An ultra vires decision
It has been reported that the Cypriot parliament rejected the agreement over the issue of the protection of Halloumi cheese. If this is indeed the case, the Cypriot representatives acted ultra vires. Appellations of origin come under the exclusive competence of the EU in the framework of the Common Commercial Policy (see also C-389/15). As a result, and because national parliaments have no say over the EU’s exclusive competence other than indirectly through their governments represented in the EU Council, the Cypriot parliament cannot invoke it in the national ratification procedure as a reason for rejecting a trade agreement. If it wants to push the Halloumi issue, it needs to give instructions to the Cypriot representative in the EU Council.

The Cypriot vote does not necessarily spell the end of the provisional application of CETA however, since the latter will only be terminated if ratification of CETA “fails permanently and definitively”. Yet it is clear that the choice to conclude CETA as a mixed agreement is starting to bite back. The Halloumi-incident is a classic illustration of how mixity makes the EU’s external action more burdensome and complicated and this, we argue, unnecessarily so. The choice for mixed action in cases of facultative mixity – as in the case of CETA – undermines democracy and the collective interests of EU Member States and citizens. It is high time that the choice for mixity in a context of shared competence be made subject to an external check.

Mixity: always good for democracy?
Mixity renders the ratification of CETA more cumbersome and uncertain. This is considered by many a price worth paying. The argument is frequently made – and permeates for instance the Namur Declaration – that the involvement of 30 or so parliaments leads to greater democratic legitimacy. The argument has intuitive appeal: national parliaments are close to EU citizens, and a greater number of parliaments will surely lead to more rigorous scrutiny.

While we are sceptical whether involving more parliaments always benefits democracy – is it democratically sound that an assembly representing less than 800 000 EU citizens can block the ratification of an agreement affecting over 450 million people? – we wish to draw attention to an often overlooked but crucial element in the discussion on the democratic legitimacy of mixity in the case of CETA: mixity was the result of a political choice by the Council. And one with a high price at that, as the choice for mixity significantly hollows out the powers of the European Parliament and frustrates a European debate about an issue that is essentially European.

Challenging facultative mixity to protect European democracy
As it befits a multi-level polity with conferred powers only, the question of the democratic legitimation of EU action needs to follow the competence question. Internally, depending on which level of governance is competent to act, the process of democratic legitimation passes through the institutions at either the EU or the Member State level. If an issue falls under exclusive or shared EU competence, the European Parliament and the Council provide the necessary democratic legitimacy through the procedures set out in the Treaties. If an issue falls under exclusive Member State competence, or if a Member State acts when the EU has not (yet) acted, that Member State acts in accordance with its own constitutional procedures. This way, direct lines of accountability run from the entity that acts towards the citizenry that holds that entity to account and the EU citizen.

Things are different in the external context where, as in the case of CETA, the EU and the Member States often act jointly. This has negative repercussions for the EU institutions, and in particular for the European Parliament, which sees its position reduced to but one out of over 30 assemblies with the power to approve or reject CETA. As a consequence, the Parliament is deprived of part of its power: it can reject the agreement, but it cannot approve it. It can vote against CETA (as did the Cypriot parliament), and that would be the end of it. If the European Parliament votes in favour of CETA (as it has done in 2017), the focus of political debate shifts to the national parliaments, where the debate now continues.

The foregoing would have been necessary had it been established that the EU lacked the necessary competence to conclude CETA alone. This is not the case however and this is why it can be claimed that the Parliament is deprived of a power. CETA is, in our view, a case of facultative mixity. As the Court of Justice of the EU (CJEU) made clear in its opinion on the EU-Singapore agreement, virtually that entire agreement, and therefore also CETA, is covered by EU exclusive competences. Even for the agreement’s provisions on Investor-State-Dispute-Settlement (ISDS), the Court has not explicitly noted that these come under exclusive national competences (which would require a mixed agreement, see Rosas).

In addition to the competence question, the policy question whether the EU and its Member States should bind themselves to rules foreseen in CETA is undoubtedly a European question. It is not the sum of one European question and 27 national questions. The national debates on CETA illustrate this since the reasons why CETA is controversial are fundamentally the same in every Member State: the benefits of free trade in general, ISDS, (the enforceability of) environmental and labour standards, the alleged chilling effect on regulation, etc. Addressing these policy questions (also) at the national level (in the name of democracy) has the perverse effect of obstructing a genuinely European debate. In Habermasian terms: CETA would be an ideal topic to discuss in the European public sphere and would incentivize the development of that public sphere, which in turn would foster a European democracy. By opting to conclude CETA as a mixed agreement, European public interests remain confined to and remain to be expressed in national public spheres.

Time for an external check
There is something inherently perverse about an arrangement whereby the EU institution representing Member State interests can unilaterally – and without being subject to an external check — decide whether to empower those same Member States, or the supranational EU institutions. As John Costonis, an American commentator, observed over half a century in an article on the then EEC’s treaty making powers (p. 452): this is a bit like having the wolves guard the sheep.

Principles of sound institutional design would seem to require that the choice to go for an EU-only or a mixed agreement in a context of shared competence be at the very least subject to ex post review. For as long as the Council retains the final word on the matter, mixity will be here to stay, even for agreements such as CETA that are covered for 99% by exclusive EU competences.

In our view, the present arrangement can hardly be considered democratic due to its negative impact on the European Parliament’s institutional position. Further, facultative mixity undermines the EU’s ability to act in pursuit of the collective interest. As Pescatore observed in 1999 (p. 388 at fn 6), ‘mixity combined with a presumption for the competence of member states is … a way of whittling down systematically the personality and capacity of the Community as a representative of the collective interest’.