

She also argues that both sides in the referendum made clear it was about the single market. But the single market was not on the ballot paper and was not often mentioned. When it was mentioned, some Leavers, like Dan Hannan, expressly declared that single market membership would not be affected. I recall well a common cut-and-paste statement from Leave supporters in Facebook posts beginning ‘The UK will not be leaving the EEA…’. Although David Cameron stated that leaving the EU meant leaving the single market, why should anyone be bound by his falsehood? And why should one claim made during the campaign be treated as politically binding, while others – notably those which appeared on the side of a bus – are not?

As for the customs union, May proposes a ‘have your cake and eat it’ version – a special deal simplifying border crossings, while being free to sign the UK’s separate international trade deals. Time will tell if this idea interests the EU.

A transitional deal

The Prime Minister accepts that the UK cannot switch immediately to a new arrangement, but cannot bring herself to support a transitional deal, saying ‘[i]nstead, I want us to have reached an agreement about our future partnership by the time the 2-year Article 50 process has concluded’. Such an arrangement would then be phased in. This time frame is unlikely, given that she wants a bespoke deal, involving special arrangements on customs and comprehensive free trade.  So what happens if the Brexit Fairy does not deliver by this deadline?

The role of parliament

Early on in the speech, May states that ‘the principle of Parliamentary Sovereignty is the basis of our unwritten constitutional settlement’. Unfortunately, these are empty words. A Martian reading this would assume that she had gone to court to try to ensure parliamentary involvement in the triggering of Article 50 – rather than to block it.

Furthermore, her speech comes in place of any white paper or any other public consultation on the best course to follow after Brexit. She ‘concedes’ that parliament will vote on the final deal, but this is not much of a choice – a free trade deal or nothing – unless there is an option to negotiate a different deal (not enough time) or to stay in the EU on the basis of another referendum on the exit terms (ruled out by the government).  In any event, it’s not a real concession: the Constitutional Reform and Governance Act of 2010 makes a form of parliamentary control a legal requirement in principle for most treaties. She made no commitment for a full Act of Parliament to approve the final deal – even though one is required for even minor changes to EU Treaties, and even for the approval of some EU legislation.

So May seeks credit for doing something she was anyway legally required to do. In fact, she deserves blame for previously threatening to ignore the law, and even now involving Parliament as little as possible and planning to offer it a fait accompli.

As for EU legislation converted to UK law, by the future Great Repeal Bill, she states that it will only be changed ‘after full scrutiny and proper Parliamentary debate’. This sounds nice superficially, but falls short of a commitment to use Acts of Parliament on key issues. Rather it sounds like an intention to use Statutory Instruments, which can’t usually be amended by Parliament and are rarely blocked. Without a commitment to use Acts of Parliament, her guarantee to uphold workers’ rights derived from EU law is worth rather less than she suggests; and there is no such commitment as regards environmental law.

The devolved administrations

The Prime Minister states that ‘we will put the preservation of our precious Union at the heart of everything we do’ and that she will ‘strengthen our precious Union’. However, her plan necessarily rejects the detailed suggestions of the Scottish government from December (discussed here) for the future EU/UK trade relationship.  So not only is the Scottish (and Northern Irish) public’s view on the desirability of Brexit is overridden, but also the Scottish government’s later views on how Brexit should take place are ignored. The Scottish government paper can hardly be ‘considered’ if it has already been overruled.

There’s a pledge not to weaken existing powers of devolved bodies, but there will surely be battles ahead over which level of government should exercise powers over devolved competences returned from the EU.  Conversely, there’s no suggestion of any granting any additional devolved powers, which might have been appropriate to address the obviously highly divergent views of Scotland, Northern Ireland and the rest of the UK. There’s another pledge to maintain the Common Travel Area between Northern Ireland and the Republic of Ireland, but this is content-free.

In short, there’s nothing here to ‘strengthen’ the Union at all. Its ‘preservation’ depends solely upon the continued argument that Scotland would be worse off outside the UK’s economic union – while simultaneously maintaining that the UK is better off outside the European version of the same.

Unity and Brexit

The Prime Minister declares that the referendum ‘victors have the responsibility to act magnanimously’, and the losers to accept the result. But she has not shown the slightest magnanimity in her speech today. She dismisses the arguments for staying in the single market made by those – like the Scottish government – who sought to remain in the EU but who believe that single market membership would be a reasonable compromise for a badly divided country.

More broadly, her emollient tone today cannot erase the memory of her conference speech in October – full of sneering references to ‘citizens of the world’ and the dreaded ‘liberal elites’ (cue the Star Wars Imperial March music). It’s a strange world in which Alexander Boris de Pfeffel Johnson – graduate of Eton and Balliol College – dismisses people like me – the grandson of a miner, the son and stepson of factory workers – as part of the ‘elite’.

Still less can her speech erase the memory of her Lord Chancellor failing in her statutory duty to defend the independence of the judiciary from screeching headlines about the ‘Enemies of the People’. And if she really believed in magnanimity in concrete terms, she could have announced a unilateral decision to let EU citizens stay in the UK.


Some of the Prime Minister’s speech is valuable – setting the right overall tone on relations with the EU, implicitly rejecting the more harmful 'WTO-only' option, and eschewing (hopefully genuinely) future derision of the 48% who took a different point of view in the referendum. But ultimately she has made the wrong decision on single market participation, putting politics ahead of the country’s economic interests. And key parts of the speech are vague, incorrect, misleading, hypocritical or fantasist. Perhaps we were better off with ‘Brexit means Brexit’.

Barnard & Peers: chapter 27
Photo credit: BBC

Saturday, 14 January 2017

Non-refoulement: is part of the EU’s qualification Directive invalid?

By Pieter Boeles, visiting professor migration law, VU University Amsterdam, emeritus professor University Leiden


The absolute character of the principle of non-refoulement – the ban on removal to an unsafe country – is widely acknowledged. Still, there is confusion and insecurity on this point with regard to the Qualification Directive, which defines how to determine if someone enjoys refugee or subsidiary protection status within the EU. On 14 July 2016, a Czech Court (the Nejvyšší správní soud) asked the Court of Justice EU whether Article 14(4) of the EU Qualification Directive, allowing for revoking, ending or refusing to renew refugee status for reasons of criminal behaviour or a security risk, is invalid in the light of the principle of non-refoulement (Case C-391/16). In its explanation, the Czech Court points out that ‘the binding nature of the prohibition on the return of persons in contravention of the principle of non-refoulement forms part of the obligation not to subject anyone to torture or inhumane or degrading treatment or punishment under Article 3 ECHR and Article 4 and Article 19(2) of the Charter, and it applies whenever there is a real risk of such treatment occurring as a result of forced deportation or extradition’.

The question of the Czech court is important. In my view, a well-reasoned answer can only be given if the ambivalent structure of the Qualification Directive on this point is acknowledged and addressed.  In this comment I will try to analyse the problems to be solved. 

One sole principle of non-refoulement in EU law

Basically, the dogmatic point of departure is simple: the EU principle of non-refoulement is anchored in Article 19(2) of the Charter of Fundamental Rights of the EU, which contains a prohibition to remove, expel or extradite any person to a State where there is a serious risk that he or she would be subjected to the death penalty, torture or other inhuman or degrading treatment or punishment. The Charter should govern the uniform interpretation of the principle of non-refoulement in Union law, both in the Treaties and in secondary legislation (like the Returns Directive and the Qualification Directive). As the prohibition of refoulement is absolute in the ECHR, it should universally be interpreted to be absolute regardless of the legal context of EU law in which it appears. Article 19(2) of the Charter corresponds to Article 3 ECHR, and so must be interpreted the same way (Article 52(3) of the Charter). See the ECtHR ruling in Chahal, and more case law in Kees Wouters, International Legal Standards for the Protection from Refoulement, Intersentia, 2009, p. 307 – 314. The Court of Justice has recognized the absolute nature of the rule in its judgment in Aranyosi (paras 85-87).  

Article 19(2) Charter, derived from Article 2 and 3 ECHR, is not only relevant for persons with subsidiary protection but equally for refugees. Refugees are protected against a danger, formulated (in Article 1.A of the Refugee Convention, as well as the EU Qualification Directive) as persecution on grounds of race, religion, nationality, political opinion or membership of a particular social group. The persecution feared must be of a certain level of severity, which will normally coincide with death penalty, torture or other inhuman or degrading treatment or punishment. Under the uniform meaning of non-refoulement in Article 19(2) of the Charter, absolute protection against refoulement must be granted on an equal basis to both refugees and subsidiary protected persons. As the principle of non-refoulement is a prevailing general principle of EU law in the sense of Article 6 TEU, secured by the Charter, it follows that the Qualification Directive grants stronger protection to refugees than the Refugee Convention alone does.  This is also what the Czech court asserts.

The stronger protection does not undermine any of the rights for refugees granted in the Refugee Convention.  But it may mean that non-refoulement has a wider meaning for refugees within the scope of EU law. Although the wording of Article 14(4) of the Qualification Directive matches the exception to the non-refoulement rule in the Refugee Convention, the Charter sets a higher standard for non-refoulement (as confirmed by reference to Chahal).

This is not unequivocally visible in the text of the Qualification Directive  The seminal Article 21(1) of the Directive, which deals with non-refoulement directly, does not say straightforwardly that Member States shall respect ‘the absolute principle of non-refoulement’, or ‘the principle of non-refoulement in accordance with Article 19(2) the Charter of Fundamental Rights’. True, the present text of the first paragraph of this article, stating that Member States shall respect the principle of non-refoulement ‘in accordance with their international obligations’, can be interpreted that way, especially because the ‘international obligations’ must be deemed to include those under Article 2 and 3 ECHR. But if that is the correct interpretation, it is difficult to understand what the reasonable meaning can be of the second paragraph of Article 21, stating that refoulement of a refugee is nevertheless allowed in some cases, ‘where not prohibited by the international obligations’. In suggesting that refoulement would nevertheless be allowed under certain conditions, Article 21(2) Qualification Directive is confusing. Obviously, Article 21(2), like Article 14(4) of the Directive, implicitly refers to the exception to the non-refoulement rule set out in Article 33(2) of the Refugee Convention, as it is phrased nearly identically.  Article 21 thus begs the question what must be considered  to be the prevailing norm informing the ‘principle of non-refoulement’. The absolute norm of Article 19(2) in the Charter - or the non-absolute norm of the Refugee Convention?

I cannot help suspecting that the Court of Justice was captured in this confusion, when earlier answering the questions asked in the case of H.T. (C-373/13), discussed here. The Court of Justice described the system of Article 21 Qualification Directive in paragraphs 41 – 44 of that judgment. In that description any explicit referral to the absolute character of the principle of non-refoulement is lacking. Strictly, the Court says nothing wrong, stating in paragraph 42 that Article 21(2) Qualification Directive, ‘whose wording essentially repeats that of Article 33(2) of the Geneva [Refugee] Convention, nevertheless provides for a derogation from that principle, allowing Member States the discretion to refoule a refugee where it is not prohibited by those international obligations (…)’.

It may be telling that the Court notes in para 65 that the principle of non-refoulement is guaranteed as a fundamental right by Articles 18 and 19(2) of the Charter of Fundamental Rights of the European Union. But the Court does not elaborate on that. Then again, I do not understand the extensive attention drawn to the alleged leeway granted by Article 21(2) of the Directive. Why suggest that there can be meaningful room for refoulement at all? Why make mention of ‘Member States, enjoying the discretion whether or not to refoule a refugee’ in paragraph 43? Why talk of a ‘derogation’ in paragraphs 42, 55, from this non-derogable principle? Why state in paragraph 72, that the consequences of applying Article 21(2) may be “very drastic” because the refugee ‘might be returned to a country where he is at risk’? Why keep secret that the whole exercise of explaining Article 21(2) is essentially futile because of the absolute character of the principle at stake?

Crucial impact of the case

The answer of the Court of Justice to the Czech court’s questions will be crucial for the future development of EU law. If the absolute character of the principle of non-refoulement is not clearly upheld now, we might be witnessing a gradual process of interpreting away the absolute character of non-refoulement. The first step could be, to frame the refugee status in the Directive as the primary status. The protection granted by Article 3 ECHR could then be downgraded as subsidiary and less important and only to be used as a safety net for persons who are not eligible for the first class protection. In such a construction, it could be argued that the international obligations referred to in Article 21 Qualification Directive are essentially obligations under Article 33 of the Refugee Convention.  This would grant Member States room for refoulement of refugees under Article 21(2) of the Directive in cases of criminal behaviour or security risks.  Then, if the ‘primary’ status would not be deemed connected to an absolute protection of non-refoulement, it could be considered weird to grant any better protection to the ‘lesser’ status.  

On the other hand, such a downgrading process may less easily develop if the importance of Article 19(2) of the Charter in this respect is acknowledged. In that respect, it is striking that the referral to the principle of non-refoulement in point 3 of the preamble of the Qualification Directive is solely linked to the Refugee Convention and not to Article 19(2) Charter. It is further conspicuous that point 16 of the Preamble mentions a whole range of relevant Charter provisions except for precisely Article 19(2).  

Complications: status, residence permit, expulsion

The discussion on this topic is complicated by a number of circumstances. First, refoulement only refers to deportation to the country where the person is in danger. Non-refoulement as such does not stand in the way of expulsion to other countries (as can be seen in Article 32 of the Refugee Convention allowing a limited possibility of removals to other countries, even if a refugee is ‘lawfully in [the] territory’). Further, the protection granted by the Qualification Directive takes the form of issuing statuses and residence permits, among other things, to protected persons.  What follows from this? Status and residence permit are two different things, as can be seen in Article 24 Qualification Directive. Revoking a residence permit cannot lead to the revocation of the refugee status (H.T. judgment, para. 74). Neither will the status of subsidiary protection be affected by revocation of the residence permit. What does the distinction between status and residence permit mean? And what is the relationship between a status and the obligation of non-refoulement?


Principally, a status is recognition. Under Article 2(e) Qualification Directive, ‘refugee status’ means the recognition by a Member State of a third-country national or a stateless person as a refugee, and under Article 2(g) ‘subsidiary protection status’ means the recognition by a Member State of a third-country national or a stateless person as a person eligible for subsidiary protection.  As a status is equated with recognition, a status can only be terminated together with the recognition. The status intrinsically comprises recognition of the existence of an obligation of non-refoulement with respect to the person concerned.

(When I use the term ‘recognised refugee’ I am only dealing here with recognised refugees under the Qualification Directive. A recognised refugee under the Qualification Directive is in a different position from a person who is not recognised but who might still be a refugee under the Convention of 1951. It is the recognition under the Qualification Directive which puts beyond doubt that an the absolute EU principle of non-refoulement is applicable to the refugee as well).

Now, if a ‘status’ should be equated with ‘recognition’ of the danger threatening the person involved, it would be an anomaly to revoke a status in cases where the danger continues to exist, just because of criminal behaviour. That would amount to an arbitrary refusal to acknowledge the real and continuing risk against which the person seeks protection. More or less like revoking the medical acknowledgement of pregnancy of a woman because she stole a book, not because the pregnancy was over. However, under the Qualification Directive, termination of a status is made possible in cases where it is not established that an absolute prohibition of refoulement is no longer applicable to the person concerned. This is especially so with refugees. On the same grounds as set out in Article 21(2), but without the proviso of the ‘international obligations’ override, Article 14(4) Qualification Directive allows for revoking, ending or refusing to renew a refugee status. 

The wordings of Article 14(4) obviously refer to Article 33(2) Refugee Convention which also inspired the creation of a possibility of ‘refoulement’ under Article 21(2). In the beginning of my exploration, I stated that a correct interpretation of Article 21(1) compels non-application of the second (and consequently the third) paragraph. So, if the second and third paragraphs of Article 21 should be declared ‘dead letters’, is there still room for applying Article 14(4)?

The answer may depend on how consistency of the system of the Directive is valued. It is strange to refuse or terminate a status on grounds that have nothing to do with the danger against which the status is meant to offer protection. But, as long as there is no refoulement, the refusal or termination of the refugee status as cannot violate the principle of non-refoulement. 

Residence permit

In the H.T. judgment, para. 95, it is claimed that, even without his residence permit, the person concerned remains a refugee and as such remains entitled to the benefits guaranteed by Chapter VII of the Qualification Directive to every refugee, including protection from refoulement, maintenance of family unity, the right to travel documents, access to employment, education, social welfare, healthcare and accommodation, freedom of movement within the Member State and access to integration facilities. The same must be true for a person with the status of subsidiary protection (Article 20(2)).

However I have some difficulties with this passage. A number of rights mentioned in Chapter VII – like travel documents for travelling abroad, freedom of movement within the territory of the host state, access to employment -  would, under the Refugee Convention only be granted to ‘lawfully’ present refugees. I doubt it whether the Court took this aspect fully into account. In Chapter VII no clear distinction has been made between the rights correlating to the ‘status’ alone and rights specifically connected to ‘lawful residence’.  

I am therefore not convinced that para. 95 of the H.T. judgment is tenable. Is it really true that a ‘status’ alone already implies lawful presence? Is it not rather so, that a ‘status’ differs from a ‘residence permit’ precisely because a ‘status’ does not in itself regulate a right to lawful presence on the territory?

Whatever the right answer to that question is, even if we fully accept para. 95 of the H.T. judgment, a ‘status‘ only protects against refoulement, not against any form of expulsion to a safe country. That is why the residence permit offers more security. As long as a residence permit is granted to a status holder, the Member State guarantees not only non-refoulement to the country of origin but also non-expulsion to any other country and a full right to inclusion in society. This applies both to refugees and to subsidiary protected persons. With a residence permit the legal position of the person may become stronger through time and eventually lead to a permanent status or to nationality of the host state.

The principle of non-refoulement is, at least in abstracto, reconcilable with revoking, non-extending or refusing a residence permit.  Under Article 24 Qualification Directive, as explained in the H.T. judgment, a non-renewal or revocation of a residence permit is possible both for a refugee and for a subsidiary protected person in case of compelling reasons of national security or public order, while the status remains unaffected. So, measures to protect public order and public security may take the form of terminating or refusing a residence permit, but they may, according to what I said above never lead to refoulement as long as it is prohibited under Article 19(2) Charter.


A thorny issue is how the concept of exclusion (as distinct from revocation of refugee status, and/or a residence permit) must be positioned under an absolute norm of non-refoulement.

The exclusion clauses of the Refugee Convention are transposed in Article 12 of the Qualification Directive. Further, in Article 17 of the Directive, the concept of exclusion is also applied to persons with subsidiary protection – a novelty. According to the wording of the Directive, exclusion may mean something different for refugees compared to subsidiary protected persons. In Article 12, a third-country national or a stateless person is excluded from being a refugee.  Thus, an excluded person ‘is’ not a refugee.  He is excluded from the definition. In the wording of the Court of Justice in B and D (paras 89, 91, 98, 100, 104 and 106), the person is excluded from refugee status. In contrast, under Article 17, a third-country national or a stateless person is excluded from being eligible for subsidiary protection. Here, the person is excluded from the ’eligibility’ to be protected, which may not be exactly the same as a status. However, for reasons of consistency with the language of the B. and D. judgment, it is perhaps preferable to see Article 17 as dealing with exclusion of the status of subsidiary protection, just like Article 12 is about exclusion from the status of refugee.

Anyhow, regardless of the potential differences between the two provisions, their common effect is that they prevent or undo the creation of a legal moment in which it must be assessed under the Qualification Directive whether the person concerned is in a situation where refoulement is prohibited, either because of well-founded fear for persecution or because of a real risk for life or of torture, inhuman or degrading treatment or punishment. Of course, non-assessment of a risk does not mean that it does not exist. Therefore, exclusion under the Qualification Directive leaves open that the principle of non-refoulement may still apply to excluded persons. Under Article 5 of the Returns Directive they still are protected against refoulement. So, it is not a priori allowed to send excluded persons back to their countries.

Above, I stated that the status embodies a recognition of an obligation of non-refoulement with regard to the person concerned. Accordingly, an exclusion from a status is equivalent to exclusion from recognition of an obligation of non-refoulement. As a consequence, there is no right to a residence permit either.

In fact, excluded persons are thrown out of the field of application of the Qualification Directive and are referred to the Returns Directive for further protection against refoulement. In accordance with the Abdida judgment, Court of Justice 18 December 2014, C-562/13, para 50 (discussed here), they must be able to avail themselves, in such circumstances, of a remedy with suspensive effect, in order to ensure that the return decision is not enforced before a competent authority has had the opportunity to examine an objection alleging infringement of the principle of non-refoulement laid down in Article 5 Returns Directive and Article 19(2) of the Charter.


In spite of its confusing content, Article 21 Qualification Directive is not invalid because it can be interpreted in accordance with the absolute prohibition of refoulement thanks to the words ‘in accordance with their international obligations’  in the first section. The effect of the correct interpretation is that refoulement of refugees is not allowed even in the cases formulated in paragraph 2. Consequently, neither the third paragraph of Article 21, dealing with revoking, ending or refusing to renew or to grant the residence permit of (or to) a refugee to whom paragraph 2 applies, can ever be applied. Correctly interpreted, the second and third paragraph of Article 21 must be regarded dead letters.

It is difficult to conclude what the ‘dead letter’ position of Article 21(2)(3) means for the validity of Article 14(4) which was based on the same Article 33(2) Refugee Convention. At any rate, Article 14(4) is not invalid for violating the absolute prohibition of refoulement as long as Article 21(2) is not applied. Still, there is something anomalous about Article 14(4). It compels Member States (‘shall’) to refuse or terminate a refugee status in cases where there is still an obligation to acknowledge the absolute prohibition of refoulement. If it would be accepted that a ‘status’ is equivalent to ‘recognition’, the ‘status’ should be maintained, also for persons committing crimes, as long as the ‘recognition’ has solid ground.    

In answering the questions of the Czech Court, the Court of Justice may also have to clarify (as it did in Abdida) how the protection against refoulement is divided between the Qualification Directive and the Returns Directive. Exclusion in the sense of Articles 12 and 17 of the Qualification Directive has the effect of throwing the applicants out of the protection system of the Qualification Directive and leaving their protection against refoulement to the Returns Directive, which entitles them to a remedy with suspensive effect for the examination of a claim of non-refoulement.

Barnard & Peers: chapter 9, chapter 26
JHA4: chapter I:5

Photo credit: Bryan Denton, New York Times

Tuesday, 10 January 2017

A Threat to Human Rights? The new e-Privacy Regulation and some thoughts on Tele2 and Watson

Matthew White, Ph.D candidate, Sheffield Hallam University


In a follow-up to last Christmas’s post, on 10 January 2017, the European Commission released the official version of the proposed Regulation on Privacy and Electronic Communications (e-Privacy Regs). Just as the last post concerned the particular aspect of data retention, this post will too.

Just as the former leaked version maintained, the proposal does not include any specific provisions in the field of data retention (para 1.3). This paragraph continues that Member States are free to keep or create national data retention laws, provided that they are ‘targeted’ and that they comply with European Union (EU) taking into account the case-law of the Court of Justice of the European Union (CJEU) and its interpretation of the e-Privacy Directive and the Charter of Fundamental Rights (CFR). Regarding the CJEU’s interpretation, the proposals specifically refers to Joined Cases C-293/12 and C-594/12 Digital Rights Ireland and Seitlinger and Others, and Joined Cases C-203/15 and C-698/15 Tele2 Sverige AB and Secretary of State for the Home Department. Aspects of the latter case is the focus of this post; the case itself has been thoroughly discussed by Professor Lorna Woods.

So, when is the essence of the right adversely affected?

Before discussing certain aspects of Tele2 and Watson, it is first important to draw attention to the provision which enables data retention in the new e-Privacy Regs. Article 11 allows the EU or its Member States to restrict the rights contained in Articles 5-8 (confidentiality of communications, permissions on processing, storage and erasure of electronic communications data and protection of information stored in and related to end-users’ terminal equipment). From Article 11, it is clear that this can include data retention obligations, so long as they respect the essence of the right and are necessary, appropriate and proportionate. In Tele2 and Watson the CJEU noted that any limitation of rights recognised by the CFR must respect the essence of said rights [94]. The CJEU accepted the Advocate General (AG)’s Opinion that data retention creates an equally serious interference as interception and that the risks associated with the access to communications maybe greater than access to the content of communications [99]. Yet the CJEU were reluctant to hold that data retention (and access to) adversely affects the essence of those rights [101]. This appears to highlight a problem in the CJEU’s reasoning, if the CJEU, like the AG accept that retention of and access to communications data is at least on par with access to the content, it makes little sense to then be reluctant to hold that data retention adversely affects the essence of those rights. The CJEU does so without making any distinction or reasoning for this differential treatment, and thus serves to highlight that perhaps the CJEU themselves do not fully respect the essence of those rights in the context of data retention.

The CJEU’s answer seems only limited catch all powers

The thrust of the CJEU’s judgment in Tele2 and Watson was that general and indiscriminate data retention obligations are prohibited at an EU level. But as I have highlighted previously, the CJEU’s answer was only in response to a very broad question from Sweden, which asked was:

[A] general obligation to retain traffic data covering all persons, all means of electronic communication and all traffic data without any distinctions, limitations or exceptions for the purpose of combating crime…compatible with [EU law]?

Therefore, provided that national laws do not provide for the capturing of all data of all subscribers and users for all services in one fell swoop, this may be argued to be compatible with EU law. Both the e-Privacy Regs and the CJEU refer to ‘targeted’ retention [108, 113]. The CJEU gave an example of geographical criterions for retention in which David Anderson Q.C. asks whether the CJEU meant that ‘it could be acceptable to perform “general and indiscriminate retention” of data generated by persons living in a particular town, or housing estate, whereas it would not be acceptable to retain the data of persons living elsewhere? This is entirely possible given the reference from Sweden and the answer from the CJEU. In essence the CJEU have permitted discriminatory general and indiscriminate data retention which would in any event respect the essence of those rights.

Data retention is our cake, and only we can eat it

A final point on Tele2 and Watson was that the CJEU held that national laws on data retention are within the scope of EU law [81]. This by itself may not raise any concerns about protecting fundamental rights, but it is what the CJEU rules later on in the judgment that may be of concern. The CJEU held that the interpretation of the e-Privacy Directive (and therefore national Member State data retention laws) “must be undertaken solely in the light of the fundamental rights guaranteed by the Charter” [128]. The CJEU has seemingly given itself exclusive competence to determine how rights are best protected in the field of data retention. It is clear from the subsequent paragraph that the CJEU seeks to protect the autonomy of EU law above anything else, even fundamental rights [129]. This is despite the ECHR forming general principles of EU law and is mentioned in Article 15(1) (refers Article 6(3) of the Treaty of the European Union (TEU) specifically referring to the ECHR as such). Article 11 of the e-Privacy Regs refers to restrictions respecting the ‘essence of fundamental rights and freedoms’ and only time will tell whether the CJEU would interpret this as only referring to the CFR. Recital 27 of the e-Privacy Regs just like Recital 10 and 30 of the e-Privacy Directive refers to compliance with the ECHR, but as highlighted previously, Recitals are not legally binding.

Is the CJEU assuming too much?

A further concern, is that had the European Commission added general principles of EU law into Article 11, the CJEU may simply have ignored it, just as it has done in Tele2 and Watson. The problem with the CJEU’s approach is that it assumes that this judgment offers an adequate protection of human rights in this context. The ECHR has always been the minimum floor, but it appears the CJEU wants the CFR to be the ceiling whether it be national human rights protection, or protection guaranteed by the ECHR. What if that ceiling is lower than the floor? The AG in Tele2 and Watson stressed that the CFR must never be inferior to the ECHR [141]. But I have argued before, the EU jurisprudence on data retention is just that, offering inferior protection to the ECHR, and the qualification by the CJEU in Tele2 and Watson does not alter this. This position is strengthened by Judge Pinto De Albuquerque in his concurring opinion in the European Court of Human Rights judgment in Szabo. He believed that:

[M]andatory third-party data retention, whereby Governments require telephone companies and Internet service providers to store metadata about their customers’ communications and location for subsequent law-enforcement and intelligence agency access, appeared neither necessary nor proportionate [6].

Of course, Judge Pinto De Albuquerque could have been referring to the type of third party data retention which requires Internet Service Providers (ISPs) to intercept data from Over The Top (OTT) services, but his description is more in line with data retention of services’ own users and subscribers.


Although the CJEU has prohibited general indiscriminate data retention, the CJEU does not seem to have prevented targeted indiscriminate data retention. If the European Court of Human Rights (ECtHR) were to ever rule on data retention and follow its jurisprudence and the opinion of Judge Pinto De Albuquerque, this may put EU law in violation of the ECHR. This would ultimately put Member States in a damned if they do, damned if they do not situation, comply with the ECHR, and violate EU law autonomy; comply with EU law and violate the ECHR. When the minimum standards of human rights protection in this context are not adhered to, because of EU law, the ECHR should prevail. As anything less is a threat to human rights, meaning that the (even if well intentioned) CJEU can also be.

JHA4: chapter II:7

Photo credit: goldenfrog.com

Expulsion of seriously ill migrants: a new ECtHR ruling reshapes ECHR and EU law

Dr Lourdes Peroni*, Postdoctoral Research Fellow, Ghent University Human Rights Centre (ECHR aspects) and Professor Steve Peers (EU law aspects)

In what is possibly one of the most important judgments of 2016, Paposhvili v. Belgium, the Grand Chamber of the European Court of Human Rights (ECtHR) has memorably reshaped its case law on when Article 3 ECHR (which bans torture or other inhuman or degrading treatment) applies to the expulsion of seriously ill migrants. In a unanimous judgment, the Court leaves behind the restrictive application of the high Article 3 threshold set in N. v. the United Kingdom and pushes for a more rigorous assessment of the risk of ill-treatment in these cases. For us at the Human Rights Centre of Ghent University, it was a thrill to intervene as a third party in such an important case. In our third party intervention we submitted that Paposhvili offered a unique opportunity to depart from the excessively restrictive approach adopted in N. We are delighted that the Grand Chamber has seized the opportunity to re-draw the standards in this area of its case law in a way that does fuller justice to the spirit of Article 3.

This main part of the post addresses the ECtHR’s interpretation of the ECHR in Paposhvili, while in the Annex to this post, Steve Peers considers its application within the scope of EU law.

The ECHR judgment

Mr. Paposhvili, a Georgian national living in Belgium, was seriously ill. He claimed that his expulsion to Georgia would put him at risk of inhuman treatment and an earlier death due to the withdrawal of the treatment he had been receiving in Belgium (for more on the facts, see my previous post). He died in Belgium last June, while his case was pending before the Grand Chamber. The Court did not strike his application out of the list. It found that “special circumstances relating to respect for human rights” required its continued examination based on Article 37 § 1 in fine ECHR (§ 133). The Court held that there would have been a violation of Article 3 if Belgium had expelled Mr. Paposhvili to Georgia without having assessed “the risk faced by him in the light of the information concerning his state of health and the existence of appropriate treatment in Georgia.” It found a similar violation of Article 8 if Belgium had expelled him without having assessed the impact of his return on his “right to respect for his family life in view of his state of health.”

Opening Up “Other Very Exceptional Cases”

The Chamber judgment in Paposhvili followed N. and Yoh-Ekale Mwanje v. Belgium where the Court had taken into account that “the applicants’ condition had been stable as a result of the treatment they had been receiving, that they were not ‘critically ill’ and that they were fit to travel” (§ 119). The Chamber thus concluded that though Mr. Paposhvili suffered from “a fatal and incurable disease … his conditions are all stable and under control at present; his life is therefore not in imminent danger and he is able to travel” (§ 120).

As readers might remember, the N. Grand Chamber established that removing a non-national suffering from a serious illness to “a country where the facilities for the treatment of that illness are inferior to those available in the Contracting State may raise an issue under Article 3, but only in a very exceptional case” (§ 42). The Grand Chamber concluded that the applicant’s circumstances in N. were not exceptional, as found in D. v. United Kingdom (§ 42). D was critically ill, close to death, and had no prospect of medical care and family support in his home country. The N. Grand Chamber, however, left a window open: it did not exclude that “there may be other very exceptional cases where the humanitarian considerations are equally compelling” (§ 43, emphasis added).

In our third party intervention, we argued that being medically stable and fit to travel as a result of the treatment received should not be a determining criterion in allowing an expulsion. We respectfully invited the Court to develop a less extreme approach, one that considered the difference between applicants’ suffering in the sending state and the suffering they would face in the receiving state. The aim, we submitted, should be to determine whether the reduction of applicants’ life expectancy and the deterioration of their quality of life would be such as to reach the level of severity required by Article 3. The applicant argued that his expulsion to Georgia would place him at risk of “a severe and rapid deterioration in his state of health leading to his swift and certain death” (§ 148). He asked the Court “to go beyond its findings in N. v. the United Kingdom” and to define “a realistic threshold of severity that was no longer confined to securing a ‘right to die with dignity’” (§ 149).

The Paposhvili Grand Chamber enters through the window N. left open. It notes that since N. no other “very exceptional cases” had been found (§ 178). It importantly recognizes that the application of Article 3 only to persons close to death has deprived those whose condition was less critical but who were still seriously ill from “the benefit of that provision” (§ 181). In a pivotal paragraph, the Grand Chamber considers

… that the “other very exceptional cases” within the meaning of the judgment in N. v. the United Kingdom (§ 43) which may raise an issue under Article 3 should be understood to refer to situations involving the removal of a seriously ill person in which substantial grounds have been shown for believing that he or she, although not at imminent risk of dying, would face a real risk, on account of the absence of appropriate treatment in the receiving country or the lack of access to such treatment, of being exposed to a serious, rapid and irreversible decline in his or her state of health resulting in intense suffering or to a significant reduction in life expectancy. The Court points out that these situations correspond to a high threshold for the application of Article 3 of the Convention in cases concerning the removal of aliens suffering from serious illness (§ 183). Emphasis added.

This is a graceful move that softens the unduly restrictive approach that had so far been followed in cases concerning the expulsion of seriously ill migrants. Paposhvili thus comes to fill what Judge Lemmens calls a “gap in the protection against inhuman treatment” (concurring opinion in Paposhvili § 3) by including as exceptional more than just cases of imminent death. My first impression is that the Court does not formally leave behind N.’s exceptional character and the high threshold of Article 3 in cases concerning the expulsion of seriously ill non-nationals (see last sentence § 183 and Judge Lemmens’ opinion § 3). Rather, it appears to open up what in practice has resulted in a limited application of the high threshold. The commendable effect of the Court’s move is, in any event, a less extreme approach more compatible with the spirit of Article 3. Elements of both our third party intervention and the applicant’s arguments are reflected positively in the Grand Chamber reasoning in this regard.

Real Rather Than Theoretical Access to “Sufficient” and “Appropriate” Care

In our third party intervention we proposed that the risk assessment should consider the adequacy of the medical care available in the receiving state and the person’s actual access to such care. The question, we argued, is not just whether adequate treatment is generally available but, crucially, whether the available treatment would in reality be accessible to the person concerned. The applicant argued that the alleged Article 3 violation should be examined “in concreto,” taking into consideration, among other things, “the accessibility of treatment in the country of destination” (§ 139).

The Grand Chamber seizes the occasion to meticulously set out a range of procedural duties for the domestic authorities in the ECHR state parties. All these duties point in one clear direction: a more rigorous assessment of the risk as required by the absolute nature of the Article 3 prohibition (Saadi v. Italy § 128). In assessing the alleged risk of ill-treatment, the domestic authorities should verify whether the care available in the receiving state is “sufficient and appropriate in practice for the treatment of the applicant’s illness so as to prevent him or her being exposed to treatment contrary to Article 3” (§ 189, emphasis added). The domestic authorities should also consider “the extent to which the individual in question will actually have access to this care and these facilities in the receiving State” (§ 190, emphasis added). Referring to existing case law, the Court points to several factors to be taken into account: “cost of medication and treatment, the existence of a social and family network, and the distance to be travelled in order to have access to the required care” (§ 190).

Duty to Obtain Assurances from the Receiving State

With reference to Tarakhel (a 2014 ECtHR ruling on the application of the EU’s Dublin rules on allocation of asylum responsibility), our third party intervention proposed that Article 3 impose on the domestic authorities in the returning state the procedural duty to seek or obtain assurances from the receiving state that the person concerned would actually have access to the treatment s/he needed. We argued that access to appropriate medical care should not be a theoretical option, but a real and guaranteed one, and the burden of proving that such a real option exists should lie on the expelling state (on assurances and the benefits of adopting this path, see Eva Brems’ commentary on Tatar v. Switzerland).

On this point, the Grand Chamber states in paragraph 191:

Where, after the relevant information has been examined, serious doubts persist regarding the impact of removal on the persons concerned – on account of the general situation in the receiving country and/or their individual situation – the returning State must obtain individual and sufficient assurances from the receiving State, as a precondition for removal, that appropriate treatment will be available and accessible to the persons concerned so that they do not find themselves in a situation contrary to Article 3 (on the subject of individual assurances, see Tarakhel, cited above, § 120).


There is so much more to say about the Court’s reasoning in Paposhvili. I have highlighted some of its most remarkable Article 3 principles. Together with others, such as the one establishing when the responsibility of the returning state is engaged (§ 192), these principles firmly move a body of the Court’s case law closer to its principles on the absolute nature of the Article 3 prohibition.

*This part of the post is reblogged with permission from the Strasbourg Observers blog

Barnard & Peers: chapter 26
JHA4: chapter I:7
Photo: OLV hospital, Belgium
Photo credit: Sapa group

Annex: the impact on EU law

By Professor Steve Peers

How does this judgment impact upon EU law?

First of all, it’s necessary to explain the existing EU law position, set in the Abdida and M’Bodj judgments of the ECJ, which was referred to in the ECtHR judgment (paras 120-22), and which I discussed further here. In short, ‘medical cases’ are not within the scope of EU asylum law, either as regards refugee status or subsidiary protection (M’Bodj). However, if the person concerned faces an expulsion order, then the Returns Directive applies. (Note that the latter Directive doesn’t apply to the UK, Ireland or Denmark.)

Although the Returns Directive was mainly intended to ensure removal of irregular migrants from the territory, in ‘medical cases’ (at least), as interpreted by the ECJ in Abdida, it has the opposite effect. According to the Court, the requirement in Article 5 of the Directive to ‘respect the principle of’ non-refoulement means that irregular migrants who fall outside the scope of EU asylum law but nevertheless face an Article 3 ECHR risk, as defined in the case law of the ECtHR, cannot be removed. Moreover, in further displays of legal alchemy, the ECJ ruled that the challenge to their removal must have suspensive effect, and they must receive the necessary health care and social benefits.  

The ECJ has not developed this case law since, although further relevant cases are pending. In MP, the Court has been asked to clarify the line between asylum cases and medical cases, where the medical conditions are more directly linked to persecution or serious harm suffered in the country of origin. In Gnandi, it has been asked to clarify the suspensive effect of a legal challenge in medical cases, following a failed asylum application. In K.A. and others, the Court has been asked about the requirement to ‘take due account’ of family life in Article 5 of the Returns Directive; its ultimate ruling might be relevant to the ‘non-refoulement’ aspect of the same clause by analogy. Equally in Nianga the Court has been asked whether Article 5 applies to the decision to issue a return decision or removal order in the first place: a crucial point because if it does not apply, the person concerned might well fall outside the scope of EU law entirely.  

What impact will the new ECtHR ruling have on the interpretation of EU law? First of all, there’s nothing to suggest it will, by itself, move the dividing line between asylum cases and medical cases, as applied by the ECJ. So we are still looking at the interpretation of the Returns Directive, if that Directive applies.

Since the ECJ committed itself to follow the case-law of the ECtHR as regards medical cases when interpreting the non-refoulement provision of the Returns Directive, it should follow that the new ECtHR ruling applies to the Directive too. Therefore this enlarges the group of people who can benefit from the specific provisions of EU law as interpreted by the ECJ, as regards suspensive effect of appeals and access to health care and social benefits.

Equally the ECtHR’s strong stress on the procedural elements of such cases logically applies by analogy to cases falling within the scope of the Returns Directive. While the ECJ in the Abdida judgment did not refer to its own jurisprudence on the right to a hearing for irregular migrants (discussed here), it is now necessary to update that approach in light of the ECtHR ruling, given the strong link which the latter judgment establishes between the procedural and substantive aspects of what I have referred to as ‘alternative protection’. The ECJ will have an opportunity to address this issue in the months to come, in the pending cases referred to above.

While the ECtHR judgment referred to a need to cooperate with the country of origin in order to check conditions there, in the EU context this might arguably in some cases entail by analogy a check on health conditions in another Member State, which would be responsible for that person under the Dublin rules. The ECJ has yet to determine how its interpretation of the Returns Directive in medical cases fits together with the application of the Dublin rules, which in principle apply if the person concerned has at one point applied for international protection (refugee status or subsidiary protection) within the EU. (Mr. Paposhvili was originally subject to the Dublin rules, but it seems that the plan to remove him to Italy pursuant to those rules petered out).  

Finally, it should be noted that the ECtHR also found a breach of Article 8 ECHR (the right to family life), on similar procedural grounds. This might be relevant to interpretation of the EU’s family reunion Directive, for those who fall within the scope of that Directive and who argue on the basis of the factors to consider during expulsion proceedings pursuant to Articles 17 and 18 of that law.

Sunday, 8 January 2017

What is ‘free trade’?

Professor Stephen Weatherill, Somerville College Oxford University

On 3 January 2017 Sir Ivan Rogers resigned as the UK’s Permanent Representative to the EU. He wrote a widely publicised letter of resignation. Among several incendiary observations directed at the government’s perceived unpreparedness for negotiations on Brexit perhaps the most headline-grabbing was:

‘Contrary to the beliefs of some, free trade does not just happen when it is not thwarted by authorities: increasing market access to other markets and consumer choice in our own, depends on the deals, multilateral, plurilateral and bilateral that we strike, and the terms that we agree’.

The use of a double negative in the opening salvo is a little clumsy, but the gist is clear. Removing obstacles created by authorities does not automatically release free trade. More is needed. Dealmaking is needed. And this, Sir Ivan stings, is contrary to the beliefs of some. Who are these ‘some’?

David Davis and Boris Johnson, possibly. Liam Fox, probably.

The International Trade Secretary delivered a speech in Manchester on 29 September 2016 in which, having begun with reference to Adam Smith’s The Wealth of Nations (1776), he expressed a desire to ‘remake the intellectual and philosophical case for free trade’. The speech is peppered with praise for free trade. But Fox never defines what he means by free trade. His speech to the Conservative Party Conference in Birmingham a week later is equally barren. The closest he comes to definitional precision in the speech delivered in Manchester is to observe, citing Adam Smith for the claim that ‘it is a moral right for people to buy whatever they want from those who sell it to them the cheapest’, that the idea that ‘governments should restrict the right of individuals to exchange their hard work for goods and services at an agreed price in an open market is one of the gravest infringements of personal liberty I can think of’. And he drew on the repeal of the Corn laws during Victorian times as a demonstration of the virtuous release of price competition to the benefit of consumers.

Fox’s vision demands that the State keep out of private transactions. The free trade which he wishes to champion is in truth unregulated trade.

But free trade in this form does not exist. Governments intervene in markets for myriad reasons and in myriad ways, and they have been doing so for a very long time. In England the composition of ale and bread has been the subject of regulation for centuries. Consumers cannot know for sure that the products on offer in the marketplace are wholesome, so the State intervenes. These are the earliest forms of modern consumer protection law, and they reflect an understanding that leaving the market unregulated will cause it to malfunction where the consumer does not possess adequate information to distinguish between products according to their quality. Ale and bread then – videogames and i-phones today, which come with statutory guarantees of quality. In similar vein the nature of the relationship between employer and employee is not simply a matter of private negotiation: State regulation supplies a floor of protection according to an assumption that, without it, the employee is vulnerable to exploitation or unfair treatment. The relationship between a landlord and a tenant is regulated for similar reasons. Markets may also malfunction on the supply-side. Competition law, in its common law form of restraint of trade and more recently in statutory guise, places restraints on the conduct of traders because of fear that a market left unregulated may become contaminated by anti-competitive agreements or the abuse of monopoly power. As Adam Smith himself remarked in The Wealth of Nations, ‘People of the same trade seldom meet together even for merriment and diversion, but the conversation ends in a conspiracy against the public, or in some contrivance to raise prices’.

Markets are neither free-standing nor inevitably self-correcting. Sometimes they require protection. So markets are and long have been built on the correcting influence of public regulation. Of course one might argue for less or for more public regulation of the market, but to speak of ‘free trade’ in a market as if it operates with no public regulation is deeply misleading and lacks historical context. For every perniciously protectionist rule such as the Corn laws there are hundreds of virtuous measures of consumer, worker or environmental protection which reflect the limits of the capacity of unregulated markets to address the pressing social concerns of the time.

This readily translates to the transnational plane. ‘Free trade’ across State borders cannot possibly entail a process of private traders arriving in foreign lands and striking deals with local buyers without regard for local laws. Those traders will be required to comply with the local rules that govern the operation of markets – rules of consumer protection, environmental protection, competition law and so on. And those rules will doubtless be different from those that apply at home: cultural specificity and historical accident dictate that regulatory diversity between States is the norm. These are non-tariff barriers to trade. They mean that trade is not ‘free’.

There are plenty of ways for States to get together in order to address such impediments to cross-border trade. In the abstract there are two extremes. One is to decide that regulatory diversity should be ignored: a product or service that is good enough for one market should be treated as good enough for another. The other is to replace that regulatory diversity by introducing rules which apply in common.  The first model directs that States may not exclude products and services even if they fall below their own locally determined preferences. This therefore involves States relinquishing regulatory authority, and in effect placing their market under the jurisdiction of a more lenient regulator in another State. This raises obviously sensitive issues associated with accountability and control. The second model requires the creation of a common rule-making body. Here too States relinquish regulatory authority, but to a new supra-State authority charged with the responsibility to select the applicable rule. Here too loom – different – issues associated with accountability and control. But neither model suggests unregulated trade. Quite the contrary. Public regulation of one kind or another is necessary to promote an integrated trading space that spans jurisdictions which are marked by varying patterns of intervention in the market.

In practice models of economic integration that fall within the two extremes are found. States sometimes accept that national measures that impede trade shall be subject to some form of review: they are typically not automatically set aside but nor, in the cause of economic integration, are they jealously guarded as inviolable expressions of local autonomy. In this vein the rules of the WTO assert that national measures that obstruct cross-border trade may be subject to review. The detail need not detain us: the point of present relevance is that there is no question of cross-border trade proceeding between private parties without reference to public regulation. The material scope of the WTO regime is limited and in any event there is room to justify national measures of regulation even where they do impede inter-State trade, for example for reasons associated with the protection of public health. 

This is remote from Dr Fox’s superficial model of ‘free trade’. Free trade agreements are recognised under WTO law and they represent a deeper commitment to realising the productive energy of cross-border trade. There are myriad versions but the free trade envisaged is not unregulated trade. Free Trade Agreements aim to achieve a freeing of trade, but they do not envisage a wholesale ejection of public regulation from the market. The case of the EU is more complex again. In the EU the extent to which control is exercised over State regulatory autonomy in so far as it obstructs inter-State trade goes still deeper than that exercised through the WTO – this is the entrancing story of Cassis de Dijon laced by ambiguities such as Keck. This is the promotion of deregulated trade within the EU but, given the vital space permitted to States to justify national measures which obstruct inter-State trade on grounds of health protection and the public interest more generally, it is certainly not a charter for unregulated trade. Moreover, the EU possesses important legislative competences which supplement the free movement rules as a means to open up the internal market. This covers legislative harmonisation supported by sector-specific rule-making activity in areas such as social policy and environmental protection. This programme of common rulemaking is designed to free trade – but not to leave it unregulated. In fact the EU, when it legislates, acts to regulate trade, according to common patterns: better, the EU re-regulates trade, in replacement for pre-existing and diverse State regulation. EU law promotes cross-border private contracting within the internal market but not on terms that exclude public regulation.

To return to Sir Ivan Rogers– ‘free trade does not just happen’. In the transnational sphere, there is no free trade, there is only freed trade and it is regulated trade. The extent to which it is freed and the terms according to which it is regulated depends on the deals struck and the enforcement mechanisms created in their support. It requires transnational negotiation, design of rules and of institutions. It – again to return to Sir Ivan – ‘depends on the deals, multilateral, plurilateral and bilateral that we strike, and the terms that we agree’. The people of the UK have voted to quit the EU and that will lead also to the UK falling out of the scope of the EU’s several dozen free trade agreements with third countries. That will not grant the UK free trade. It will diminish the UK’s enjoyment of freed trade. And it will demand that the UK does a great deal of dealmaking even to begin to replace what it has lost. In Victorian times, to which Dr Fox pays much wistful attention in his Manchester speech, the UK’s economic and political strength allowed it to swagger its way to oceans of beneficial trading activity. The balances of power are different today. It is not conceivable that post-Brexit the UK will get anything other than a worse deal with the EU-27 than it enjoys currently, given that the percentage of total export trade which the UK does with the EU-27 is so many times higher than that which the EU-27 does with the UK. The UK needs the EU a great deal more than the EU needs the UK. And equally it is not conceivable that post-Brexit the UK will get anything other than a worse deal than it enjoys currently with trading partners elsewhere in the world, given that the UK, population 64 million, has so much less clout and so much less to offer than the EU-27, population 440 million.

Trade today cannot be effectively promoted by unilateral action. Dealmaking is required. And that requires concessions. As the House of Lords European Union Committee felicitously put it in its December 2016 report on Brexit: the Options for Trade, ‘there is always an inherent trade-off between liberalising trade and the exercise of sovereignty’ (page 3). It adds that as a general rule ‘the deeper trade relationship, the greater the loss of sovereignty’ (page 76). I would treat ‘sovereignty’ with as much suspicion as I treat ‘free trade’: as a label, it is at best unhelpfully imprecise, at worst an anachronism. Better to frame the discussion in terms of power in practice as distinct from power in principle. In the current conditions of interdependence among States in Europe the State that insists on exercising its power unilaterally may pride itself on its adherence to principle but it will find that in practice its ability to address problems that spill over borders – climate change, migration, trade, and so on – is seriously diminished. It needs to co-operate with other States to find solutions. That co-operation extends its power in practice. All States gain from agreeing to be locked into a mutually agreed framework for addressing problems: each gives up a degree of autonomy in principle but in return knows that all other participants have made precisely the same concession.

Voting rules are vividly emblematic of the trade-offs at stake. A rule of unanimity preserves a veto but it is one that is held by all members of the bloc, so that difficult decisions are unlikely to be taken: this is to privilege ‘sovereignty’ in principle over the facilitation of practical dealmaking. Embrace of majority voting acquiesces in the possibility of being outvoted in return for a power also to outvote: deals will be struck more readily, though on occasion under sufferance. Voting rules are one of the design choices that have to be made by co-operating States. So engaging in multilateral trade deals typically involves some degree of self-restraint measured in commitments to comply with binding norms located at the transnational level – promises not to obstruct trade (typically except in defined circumstances), promises to apply rules agreed in common. This is to yield power (to act unilaterally) in principle yet it enhances power in practice. ‘Free trade’ is really about regulated trade – negotiating the terms of ‘free trade’ is really about negotiating the patterns of regulation that will provide the foundation for trade.

This is ‘free trade’ in a modern world of densely regulated markets and unavoidably interdependent States. It is an agenda of multilateralism. To suppose that free trade simply happens when governments get out of the way is an exercise in evading complexity.

Seen from the outside one has the increasing impression that those who drove the people of the UK to vote for Brexit and who are now in charge of plotting the future do not even understand the first thing about what ‘free trade’ means today, in the EU or more generally. Sir Ivan’s comments suggest that that is what it looks like from the inside too.

Barnard & Peers, European Union Law: chapter 27, chapter 11
Photo: Adam Smith

Photo credit: Adam Smith Institute, www.adamsmith.org