The EU's future trade policy starts to take shape: the Opinion on the EU/Singapore FTA

Professor Steve Peers

What is the scope of the EU’s powers over trade agreements? The issue has been disputed for decades in the case law of the ECJ, for it has a significant impact on the allocation of powers between the EU and its Member States as regards external economic policies. A number of Treaty amendments over the years – in particular the Treaty of Lisbon – have amended the rules.

The issue has gained added salience given the controversies surrounding some EU trade negotiations (in particular with Canada and the USA), and the trade talks between the UK and EU in light of Brexit. Today’s opinion of an ECJ Advocate-General is not binding, but is very thorough and will likely have a significant impact on the Court’s final judgment, expected in the spring.

This post will summarise the lengthy opinion succinctly and suggest its likely implications for the FTAs with Canada, the USA and the UK in particular. For further reading, see the earlier posts on this blog on the background to the Opinion and on the hearing before the ECJ.  

Background

The Court has been asked to rule on whether the various provisions of the EU’s draft trade deal with Singapore fall within the scope of the EU’s exclusive powers, or whether powers are shared with the Member States, or whether only Member States can conclude them. If the EU only can conclude them, there can be no national ratification and also probably (depending on the exact content of the agreement) the EU will approve the deal by qualified majority, ie Member States will not have a veto.

If both the EU and its Member States can conclude the provisions, the agreement is ‘mixed’, but the EU has a choice to conclude the agreement without the Member States, if a qualified majority (assuming, again, that no veto applies due to the subject matter) agree to this.

If an issue is within exclusive Member State competence, then Member States must be parties to the treaty in order to conclude it. National ratification, and a de facto national veto for each Member State, therefore applies.

When is a power exclusive to the EU? Article 3(1) of the Treaty on the Functioning of the European Union (TFEU) lists a number of powers that are inherently exclusive, including the common commercial (ie trade) policy (CCP) and fisheries conservation. The CCP is further defined in Article 207 TFEU: it particularly applies to ‘goods and services’, the commercial aspects of intellectual property’ and ‘foreign direct investment.’ The EU/Singapore case concerns the interpretation of each of these aspects.

Besides Article 3(1), Article 3(2) TFEU goes on to provide that exclusive EU powers over an international treaty can also derive from the exercise of EU internal powers, in three cases: (a) ‘its conclusion is provided for in a legislative act of the Union’ or (b) it ‘is necessary to enable the Union to exercise its internal competence’, or (c) ‘in so far as its conclusion may affect common rules or alter their scope.’ The EU/Singapore case concerns the interpretation of both (a) and (c), which I will refer to as the ‘legislative authorisation’ ground and the ‘affect common rules’ ground.  (Note that ground (b) is rarely applied, as the ECJ case law interprets it very narrowly).

Summary of the opinion

The Commission argues that the EU has exclusive competence to conclude the deal. It’s supported by the European Parliament, which will have the power to consent to the deal as long as part of it relates to the CCP, or indeed to most other EU powers. Member States argue for mixed competence of much of the agreement, and exclusive national competence for some parts of it.

In general, the Advocate-General argues that much of the agreement is solely within the EU’s exclusive powers, mostly (but not entirely) as part of the CCP. A significant part falls within the EU’s mixed competence, while a small part is purely national competence.

First of all, she makes some general points about the scope of the CCP. She restates prior ECJ case law: the CCP applies to a measure which regulates and has direct effect on trade; mere implications for trade are not sufficient. She also interprets the exceptions in Article 207(6) TFEU, which states that the CCP ‘shall not affect the delimitation of competences between the Union and the Member States, and shall not lead to harmonisation of legislative or regulatory provisions of the Member States in so far as the Treaties exclude such harmonisation.’ In her view, this clause must be narrowly interpreted and has limited effect: for instance, it does not restrict the EU from agreeing measures on trade in culture and health services, as long as it does not harmonise the laws on those issues within the EU.

The opinion does not address the potentially important exceptions in Article 207(4) TFEU, which call for unanimous voting where ‘unanimity is required for the adoption of internal rules’ or ‘(a) in the field of trade in cultural and audiovisual services, where these agreements risk prejudicing the Union's cultural and linguistic diversity’, or ‘(b) in the field of trade in social, education and health services, where these agreements risk seriously disturbing the national organisation of such services and prejudicing the responsibility of Member States to deliver them.’

On the other hand, the opinion does discuss the exception in Article 207(5) TFEU, which states that the CCP does not apply to agreements concerning transport. As a general rule, the Advocate-General argues that this exception applies whenever a treaty has rules ‘specifically concerning transport’. The further implications of this are discussed below.  

The Opinion then examines the specific provisions of the EU/Singapore deal. First of all, the opening provisions of the FTA, referring to the creation of a free trade area, fall within the scope of the CCP. Next, following pre-Lisbon case law, the Opinion concludes that the FTA provisions on trade in goods are also within the scope of the CCP (Paras 144-155).

Thirdly, the Opinion examines the FTA provisions on services, establishment and e-commerce (paras 195-269). In general, other than transport issues, these fall within the scope of the CCP powers over services. In particular, immigration of service providers falls within the scope of the services powers, and therefore not under the immigration powers of the EU, where the UK and some other Member States have an opt-out (para 203). Financial services are covered by the CCP (para 204), since its scope is not dependent on prior harmonisation of the relevant law by the EU (unlike Article 3(2) TFEU). Professional qualifications are also covered (para 205).

As for the transport exception from the CCP, it applies not just to the services themselves, but those indissolubly linked to those services – ie cargo handling, transport repair, and computer reservation – but not to customs clearance, since that applies also to trade in goods.  But does the EU have exclusive power over the transport issues, by applying Article 3(2) TFEU instead? As regards aircraft repair, the ‘legislative authorisation’ ground doesn’t apply, since the EU legislation creating an aircraft safety agency doesn’t address this issue in detail. As for the ‘affect common rules’ ground, there is insufficient internal harmonisation as maritime transport, air transport (other than computer reservation systems), and inland waterways – but sufficient internal harmonisation as regards road and rail transport for the powers to become exclusive as regards the EU/Singapore FTA. Other aspects of transport remain a shared competence.

Fourthly, on the issue of investment (paras 305-398), the opinion again examines both the CCP and Article 3(2) TFEU. The opinion offers a definition of the EU’s CCP powers over foreign direct investment: investments ‘which serve to establish or maintain lasting and direct links, in the form of effective participation in the company’s management and control, between the person providing the investment and the company to which that investment is made available in order to carry out an economic activity. In applying that definition, I consider that the fact that the direct investor owns at least 10% of the voting power of the direct investment enterprise may offer evidentiary guidance but is certainly not determinative’. Crucially, the opinion argues (paras 324-342) that the CCP power covers the issue of investor protection.

As for other forms of investment – referred to as ‘portfolio investment’, it was agreed that the CCP didn’t apply. Could Article 3(2) TFEU apply, though? Here, there was no legislation on the issue, but there are EU Treaty provisions on capital movements to non-EU countries, which the Commission believes fall within the scope of the ‘affect common rules’ ground. However, the Opinion argues in principle that this ground for exclusive competence can only apply where the prior EU harmonisation results from legislation, not the Treaty. But the EU and its Member States still shared competence on most investment issues, except for the termination of bilateral investment treaties.

Fifth, on the issue of government procurement, previous prior case law said that the CCP only applied to procurement relating to goods and limited aspects of services. The Opinion concludes that in light of the Lisbon Treaty provisions made to the scope of the CCP, that EU power now fully applies to government procurement issues – other than those within the scope of the transport exception (paras 401-408).

Sixth, the Opinion examines the scope of the CCP power relating to intellectual property (paras 424-456). Although prior case law had concluded that the CCP fully applied to the ‘TRIPS’ (ie the intellectual property deal forming part of the World Trade Organisation system), the Opinion argues that this ruling did not necessarily apply by analogy to intellectual property rules in the EU’s FTAs (IP rules found in FTAs are often called ‘TRIPS+’ clauses).

To determine if a TRIPS+ clause falls within the scope of the CCP, the test (para 435) is not based on the remedy which applies, but rather whether: the substantive obligation governs trade rather than harmonises IP law; there is a direct and immediate effect on trade; and if the measure aims to avoid distortions to trade caused by monopolies. Again, application of the CCP does not depend on whether the EU has harmonised an IP issue internally. The Opinion also argues that rules on court procedures do not necessarily fall outside the scope of the CCP.

Appling this test to the facts: enforcement and plant variety rights are part of the CCP, but some parts of the draft EU/Singapore are not: namely moral rights, which also are not covered by Article 3(2) because the EU has not harmonised them internally. But the EU does have shared competence over this issue, since it could harmonise them on the basis of its internal market powers.

Seventh, the Opinion looks at competition law (paras 459-466). The FTA rules on this issue fall within the scope of the CCP, since they extend EU rules to Singapore and there is a a strong link with trade in goods and services.

Eighth, the Opinion looks at the FTA provisions on environment and sustainable development (from para 478). Here the rules on renewable energy fall within the scope of the CCP, since there is a strong link to trade and investment. However, the rules on labour and environmental standards are not closely linked with trade, so the EU shares competence with its Member States (no one had made an argument that Article 3(2) applied). The rules on fish stocks fell within the scope of another EU exclusive competence: fisheries conservation.

Finally, the rules on transparency and judicial review were ancillary to the substantive provisions of the FTA (paras 508-13). So were the rules on dispute settlement and mediation (paras 523-44); here the Opinion points out that the controversial rules on investor-state dispute settlement were not at issue in this case (para 536). (Note that Belgium has promised to ask the Court about the relevant provisions in the EU/Canada FTA). And the final provisions are either accessory or minor, so change none of the legal assessment (paras 548-553).

Comments

The Advocate-General’s analysis as regards goods, services and intellectual property is unsurprising in light of prior case law. However, the analysis as regards the fresh issue of investment is more disputable. Her case that investor protection falls within the scope of the CCP is convincing, on the grounds that people might not invest in the first place without adequate protection (ie, there is a link back to market access). On the other hand, the analysis relating to portfolio investment puts form over substance: why should it matter that ‘common rules’ derive from the Treaties, rather than EU legislation? Also, the termination of bilateral investment treaties should more logically be seen as the corollary of the exercise of the EU’s other (exclusive or shared) competence, rather than a purely national competence. And it is unfortunate that the Commission missed this opportunity to ask the Court to rule already on the controversial investor-state dispute settlement rules.

What are the implications for other FTAs, and for Brexit? That depends in part on the exact commitments in those other treaties, since this Opinion analyses the commitments that would be made under the EU/Singapore FTA, and commitments under other treaties might differ. In particular, it’s conceivable that other FTAs might arguably require unanimity on the basis of Article 207(4) TFEU, discussed above, which was not at issue in this case.

In general, for other FTAs it seems likely that a mixed agreement may be necessary, in light of the interpretation here relating to the transport exception, portfolio investment, and labour and environmental standards. Apart from the question of termination of investment treaties, then, it will be a purely political question whether Member States are content to agree those trade treaties on behalf of the EU alone, or will continue to insist (as they traditionally have done) on Member States being parties as well.

As for a post-Brexit FTA in particular, different issues may arise. The UK and the EU might not have any interest in negotiating measures relating to investment or intellectual property, at least in the form that EU FTAs now address them. So if the UK and EU want to focus on goods and services only, then the EU’s exclusive CCP competence would apply except as regards transport – and the EU often signs separate transport agreements with non-Member States.  It could be argued that a deal might need unanimity on the basis of Article 207(4) TFEU, but the counter-argument is that a post-Brexit trade deal would simply be preserving (some of?) the existing UK market access into the EU, so could not threaten health or audiovisual services.

Even on transport issues, or as regards labour and environmental standards, case law suggests that exclusive competence on the basis of Article 3(2) applies where the EU seeks to extend its own laws to non-EU states. If the UK is willing to sign up to a treaty that preserves market access in return for compliance with EU rules, it would follow that today’s opinion – if followed by the ECJ – has possibly drawn a road map for the negotiation of an agreement based on free trade in goods and services and compliance with selected EU legislation which could avoid national ratification and (depending on the subject matter) national vetoes.

Barnard & Peers: chapter 24, chapter 27

Photo credit: Singapore Hotels and Guide

The Court of Justice and EU Foreign Policy: what jurisdiction should it have?

Luigi Lonardo, PhD student, King’s College London

The second paragraph of Article 24(1) Treaty on the European Union explains that “the Common Foreign and Security Policy (CFSP) is subject to specific rules and procedures”, and ends with the rather explicit sentence “the Court of Justice of the European Union shall not have jurisdiction with respect to these provisions”.

Lawyers are currently discussing whether the sentence “the CJEU shall not have jurisdiction” means “the Court has some jurisdiction”. Seriously. AG Wahl elegantly phrased it this way: “The main question could be framed as follows: does the exclusion from the CJEU’s jurisdiction cover, in principle, all CFSP acts or only certain categories of CFSP acts?” (Case C‑455/14 P H v Council and Commission AG Opinion, Par 52).

The question is of fundamental constitutional importance because an answer will enable lawyers to understand with clarity what EU foreign policy acts are excluded from the Court’s judicial review – a legal issue that the Court has not yet had the opportunity to adjudicate upon. While Art 19 TEU confers on the Court jurisdiction to ensure that in the interpretation and application of the Treaties the law is observed, Article 24, as recalled, introduces an exception. The scope of this exception, however, has not been fixed. In Case C- 658/11 the Court said that the exception “must be interpreted narrowly” because it introduces an exception from a general rule (par 70). In Opinion 2/13 (on ECHR accession), it only concluded, without further specification, that “as EU law now stands, certain acts adopted in the context of the CFSP fall outside the ambit of judicial review by the Court of Justice” (par 252). To further complicate the issue, however, Article 24 TEU also introduces an exception to the exception: the Court has jurisdiction to monitor compliance with Article 40 TEU (the division between foreign policy and other EU measures) and to review the legality of sanctions.

So, when does the Court of Justice of the European Union (CJEU) have jurisdiction? Two cases may offer guidance with respect to this issue. One case, H v Council and Commission, was decided by the Court in July, and another, Rosneft, is currently pending.

H v Council

In H, an Italian magistrate sought annulment, before the General Court (Order in H v Council and Others, T‑271/10), of the decision of a Head of an EU Mission established under CFSP. The contested decision concerned the transfer of H, a seconded Legal Officer of the EU Police Mission in Sarajevo, to the post of Prosecutor in another regional office of the same country. The General Court (GC) held that it lacked jurisdiction to hear the complaint and therefore found that the action was inadmissible. The GC reasoned that the exclusion of jurisdiction under Art 24(1) TEU only encounters two exceptions: monitoring compliance with Article 40 TEU (ie the division of competence between CFSP and non-CFSP external measures) and the review of the legality of sanctions under the second paragraph of Article 275 TFEU.

The General Court took the view that the appellant’s situation did not fall under one of the exceptions to the general rule that EU Courts do not have jurisdiction in CFSP matters (it was not, therefore, one of the two “exceptions to the exception”). The General Court considered that the contested decisions were adopted by the Head of Mission pursuant to powers that had been delegated to him by the Italian authorities. It thus concluded that it was for Italian courts to review the legality of the contested decisions and to hear the action for damages. It finally added that, should the Italian court having jurisdiction consider the contested decisions unlawful, it could make that finding and draw the necessary conclusions, even with respect to the very existence of those decisions.

The applicant appealed the decision before the ECJ. Applicant, Council, and Commission all wanted to set aside the GC’s order, albeit each for different reasons, which will be briefly outlined below with regards to the issue of the extent of the Court’s jurisdiction on CFSP matters.

The position of the Applicant

The Applicant took the view that the exclusion of the Court’s jurisdiction does not cover merely administrative measures (such as the decision at stake in the present case) but only the acts provided for in Article 25 TEU: general guidelines, decisions on actions and positions to be taken by the EU (and implementation thereof), and acts of systemic cooperation between Member States

The position of the Council

Par 32 of the Advocate General opinion explains that “The Council is of the view that the statement of reasons in the order under appeal does contain two legal errors. First, in deciding to relocate H, the Head of Mission did not exercise powers delegated to him by the Member State of origin, but by the competent EU institution (the Council itself). Second, the national court hearing the case does not have the power to annul the act challenged. Nevertheless, those errors do not — in the opinion of the Council — invalidate the conclusion reached by the General Court”

The position of the Commission

The Commission argued that the Court lacks jurisdiction only on acts that are “expression of sovereign foreign policy”, thus leaving the Court empowered, for example, to review the lawfulness of (a) acts of implementation, or (b) adopted in the framework of the CFSP when the alleged invalidity stems from a possible infringement of non-CFSP provisions. The Commission took the view, nonetheless, that the contested decision was not an implementing act.

The findings of the Court

Somewhat unsurprisingly, the Court reversed the order of the GC and found that the circumstance that the decision was a CFSP measure “does not necessarily lead to the jurisdiction of the EU judicature being excluded” (par 43).

The Court interpreted the exclusion of jurisdiction very narrowly. It gave a systematic reading of the general provisions of EU law (Article 2) and of CFSP (Articles 21 and 23 TEU) to recall that the EU is founded, in particular, on the values of equality and the rule of law ( Segi and Others v Council; Opinion 2/13). It stated that “The very existence of effective judicial review designed to ensure compliance with provisions of EU law is inherent in the existence of the rule of law (Schrems)” (par 41).

In the current case, the Court considered that the decision of the Head of Mission was subject to legal scrutiny because under Article 270 TFEU the EU judicature has jurisdiction to rule on all actions brought by EU staff members having been seconded to the EUPM. They remain subject to the Staff Regulations during the period of their secondment to the EUPM and, therefore, fall within the jurisdiction of the EU judicature, in accordance with Article 91 of those regulations (even though H was seconded by a Member State, the two situations were considered similar). The decision of the Head of Mission was considered to be merely “staff management”.

Therefore, the Court concluded, “the scope of the limitation, by way of derogation, on the Court’s jurisdiction, which is laid down in the final sentence of the second subparagraph of Article 24(1) TEU and in the first paragraph of Article 275 TFEU, cannot be considered to be so extensive as to exclude the jurisdiction of the EU judicature to review acts of staff management relating to staff members seconded by the Member States the purpose of which is to meet the needs of that mission” (par 55).

The ECJ concluded that “[the] jurisdiction stems, respectively, as regards the review of the legality of those acts, from Article 263 TFEU and, as regards actions for non-contractual liability, from Article 268 TFEU, read in conjunction with the second paragraph of Article 340 TFEU, taking into account Article 19(1) TEU and Article 47 of the Charter of Fundamental Rights of the European Union” (par 58). It therefore referred the case back to the GC.

Rosneft

A similar question recurs in Rosneft, the first request ever for a preliminary ruling on a CFSP act, currently pending before the Court. The case stems from a Russian gas company, Rosneft, challenging sectorial measures (not target sanctions) prohibiting EU natural or legal persons, from engaging in contractual relations with certain Russian state-owned companies and banks, and from providing such companies and banks access to financial markets.

A central question is the admissibility, as discussed at the hearing and in paragraphs 32-76 of AG Wathelet’s opinion.

The AG believes that the measure can be reviewed if it meets these cumulative two conditions: if (a) it relates to Articles 23 to 46 TEU (the foreign policy rules) and or EU acts adopted on the basis of those provisions; and if (b) its substantive content also falls within the sphere of CFSP implementation.

The first condition is derived, for Wathelet, from the consideration that the last sentence of the second subparagraph of Article 24(1) TEU excludes the Court’s jurisdiction only ‘with respect to these provisions’, and the reference thus made is to Chapter 2 of Title V of the EU Treaty, entitled ‘Specific provisions on the common foreign and security policy’, of which Article 24 forms part.

In the AG’s opinion, in particular, the court should have jurisdiction to hear actions for annulment and preliminary rulings on decisions providing for restrictive measures against natural or legal persons adopted by the Council on the basis of Chapter 2 of Title V of the EU Treaty – and not, therefore, regulations implementing them. For the AG, therefore, the Court has jurisdiction, but the challenged decision, to the extent that it is directly addressed to Rosneft, is not invalid. The very long opinion explains in detail why, but here we limit the scope of the analysis to the question on jurisdiction.

Comment

Judicial protection and uniformity of interpretation of EU law

The decision of the ECJ in H should be welcomed because it avoids the potential deterioration of the protection of fundamental rights which would derive from each national court being able to monitor CFSP decisions in the absence of a centralised mechanism. If national Courts had jurisdiction when the CJEU does not, this might lead to diverging and potentially even conflicting interpretations of the same CFSP measure.

Uniformity of interpretation of EU law would be further guaranteed if the Court affirmed jurisdiction to hear requests for preliminary rulings (and AG Walthelet in paras 61-62 of his opinion in Rosneft suggests that the Court can rule on CFSP preliminary rulings). The importance of judicial dialogue between the CJEU and national courts has been repeatedly affirmed in the Court’s case law (Opinion 1/09; CILFIT; Adeneler; Kamberaj). Moreover, absence of the Court jurisdiction to hear on preliminary rulings would be at issue with the third paragraph of Article 267 and the CILFIT doctrine.

The prohibition of judicial dialogue and cooperation between national and EU courts in CFSP may very well be a breach of the right to effective judicial remedy as enshrined in Article 47 EU Charter of Fundamental Rights. Article 47 Charter creates what has been described as a “composite, coherent, and autonomous” standard of EU judicial protection. Pursuant to Article 19(1) TEU, national Courts shall provide remedies sufficient to ensure effective legal protection in the fields covered by Union law, with the standard set and as determined by the CJEU (which has the final saying on interpretation and application of the Treaties). Completely excluding the Court’s jurisdiction from an area of EU law such as CFSP would seriously hinder the system of judicial protection (see to a similar effect  Gestoras Pro Amnistía and Others v Council par 53; Segi and Others v Council par 53).

Even though it is left to the discretion of national courts to decide whether to make a reference for a preliminary ruling as well as the questions to be referred, completely ruling out the opportunity for an applicant (or the national court) to make such a request may indeed be against Article 47 Charter. All the more so if one accepted the reading proposed by the Council in its appeal in H, that is, that the national court does not have the power to annul the CFSP decision. This would leave a legal vacuum in the annulment of the provision (unlike what happened in C-583-11 Inuit, where the Court found that existence of alternative legal remedies allowed for a restrictive rule on judicial remedy).

Political questions doctrine

The preferable option seems to be that only genuinely political acts of CFSP cannot be subject to the Court’s substantial judicial review, although the Court should be able to monitor compliance with the procedural rules of the Treaty and compliance with fundamental human rights. This position is very similar to that expressed by the Commission in H, where it said that only sovereign acts of foreign policy cannot be scrutinised by the Court – without saying anything of formal control.

In H, the Court seemed to conclude that if there was any other reason for the which the Court should have jurisdiction, that reason takes precedence over the exclusion of Article 24, and then the Court does have jurisdiction. This is too broad an understanding of the Court’s powers.

In its judgment in Manufacturing Support & Procurement Kala Naft v Council, the CJEU ruled that it does not have jurisdiction on a CFSP provision which is not a restrictive measure against natural or legal persons pursuant to Article 275 TFEU, and the substantial result might be similar in Rosneft (par 85 AG opinion).

For the reasons explained above, the Court should accept the request on the preliminary ruling in Rosneft, but should then take the opportunity to draw a clear distinction: on one hand, (a) EU acts which are purely political and diplomatically sensitive acts of sovereign foreign policy; on the other hand, (b) all remaining CFSP decisions, all acts of implementation, and provisions of general application.

On (a), which I submit should be assessed on a case by case basis and on their substantial content: the Court should recognise it lacks power of judicial review. Those acts, determined with a “substance over form” rule (see Les Verts par 27; AG Wathelet seems to be taking this position in paras 49-50 of his opinion in Rosneft; see also Gestoras Pro Amnistía and Others v Council par 54; Elitaliana v Eulex Kosovo par 48-49) will have too indirect an effect on individuals (as the case law on Article 263(4) TFEU now stands)

Such acts also have such a discretionary content that courts should defer to the decision of the political actors who adopted them. The latter element, which American constitutional lawyers refer to as the “political question doctrine” is present in many jurisdiction (see par 52 AG Opinion in Rosneft): deference toward the so called “actes de gouvernement”. The Commission proposed this thesis in its written submission and at the oral hearing in Rosneft. The “political question doctrine” is the attitude of courts not to review issues which are inherently political, are best left to the discretion of the actor who took the decision, and are ultimately non-justiciable.

In the leading case on the issue, Baker v Carr, the US Supreme Court held that a question is eminently political if it presents some characteristics such as “a textually demonstrable constitutional commitment of the issue to a coordinate political department”, or “an unusual need for unquestioning adherence to a political decision already made”. While in some cases involving foreign policy decisions the need for adherence to a political decision is evident (ie the ECJ could hardly decide that the EU cannot prohibit commerce with certain Russian companies involved in Crimea at all), arguably the retention of CFSP provisions in the TEU, the preference for intergovernmental institutions in that domain, the scant role of the European Parliament in the decision-making process, not to mention the exclusion of the Court’s jurisdiction, all militate in favour of a strong constitutional preference for CFSP to be resolved by purely political departments. The doctrine could very well be embraced for the first time by the ECJ in deciding Rosneft.

On the other hand, as regards category (b), which includes the case of the “decision on staff management” in H, the Court should exercise its powers of judicial review.

Barnard & Peers: chapter 24

Art credit: The Economist, Peter Schrank

The new Opinion on Data Retention: Does it protect the right to privacy?

Matthew White, Ph.D candidate, Sheffield Hallam University

Introduction

Has an Advocate General (AG) in the Court of Justice of the EU unleashed the power of the European Convention on Human Rights (ECHR)? On 19 July 2016, the AG gave his Opinion in the joined cases of C‑203/15 and C‑698/15 Watson and Tele2. The AG felt that general data retention obligations  imposed by Member States may be compatible with fundamental rights enshrined in EU law, provided that there are robust safeguards (para 7). This post briefly outlines the background (for a more detailed background, see Professor Lorna Woods’s take) to this case whilst highlighting aspects relating to the ECHR and that some of the AG’s conclusions become self defeating for requiring EU law to be no less stringent than the ECHR.

Background

Case C‑203/15

A day after Digital Rights Ireland (where the Court of Justice of the European Union (CJEU) ruled that the EU’s Data Retention Directive (DRD) was invalid for being incompatible with the Charter of Fundamental Rights (CFR)), the first claimant, Tele2, notified the Swedish Post and Telecommunications Authority (PTS) of its decision to cease retaining data in Chapter 6 of the LEK (the relevant Swedish law) with the aim of deleting (para 50). The National Police Board (RPS) complained to the PTS about Tele2’s actions as having serious consequences for law enforcement activities (para 51). PTS ordered Tele2 to resume retention in accordance with Chapter 6 (para 52), to which Tele2 appealed to the Stockholm Administrative Court (SAC) but lost (para 53).  Tele2 then sought to appeal against the SAC (para 54), but the Stockholm Administrative Court of Appeal (SACA) felt making a preliminary reference to the CJEU would be more appropriate where it asked:

·         Is a general obligation to retain all traffic data indiscriminately compatible with Art.15(1) of the (ePrivacy) Directive (Directive) and Articles 7, 8 and 52(1) of the CFR?

·         If no, is such an obligation nevertheless permitted  where:

§  access by national authorities was governed in a specified manner, and

§  the protection and security of data are regulated in a specified manner, and

§  all relevant data is retained for six months?

Case C‑698/15 

I previously blogged on the situation in the UK, but will make a quick summary for the purposes of this post (or alternatively see paras 56-60 of the Opinion). The UK responded to Digital Rights Ireland by introducing the Data Retention Investigatory Powers Act 2014 (DRIPA 2014). This was successfully challenged in the High Court by Tom Watson MP and David Davis MP. But the success was short lived when the Court of Appeal disagreed with the High Court, but made a preliminary reference to the CJEU asking:

·         Did the CJEU in Digital Rights Ireland intend to lay down mandatory requirements of EU law with which the national legislation of Member States must comply?

·         Did the CJEU in Digital Rights Ireland intend to expand the effect of Articles 7 and/or 8, EU Charter beyond the effect of Article 8 ECHR as established in the jurisprudence of the ECtHR?

AG’s Opinion

Asking the wrong question?

The AG initially dealt with the question regarding whether Digital Rights Ireland extended the scope of Article 7 and/or Article 8 of the CFR beyond that of Article 8 of the ECHR. The AG considered this question inadmissible (para 70 and 83) because that possibility was not directly relevant to the resolution of the current dispute (para 75). The AG admitted that the first sentence of Article 52(3) (which lays down rules of interpretation) of the CFR makes clear that any corresponding rights must be the same in meaning and scope to that of the ECHR (para 77). But highlighted, the second sentence of Article 52(3), can permit CJEU to extend the scope of the CFR beyond that of the ECHR (para 78). The ECHR has always been a minimum benchmark as in Trucl and Others v Slovenia it was noted that ‘rights guaranteed by the Convention represented minimum standards’ (para 115). Thus if the EU did acceded to the ECHR (and even if it did not), with or without the second sentence Article 52(3), the CJEU would be free to extend the scope CFR as it saw fit. Therefore in agreement with the AG, the Court of Appeal asked the wrong question.

Lack of corresponding right means rules of interpretation does not apply?

Another important aspect was  pointed out by the AG, who maintained that Article 8 of the CFR has no ECHR corresponding right and therefore the rules of interpretation laid out in the first sentence of Article 52(3) does not apply (para 79). However, there is cause for slight disagreement on this interpretation of Article 52(3). While the High Court admitted that protection of personal data fell within the ambit of Article 8 of the ECHR, they felt Article 8 of the CFR went beyond this because it was more specific and the ECHR had no counterpart (para 80). However, the High Court did so without actually considering Article 8 ECHR case law, therefore their conclusions did not appear to based on anything but mere conjecture and the wording of Article 8 CFR.  This was also questioned by Stalla-Bourdillon because it appeared the High Court followed this interpretation based on there not being an ECHR counterpart. But on closer inspection, as Stalla-Bourdillon highlighted, there is extensive Article 8 case law on the protection of personal data, which is suggested, does in fact correspond with Article 8 CFR. Therefore, both the High Court and AG has fallen prey to only considering the provisions of the ECHR and not the European Court of Human Rights’s (ECtHR) interpretation of those provisions, thus substance over form seemingly prevailed.

It is suggested because there is such extensive case law on the protection of personal data in light of Article 8 ECHR, it is only right that it should be used as a guide when considering Article 8 CFR. Article 52(3) notes that ‘the meaning and scope of those rights shall be the same as those laid down by the said Convention.’ In PPU J McB v LE the CJEU held that not only does the rights set out in ECHR are to correspond, but also the meaning given through the ECtHR’s jurisprudence (para 53) (see also). In Schecke the CJEU held that:

[T]he right to respect for private life with regard to the processing of personal data, recognised by Articles 7 and 8 of the Charter, concerns any information relating to an identified or identifiable individual...and the limitations which may lawfully be imposed on the right to the protection of personal data correspond to those tolerated in relation to Article 8 of the Convention.’(para 52). 

It has been maintained that such an interpretation can be problematic because the CJEU has allowed Article 8 CFR to be absorbed by Article 7. However, this does not and would not weaken the stance that Article 8 CFR as a standalone right should be interpreted (where possible) in accordance with principles of data protection embedded within the ECtHR’s jurisprudence. Read as a whole, Article 52(3) would therefore be properly adhered to, and would also allow the CJEU to deviate, if need be, to offer a higher standard of protection.

A general obligation to retain:

The AG then considered whether Article 15(1) of the Directive allowed Member States to impose a general data retention obligation (para 84) by establishing whether such an obligation fell within the scope of the Directive (para 86). The Czech, French, Polish and United Kingdom Governments all contended that data retention was excluded by Article 1(3) (which excludes matters such as public security, defence, State security from the scope of the Directive) (para 88). However, the AG rejected this by highlighting that:

·         Article 15(1) governed precisely that (retention of data) (para 90),

·         Provisions of access falling within Article 1(3) does not preclude retention from falling within Article 1(3) (para 92-94),

·         The approach taken by the CJEU in Ireland v Parliament and Council meant that general data obligations were not within the scope of criminal law (para 95).

When it came to the issue of whether the Directive applied the AG referred to the Member States ‘entitlement’ under Article 15(1) i.e. Member States have a choice (para 106). The AG then referred to Recital 11 of the Directive which did not alter the balance between an individual’s right to privacy and the possibility of Member States to take measures necessary for the protection of public security etc (para 107). Moreover, the AG highlighted that the Directive did not alter the ability of Member States to carry out lawful interception of electronic communications, or take other measures, if necessary for any of these purposes and in accordance with the ECHR (para 107). The AG opined that general data retention obligations were consistent with the Directive and therefore Member States were entitled to avail themselves of that possibility under Article 15(1), subject not only to its requirements, but that of the CFR in light of Digital Rights Ireland (para 116). Although the AG felt that general obligations of data retention were permissible under EU law (subject to restrictions), an avenue was created for testing the general obligations itself under the ECHR.

In accordance with  the law? But does this not defeat the AG’s premise?

When the AG considered the requirement for legal basis in national law, he invited that CJEU to confirm that the interpretation of ‘provided for by law’ in Article 52(1) CFR accorded with that of the ECtHR’s jurisprudence on a measure being ‘in accordance with the law’(para 134-137). The AG highlighted that the ECtHR has developed a substantial body of jurisprudence on the matter which could be summarised as follows:

·         A legal basis that is adequately accessible and foreseeable i.e. the law is formulated with sufficient precision to enable the individual — if need be with appropriate advice — to regulate their conduct,

·         This legal basis must provide adequate protection against arbitrary interference, and

·         Must define with sufficient clarity the scope and manner of exercise of the power conferred on the competent authorities (para 139).

The AG was of the opinion that ‘provided for by law’ in Article 52(1) CFR needs to be the same as that ascribed to it in connection with the ECHR (para 140). The AG’s reasoning was as follows:

·         Article 53 CFR explains that its provisions must never be inferior to what is guaranteed by the ECHR and therefore the CFR must at least be as stringent as the ECHR (para 141),

·         It would be inappropriate to impose different criteria on the Member States depending on which of those two instruments was under consideration (para 142).

The AG felt that general data retention obligations must be founded on a legal basis that is adequately accessible and foreseeable and provides adequate protection against arbitrary interference (para 143). This would solve the problem of the CJEU falling into ‘the trap of tautologically regarding a legal norm, the validity of which is being questioned, as being allegedly in accordance with the law because it is a law.’

This then raises the interesting issue, if this is the preferred interpretation, how could a general obligation to retain data not amount to arbitrary interference? The AG later admits that the disadvantage of this general obligation arises ‘from the fact that the vast majority of the data retained will relate to persons who will never be connected in any way with serious crime’ (para 252). If the vast majority of data retained is of individuals who are unrelated to any serious crime, how could this even be suggested to not be arbitrary?

If in line with the ECtHR’s jurisprudence, that for a measure to be in accordance with the law, a measure must be sufficiently precise so individuals can regulate their conduct, how could this square with general obligations to retain data which occurs irrespective of conduct? The ECtHR’s Grand Chamber in Zakharov v Russia maintained that the ‘automatic storage for six months of clearly irrelevant data cannot be considered justified under Article 8’ (para 255). As the AG indicated, most data retained will have no relation to the fight against serious crime and therefore, in line with Zakharov, cannot be justified under Article 8. Member States would then have to justify why most data unrelated to serious crime is relevant to the fight against serious crime. In stressing that Article 52(1) should reflect the ECtHR’s jurisprudence the AG may have undermined his own position when believing that general obligations to retain data were permissible under EU law by unleashing the ECHR in terms of Recital 11 and the interpretation of ‘provided for by law.’

Data retention does not adversely affect the essence of the right, or does it, or should it?

The AG listed six requirements a general data retention obligation must meet to be justified, one of such is that it ‘must observe the essence of the rights enshrined in the Charter’ (para 132). The AG recalled that Article 52(1) CFR provides that any limitation to the rights enshrined must respect the essence of those rights and freedoms (para 155). The AG referred to para 39 of Digital Rights Ireland where the CJEU held that the DRD did not adversely affect Article 7 CFR since it did not permit the acquisition of knowledge of the content of the electronic communications as such (para 156). The AG felt this also applied to the current case (para 157) and this was equally the case for Article 8 CFR (paras 158-9) but ultimately left it for the CJEU to decide (para 160).

However, the AG later contradicts his own opinion when considering the disadvantages of data retention. The AG accepted that ‘a general data retention obligation will facilitate equally serious interference as targeted surveillance measures, including those which intercept the content of communications’ (para 254). The AG stopped short of referring to data retention as mass surveillance, but instead referred to it as mass interference (para 255) and that it affected a substantial portion, if not all of the relevant population (para 256). The AG even went further by describing with the example of an individual who access retained data (instead of analysing content) to screen out those within the Member State who have a psychological disorder or any field specialist medicine (para 257). The AG continues, this same person who sought to find out who opposed government policies, could do so with the possibility of identifying individuals taking part in public demonstrations against the government (para 258). 

The AG agreed with the position of several civil society groups, the Law Society and United Nations High Commissioner for Human Rights that the ‘risks associated with access to communications data (or ‘metadata’) may be as great or even greater than those arising from access to the content of communications’ (para 259). The AG further added that the examples given demonstrate that ‘metadata’ can facilitate ‘the almost instantaneous cataloguing of entire populations, something which the content of communications does not’ (para 259). The AG also added that there is was nothing theoretical about the risks of abuse or illegal access to retained data (based on the number of requests by Swedish and UK authorities) and that such risk of ‘illegal access on the part of any person, is as substantial as the existence of computerised databases is extensive’ (para 260).

Considering the incredible detail the AG went to describe the risks posed by the retention of data, it makes little sense to have the opinion that a general data retention obligation does not adversely affect the essence of the right. The AG and CJEU in Digital Rights Ireland premise of this was based on the idea that communications data would not permit acquisition of knowledge of the content of the electronic communications. Yet the AG described in great detail the amount of knowledge that could be gained from communications data. And it is this acquisition of knowledge that is the important factor, the AG described the example of the ability of gaining sensitive knowledge without analysing the content. And so the AG, like the CJEU has created an arbitrary distinction that although the same knowledge can be gained from communications data or content, it is only access to content that could adversely affect the essence of the right (para 94). If it is acknowledged that similar knowledge can be gained from both measures, the CJEU and indeed the AG has not sufficient explained this differential treatment. Furthermore, by only considering that access to content adversely affects the essence of the right, this would promote the use of retention and access to communications data to a greater degree which as the AG admits, can provide far richer information than content.

Indiscriminate data retention maybe EU compliant, but not ECHR compliant

The AG highlighted that the CJEU in Digital Rights Ireland pointed out that the DRD covered all users and all traffic data without differentiation or limitation (para 197). The AG described what the CJEU considered the practical implications of the absence of differentiation i.e. concerning those with no link to serious crime, no relationship between retention and threat to public security, and no temporal, geographical and associate based restriction (para 198). The AG concluded that the CJEU did not hold that the absence of differentiation in itself went beyond what was strictly necessary (para 199).

The AG justified this one four grounds, firstly, the CJEU ruled the DRD as invalid because of the cumulative effects of generalised data retention and the lack of safeguards which sought to limit what strictly necessary for the interference with Article 7 and 8 CFR (paras 201-202). Secondly, in light of Schrems (para 93) the AG inferred again that only general data retention obligations accompanied by sufficient safeguards would be EU law compatible (para 205). Thirdly, the AG felt national measures should be scrutinised at a national level, where the national courts should rigorously verify whether general data retention obligations are the most effective at fighting serious crime i.e. whether there are other less intrusive alternatives (paras 209-210). Fourthly, the AG agreed with the Estonian Government that limiting data retention to a particular geographical area may cause a geographical shift in criminal activity (para 214).

Considering indiscriminate data retention as permissible under EU law if there is a sufficiently robust safeguard mechanism creates problems with the ECHR. In the case of S and Marper v United Kingdom the issue at hand was the retention of finger print and DNA records. In finding the retention regime incompatible with Article 8 (para 126) the ECtHR was struck by blanket and indiscriminate nature of the power because:

[119] material may be retained irrespective of the nature or gravity of the offence with which the individual was originally suspected or of the age of the suspected offender...

[122] Of particular concern in the present context is the risk of stigmatisation, stemming from the fact that persons in the position of the applicants, who have not been convicted of any offence and are entitled to the presumption of innocence, are treated in the same way as convicted persons...

[125] In conclusion, the Court finds that the blanket and indiscriminate nature of the powers of retention of the fingerprints, cellular samples and DNA profiles of persons suspected but not convicted of offences, as applied in the case of the present applicants, fails to strike a fair balance between the competing public and private interests and that the respondent State has overstepped any acceptable margin of appreciation in this regard. Accordingly, the retention at issue constitutes a disproportionate interference with the applicants’ right to respect for private life and cannot be regarded as necessary in a democratic society. This conclusion obviates the need for the Court to consider the applicants’ criticism regarding the adequacy of certain particular safeguards, such as too broad an access to the personal data concerned and insufficient protection against the misuse or abuse of such data.

 S and Marper’s significance has been linked to data retention, and therefore it is important to apply the principles to the AG’s Opinion. The ECtHR criticised the UK regime for not distinguishing between those who had been suspected and those who had committed offences. Regarding data retention obligations, this indiscriminate power is more profound because suspicion would not be a necessary component for the justification of retention. As the AG highlighted, most data retained is of individuals who bare no relation to serious crime and therefore creates issue with the presumption of innocence to an unacceptable level. The most important aspect of the ECtHR’s reasoning in S and Marper was that the retention itself was contrary to the Convention without having to consider the safeguards that may have been in place. This is direct contrast with Digital Rights Ireland and the AG’s Opinion.

Regarding the fourth point, it is submitted that the Estonian Government and the AG misunderstood how data retention and location data works in practice. It is not the physical area that is the important factor, but the location of the device in question at a particular time. This was apparent in Uzun v Germany when the ECtHR described Global Positioning System (GPS) as allowing ‘continuous location, without lapse of time, of objects equipped with a GPS receiver anywhere on earth, with a maximum tolerance of 50 metres at the time’ (para 12-13). This is all the more relevant as location data is becoming more and more sophisticated. Therefore applying a data retention obligation in a specific geographical area creates a false premise as the obligation on the service provider is to keep record of the location data of a device when it’s service is used (which will indicate where an individual might be) irrespective of geographical area. Furthermore a targeted data retention approach would not be confined to a geographical as such, but to criminal activity (based on individual use of device and service) within a particular area.      

  

Six months retention is reasonable?

The issue of retention period was also considered by the AG (para 242) who felt that according to Zakharov a period of six months would be reasonable provided irrelevant data was immediately destroyed (para 243). However, by making this connection, the AG created a false analogy of what the ECtHR held. Zakharov concerned judicially authorised interception and monitoring of communications data of individuals for six months (para 44-48). Therefore the analogy with targeted measures and that of general data retention begins to falter, as in the AG’s own words ‘metadata’ facilitates the almost instantaneous cataloguing of entire populations, something which the content of communications (via interception) does not (para 259).

Conclusion  

Although most of the finer details, in the Opinion of the AG should be left to national courts (para 263) the issue of data retention as a challenge to fundamental rights persists. The AG, by placing great significance on the ECHR and the ECtHR’s jurisprudence unwittingly undermined some of his own key points because they do not accord with the ECHR. It is unlikely that the CJEU are going to rule per se that a general obligation to retain communications data is incompatible with EU law, and therefore maybe an issue for the ECtHR to decide themselves. In light of S and Marper it is possible that the ECtHR would produce a ruling that is in contrast to the CJEU. The United Nations General Assembly has affirmed that same rights that people have offline must also be protected online. The late Caspar Bowden once described data retention as akin to having CCTV inside your head. And so the question becomes, would the AG/CJEU consider that CCTV inside every home would be compatible with EU law provided that access to that footage would be circumscribed by adequate safeguards?

Barnard & Peers: chapter 9

JHA4: chapter II:7

Photo credit: xgtnigeria.com