1Introduction

A narrative of the European integration project emerges under forms of horizontal coordination between the actors of the European Judicial system: The Court of Justice of European Union (hereinafter, ‘the Court’ or ‘the CJEU’), national courts, prosecutors, bailiffs, etc. The rise of these horizontal coordination techniques can be clearly perceived within the Civil Procedure framework.L. Cadiet, ‘The Emergence of a Model of Cooperative Justice in Europe: Horizontal Dimension’, Distinguished lecture at the Center for Judicial Cooperation, European University Institute, Florence, 5 Dec. 2013. For instance, the Council Regulation (EC) No 1206/2001 of 28 May 2001 on cooperation between the courts of the Member States in the taking of evidence in civil or commercial matters enables national judges to take evidence directly from another Member State. Accordingly, article 12.1 states that representatives of the requesting court have the right to be present in the performance of the taking of evidence by the requested court. The entrenchment of the mutual assistance between European jurisdictions in civil matters driven by EU Regulations,See also Council Regulation (EC) No 2201/2003 of 27 November 2003 concerning jurisdiction and the recognition and enforcement of judgments in matrimonial matters and the matters of parental responsibility; EU Regulation n° 1215/2012, 12 December 2012 on jurisdiction, recognition and enforcement of civil and commercial judgement. coupled with the existing judicial networks – the European Judicial Network in criminal, civil and commercial matters, and the European Judicial Training Network – provokes a gradual harmonization of the national procedural rules.See L. Cadiet, ‘The Emergence of a Model of Cooperative Justice in Europe: Horizontal Dimensions’, op. cit., p. 16.

In taxation, the aforementioned cross-border cooperation phenomenon is being implemented between national tax authorities with the aim to ensure that taxes are correctly assessed as well as combating tax fraud and tax evasion. The Council Directive 2011/16/EU of 15 February 2011 on administrative cooperation in the field of taxation reveals the need of widening the exchange of information between Member states, regardless of the protection of taxpayer’s right to be heard or to take part in examination of witnesses in such exchanges of information requests.See Case C-276/12, Sabou [not yet published] dated 22 October 2013. From 1 January 2015 onwards, this Directive stretches the automatic exchange of information over the following income categories: income from employment, director’s fees, and life insurance products not covered by other Directives, pensions, ownership of and income from immovable property. Together with the exchange of information measures, the cooperation network programme Fiscalis 2002ec.europa.eu/taxation_customs/taxation/tax_cooperation/fiscalis_programme/fiscalis_2020/
index_en.htm. contributes to enhance the horizontal cooperation in tax matters between tax administrations.

One could argue that the preliminary reference procedure enshrined in article 267 TFEU responds to this particular horizontal narrative. Indeed, Costa v. EnelCase 6/64, Costa v. Enel [1964] ECR 01141.heralded the separation of functions principle:

‘Article 177 is based upon a clear separation of functions between national courts and the Court of Justice, it cannot empower the latter either to investigate the facts of the case or to criticize the grounds and purpose of the request for interpretation’.The formula of the separation of functions is highly reproduced in the early cases of the Court: Case 20/64, Albatros [1965] ECR 29; Case 100/64, Van der Veen [1964] ECR 1105; Case 20/67, Kunstmühle Tivoli [1968] ECR 293; Case 13/68, Salgoil [1968] ECR 661; Joined Cases 36/80 y 71/80, Irish Creamery Milk Suppliers Association [1981] ECR 00735; Case C-338/85, Pardini Fratelli [1988] ECR 2041.

That means, recalling professor Lenaerts:K. Lenaerts, ‘Form and Substance of the Preliminary Rulings Procedure’, in: N.M. Blokker & S. Muller, Institutional dynamics of European integration: essays in honour of Henry G. Schermers, Vol. II (Dordrecht 1994), p. 355.

‘There is no hierarchical relationship between the Court of Justice and national courts, but only one of cooperation, each court performing its proper function: the Court of Justice interprets Community law at the request of national courts which apply that law – as interpreted – to the cases they have to decide’

Therefore, the separation of functions paves the way to a cooperation between two equal partners – the Court and the national courts – within the preliminary reference procedure. This cooperation, based on each court performing its allocated function, cannot impair the primacy of EU law.See M. Avbelj, ‘Supremacy or primacy of EU Law: (Why) does it matter?’  [2011/vol. 17/n. 6] European Law Journal, p. 750: ‘The autonomous legal orders of European integration constitute a common whole, but are not part of a single European classical hierarchical pyramid of legal sources. The relationship between them is heterarchical and is instead of principle of supremacy governed by the principle of primacy. The principle of primacy is a trans-systemic principle, which regulates the relationship between the autonomous legal orders, ie between the two sovereign levels in European integration. Primacy is a unique feature of EU law, the consequence of its recognised special autonomous nature. Without primacy, the uniform executive force of EU law in every Member State as well as its effectiveness as a prerequisite of being law (effet utile), would be hampered and its very existence would be called into question. To avoid that the principle of primacy requires that in the case of a conflict between the EU and national law the conflicting national legal provisions, irrespective of their form and time of entry into force, must be disapplied.’ As conveyed in Simmenthal,Case 106/177, Simmenthal [1978] 00629, paragraph 17. the primacy of EU law allows the national judge to dis-apply the national law in breach of EU law. Therefore, since the CJEU is confined to give an interpretation of EU law by article 267 TFEU, the ruling received from Luxembourg will serve the referring national court in applying the national provision as interpreted by the CJEU to solve the dispute at stake. The separation of functions proclaimed in Costa v. Enel pursues to safeguard a horizontal cooperation on equal footing, dubbed l’horizontalité in this contribution, between the national court and the CJEU. Here lies the great success of article 267 TFEU: only the national court is entitled to assess the compatibility of national law with EU law.This idea is clearly stated in J.H.H. Weiler, ‘A quiet revolution, The European Court of Justice and its interlocutors’ [1994/vol. 26/ nº 4] Comparative Political Studies, p. 515: ‘What is important in the procedure, indeed crucial, is the fact that it is the national court which renders the final judgment. The main result of this procedure is the binding effect and enforcement value that such decision has on a member state – coming from its courts’; and, J.H.H. Weiler, ‘Journey to an unknown destination: a retrospective and prospective of the European Court of Justice in the arena of political integration’ [1993/ Vol. 31, nº 4] Journal of Common Market studies,
p. 417-446: ‘When a national court accepts the ruling, the compliance pull of Community law becomes formidable. It is an empirical political fact, the reasons for which need not concern us here, that governments find it much harder to disobey their own courts than international tribunals’. The primacy of EU law is effective provided that the separation of functions within the preliminary reference procedure is safeguarded.

Nevertheless, this article constitutes an attempt to challenge l’horizontalité of the relationship between the CJEU and the national courts on the grounds of the case law in direct taxation. In this particular field, the virtual absence of harmonized legislation as a result of the unanimity rule (article 115 TFEU) places the preliminary reference procedure as the most frequent channel to construe EU direct taxation. By analysing the case law in direct taxation, and although the Court still refers to the separation of functions formula in the wording of Costa v. Enel, the constructive cooperation entails that the CJEU assumes discretionary powers to steer the procedural framework envisaged by article 267 TFEU. In order words, this contribution is devoted to stress the manifest tension between the formula of the separation of functions and the effective interplay between the CJEU and the national courts in this field of direct taxation. Therefore, I will make the claim that the constructive cooperation permits the Court to discipline the dialogue with the national courts.

The above-mentioned claim is subject to the following caveat: the field of direct taxation is not specific compared to other fields of EU law. In other words, the features, techniques and tensions that articulate the constructive cooperation in direct taxation can be found scattered throughout the case law of the Court. The novelty of this article is precisely to link all these elements to an area devoid of positive integration such as direct taxation. In this field, we can get a sharply-focused image of the constructive cooperation that shed light on the particular relationship between the national court and the CJEU.

To accomplish this task, in the next section, this contribution attempts to develop the main features of the constructive cooperation in the context of direct taxation. Whereas the second section is devoted to shed light on the rationale of this constructive cooperation, the third section deals with the perils and difficulties that beset it.

2The Constructive Cooperation in Action

Assessing the nature of the cooperation between the CJEU and the national courts is not an easy task. It suffices here to mention how RosasA. Rosas, ‘The European Court of Justice in Context: Forms and Patterns of Judicial Dialogue’ [2007, Vol. 1, nº 2]  European Journal of Legal Studies. classified it as a ‘special relationship’, preventing the horizontal/vertical dichotomy. In the field of direct taxation, I will coin the term constructive cooperationI borrow the term of ‘constructive cooperation’ from L. Weitzel, ‘La reformulation de la question préjudicielle’, in: V. Christianos, Evolution récente du droit communautaire, Vol I (Maastricht 1994) 83-87. to describe the relationship between the national courts and the CJEU within the preliminary reference procedure.

If national courts have the freedom to refer a question for preliminary ruling with the exception of the duty of the courts of last resort when acte clair doctrine does not apply (article 267.3 TFEU), the constructive cooperation will reveal how the CJEU oversees the initial monopoly of the national court to refer a question.The monopoly of the national court is encapsulated in this formula. See for instance, C‑379/98, PreussenElektra, Rec [2001] ECR I‑2099, paragraph 38. In other words, once the trigger of article 267 TFEU is pulled by the national court, the CJEU becomes the absolute protagonist. The constructive cooperation entails thus a transfer of the monopoly of the question posed from the referring national court to the CJEU. Throughout the case law in the field of direct taxation, the constructive cooperation means that: (i) The Court is provided with discretionary powers to discipline the dialogue with the national court; (ii) The Court handles discretionary powers to build up the substantive outcomes of European direct taxation. Yet, the development of substantive principles that articulate this particular area of EU law cannot be understood as being synonymous with the Court adopting an always-pro-integration approach, and therefore producing judgments contrary to the national tax sovereignty. Therefore, there are issues such as the juridical double taxation in which the Court systematically refrains from removing this hurdle despite the undisputed adverse impact on the internal market.The Court has left the solution of juridical double taxation up to the Member States. See Case C-513/04, Kerckhaert-Morres [2006] ECR I-10967 and Case C-128/08, Damseaux [2009] ECR I-06823. Further criticisms can be read in A. Rust, Double Taxation within the European Union (Alphen aan den Rijn 2011). As the Court ruled in Block,Case C-67/08, Block [2009] ECR I-00883, paragraphs 28-30. the fiscal disadvantages as a result of the exercise in parallel by two Member States of their fiscal sovereignty – the same income is taxed twice in two different Member States – do not come under the scope of EU law at its current stage of development. Despite the negative effects to the functioning of the internal market, Member States are the only ones competent to eliminate juridical double taxation.

In the next sub-sections, I will stress how the constructive cooperation works regarding three fundamental steps within the preliminary reference procedure: admissibility, framing and solving the questions posed by the national courts.Due to space constraints of this article, I have selected singular issues concerning those three steps of the procedure.

2.1Admissibility of the Questions: the Notion of an Independent Court

Whereas national courts assume responsibility to determine the relevance of submitting a question for preliminary ruling, the CJEU undertakes the duty to review whether the question posed is admissible. Insofar as only courts or tribunals are entitled to refer a question for preliminary ruling, one of the frequent reasons to reject a reference lies in the non-judicial nature of the referring body.

Since Vaassen-Göbbels,Case 61/65, Vaassen-Göbbels [1966] ECR 261. the notion of a court or tribunal which is empowered to refer a question for a preliminary ruling falls within the scope of EU law.See paragraph 18 of the Opinion AG Jacobs in C-53/03, Syfait [2005] ECR I-4609: ‘it is clear from the Court’s case-law that the concept of ‘court or tribunal’ is one of Community law’. Insofar as the CJEU is not confined to the concept of court or tribunal provided by national law, this concept has been enlarged to include bodies or institutions that are not embedded within the judiciary of the Member states.See paragraph 56 of the Opinion of AG Sharpston in Case C-196/09, Paul Miles and others v. European schools [2011] ECR I-05105: ‘herefore, Article 234 EC should be construed purposively and the words a ‘court or tribunal of a Member State’ should be given a broad interpretation’. However, it should be underlined that what the CJEU purported with the Vaassen-GöbbelsThe Vaassen-Göbbels criteria are: (I) the body is established by law; (ii) it is permanent; (iii) it applies rules of law; (iv) its jurisdiction is compulsory; (v) it is independent; (vi) it is called upon to give judgment in proceedings intended to lead to a decision of a judicial nature and; (vii) its procedure is inter partes (adversary procedure). criteria and the final outcome obtained after the reading of the case law are apparently far away enough to be qualified as compelling or rigorous.See paragraph 14 of the Opinion of AG Jarabo-Colomer in C-17/00, De Coster [2001] ECR I-9445. The critics of this lack of clarity in the CJEU jurisprudence can be found in M. Broberg & N. Fenger, Preliminary references to the European Court of Justice (Oxford 2010), p. 59-64;
K. Lenaerts, D. Arts & I. Mosels, Procedural Law of the European Union (London 2006), p. 33-45; M. Broberg, ‘Preliminary References by Public Administrative Bodies: when are Public Administrative Bodies competent to make preliminary references to the European Court of Justice’ [2009/15, nº 2] European Public Law 15, p. 207; T. Tridimas, ‘Knocking on Heaven’s door: fragmentation, efficiency and defiance in the preliminary reference procedure’ [2003/40] Commom Market Law Review, p. 27; D.W.K. Anderson & M. Demetriou, References to the European Court (London 2002), p. 31; M. Baptista, Manuale del rinvio pregiudiziale (Milano 2000), p. 47; H.G. Schermers, C.W.A. Timmermans & A.E. Kellermann, Article 177 EEC: Experiences and Problems (The Hague 1987), p. 135; M. Jimeno Bulnes, La Cuestión Prejudicial del artículo 177 TCE (Zaragoza 1996), p. 183. This lack of clarity is particularly relevant in the field of taxation, provided that many disputes are challenged by appeal administrative bodies embedded within the executive power prior to the judicial review. In the following paragraphs, this contribution will try to discuss the lack of consistency of the notion of independence by investigating the responses of the Court to questions referred by appeal administrative tax tribunals. This lack of certainty clearly benefits the discretionary power of the Court to discipline the dialogue with the national court.

In Corbiau,Case C-24/92, Corbiau [1993] ECR I-1277, paragraphs 15-16. the CJEU held the reference inadmissible on the grounds that there was a clear organizational link between the referring body (Directeur des Contributions Directes et des Accises) and the body adopting the tax settlement in dispute. After this decision, inasmuch as this criterion would hamper the access of the references requested by tax appeal administrative bodies, the Court switched from a strict organizational link to a functional link.

In Gabalfrisa and others,Joined Cases C-110/98-147/98, GabalfrisEa and others [2000] ECR I-1577, paragraphs 39-40. the Court started to check whether the Spanish Appeal Board (‘Tribunal Económico-Administrativo Regional de Cataluña’) was functionally independent to perform its review of the fiscal complaints brought by taxpayers. Accordingly, despite the fact that the Spanish Appeal Board (‘Tribunal Económico-Administrativo Regional de Cataluña’) was incorporated into the Ministry of Economic Affairs, the Court found that the Spanish regulations safeguarded the independence of these bodies, preventing them from receiving instructions from the Ministry of Economic Affairs.

The wide margins put forward in Gabalfrisa and others, despite Advocate General Saggio’s Opinion, seem to break down when the CJEU faced the SchmidCase C-516/99, Schmid [2002] ECR I-4573, paragraphs 36-41. case.Although the Appeal Chamber of the regional finance authority for Vienna, Niederösterreich and Burgenland (the Austrian Appeal Chamber) functioned in a similar way to the Spanish Tribunales Económicos-Administrativos Centrales, the response of the CJEU was completely the opposite. In Schmid, the members of the Appeal Chamber were not in principle bound by any directions in the exercise of their functions and the members swore on oath to take impartial decisions. However, the CJEU disregarded the Appeal Chamber as a functional separate body from the administrative authority whose decision was challenged on the grounds that:

(I) ‘The President of the regional finance authority has the power to nominate members of the appeal chambers on the basis of the lists of appeal commission members. There is no legislative provision to prevent him from modifying, at his discretion, the composition of an appeal chamber for the inquiry into each complaint, or even in the course of the inquiry into a complaint’.Ibid., paragraph 41.; (II) ‘Finally, and above all, the President of the regional finance authority may – and here he is subject to possible directions from the Finance Minister – bring an appeal against a decision of an appeal chamber (Paragraph 292 of the BAO) and on that occasion defend a point of view different from that adopted by the chamber of which he is president’.Ibid., paragraph 42.

It should be noted that the composition, appointments and removal/withdrawal of the members of the Appeal Chamber were the same that were applicable to the Tribunales Económicos-Administrativos in Gabalfrisa and others.See article 16 of the Royal Decree Nº 391/1996. Likewise, Spanish tax authorities were legitimized to bring an appeal against the resolution given by the Tribunal Económico-Administrativo Regional before the Tribunal Económico-administrativo Central, in the case that the resolutions adopted were contrary to the general interest of the tax administration.See article 11.1 d) of Legislative Decree No 2795/1980 and Article 120 of Royal Decree
Nº 391/1996. As such, in the wording of Gabalfrisa and others, the Court neglected mentioning the above Spanish provisions that could have led to the same outcome of Schmid. Whereas in Gabalfrisa and others the Court considered the ‘Tribunales Económicos-Administrativos’ as courts or tribunals within the meaning of article 267 TFEU, in Schmid the CJEU overturn the previous reasoning and declined to give a preliminary ruling to the questions raised by the Appeal Chamber.

Finally, in Nidera HandelscompanieCase C-385/09, Nidera Handelscompagnie [2010] ECR I-10385. the Court accepted the questions referred by the Lithuanian Tax Disputes Commission. Although the members of this Tax Disputes Commission are appointed by the executive, there are the following elements of independence: (i) they are people of irreproachable reputation, (ii) the members hold office only in the Commission; and finally, (iii) they are appointed for a term of six years. Curiously, the Court does not examine the fact that the members of the Commission can be dismissed before their term of office expires when they seriously violate their work duties. To the extent that the appraisal of this event fits into the discretionary power of the executive, the independence of the members to ward off the pressures from the executive can be put in danger. Although the Lithuanian Law on tax administration envisages a duty of cooperation with the Ministry of Finance, the Lithuanian Government stated at the hearing that ‘there are no cases in which the Tax Disputes Commission had received from the Ministry, instructions or guidance as to the solution which would be preferred in a certain case’. Note that the CJEU has accepted a factual argument employed by the Lithuanian Government at the hearing, without analysing in depth whether or not the legal duty of cooperation with the Ministry of Finance affects the independence of the appeal board specialized in taxation.

In cases concerning other areas of EU law, the casuistic approach prevails. For instance, in TDC,Case C-222/13, TDC [not yet reported] dated 9 October 2014, paragraphs 34-38. unlike Nidera Handelscompanie, the fact that the Danish Government at the hearing confirmed that the members of the Teleklagenævnet could be removed from office by the Minister, who also has the power to appoint them, was relevant enough to conclude that the Court did not have jurisdiction to answer the questions referred. In other rulings such as Broekmeulen,Case 246/80, Broekmeulen [1981] ECR 2311, paragraphs 16-17; Case C-394/11, Belov [not yet reported] date 31 January 2013, paragraph 52. rather than handling a Vaassen-Göbbels approach, the Court wonders whether rejecting the reference made by the national body can jeopardize the effectiveness of EU law. Therefore, whenever the decision of the referring body can be subject to appeal before a body clearly embedded within the judiciary, the Court declines jurisdiction to rule. In those cases, the effectiveness of EU law would be ensured by an eventual reference requested by the appeal body. Had the Court followed this approach in Nidera Handelscompanie, the administrative appeal boards would have been excluded from the concept of court within the meaning of article 267 TFEU on the grounds that their decisions can always been challenged before the judiciary.

Both in taxation and other fields of EU law, the Court generally uses a case-by-case approach to determine whether an administrative board is entitled to refer a question for preliminary ruling, thereby analysing not only the national norms that regulate the composition and functioning of these bodies, but also the observations put forward by the Governments. Since the information the Court checks is provided in the order or reference, the outcome is casuistic and devoid of consistency. Why did the Court ignore the hypothetical pressures that Tax Disputes Commission can suffer from the Ministry of Finance in Nidera Handelscompanie? Why did the Court neglect the fact that the members of the tax appeal administrative boards can be easily dismissed by the Ministry of Finance before their term expires, whereas in the recent TDC this circumstance is crucial to decline jurisdiction? The absence of a substantive definition of a court or tribunal under article 267 TFEU leads to a clear result: the Court holds a discretionary power to decide what a Court is (a sort of opening/closing valve which controls the flow of references).T. De la Mare, ‘Article 177 and legal integration’, in: P. Craig & G. de Burca, The Evolution of EU law (Oxford 1999), p. 220: ‘The Court has set the valve fairly wide open by taking a broad, purposive view of what is a court’.

The above examples reflect the constructive cooperation in action. Firstly, the discretionary powers handled by the Court discipline the dialogue with the national court. As the upper courts operate in national jurisdictions, the CJEU also rules on the admissibility of the referred questions by looking into the characteristics of the referring body. Secondly, the lack of consistent and clear criteria to define an independent body entitled to refer a question for preliminary ruling permit the Court either to open or to close the valve depending on the case at stake. Perhaps, due to the difficulties associated with the definition of court or tribunal, the CJEU should replace the Vaassen-Göbbels approach for the one held in Broekmeulen or Belov. The admissibility of the questions would be anchored in the need for preserving the effectiveness of EU law. Therefore, administrative appeal boards whose decisions can be appealed to judicial courts would be carved out of the definition of court within article 267 TFEU.

2.2Framing the Question Posed by the National Court

Once the CJEU is seized by a preliminary reference requested by a national court, the main endeavour to accomplish is to frame the question into the proper EU law context. Pursuant to article 94 of the Rules of Procedure of the Court, the request for a preliminary ruling must contain an account of the facts on which the questions are based as well as ‘a statement of the reasons which prompted the referring court or tribunal to inquire about the interpretation or validity of certain provisions of European Union law’. As long as the national court keeps the monopoly of the questions referred, due to the separation of functions principle proclaimed in Costa v. Enel, any doubt stemming from the interpretation of those questions posed must trigger a request for clarification addressed to the referring court according to article 101 of the Rules of Procedure of the Court.

However, the reality is quite different. The Court often recasts the question addressed by the referring court in a different EU legal context, taking into account the facts of the case, the current developments of EU law and mainly without sending a request for clarification to the referring body.See the formula of reformulation, for instance in Case C‑452/14, Agenzia Italiana del Farmaco [not yet reported] dated 1 October 2015, at 33: ‘It must be recalled in this regard that, in the procedure laid down by Article 267 TFEU providing for cooperation between national courts and the Court of Justice, it is for the latter to provide the national court with an answer which will be of use to it and enable it to determine the case before it. To that end, the Court may have to reformulate the questions referred to it. The Court has a duty to interpret all provisions of EU law which national courts require in order to decide the actions pending before them, even if those provisions are not expressly indicated in the questions referred to the Court by those courts’. The reformulation of questions becomes particularly relevant in sensitive areas such as fundamental rights. In a landmark decision, DEB,Case C-279/09, DEB [2010] ECR I-13849. the referring body was wondering whether the refusal of legal aid to an undertaking for the pursuit of an action seeking to establish State liability under EU Law might be contrary to the principle of effectiveness.Ibid., paragraph 25. However, inasmuch as the principle of effective judicial protection is enshrined in article 47 of the Charter of Fundamental Rights, the question was reframed with the aim to appraise whether the implementation of the right of effective judicial protection laid down in the Charter of Fundamental Rights could be opposed to a national rule under which the pursuit of a claim before the courts is subject to the making of an advance payment. Therefore, the Court is not constrained to the EU law provisions set out by the national court in the order of reference. Indeed, reframing the questions posed by the national courts permits EU law to progress in the effective implementation of the Charter of Fundamental Rights.

In direct taxation, since article 115 TFEU requires unanimity to harmonize, reframing the questions posed by the national courts could enable the Court to go beyond the initial EU legal framework proposed by the national court with the aim to foster the developments of EU direct tax law. The following two cases – Daily MailCase C-81/87, Daily Mail [1988] ECR 05483. in 1988 y National GridCase C-371/10, National Grid [2011] ECR I-12273. in 2010 – demonstrate that the same facts are reframed differently by the Court. Both cases deal with the right of a company incorporated under the legislation of a Member State to transfer its central management and control to another Member State.

In 1988, the facts of Daily Mail were the following: A Company incorporated under the legislation of United Kingdom purported to move its central management and control to the Netherlands. The principal reason for this transfer was to sell a significant part of its assets in the Netherlands for the purpose of minimizing the final taxation of the transaction. That is, in the Netherlands, the foreseen transaction would be only subject to the capital gain which has been accrued since the transfer of the management took place. To the extent that the legislation in United Kingdom required the consent of the Tax Authorities to transfer the management to other Member States, and after a long period of negotiations, a final resolution was issued that forced the undertaking to sell at least part of the assets before transferring its residence for tax purposes to the Netherlands. Therefore, this company lodged a claim against this resolution on the grounds that the regime of prior authorization before transferring the tax residence to other Member States was incompatible with Articles 52 and 58 of the EEC Treaty. The questions referred to the CJEU for a preliminary ruling disclose the tax nature of the claim because the change of residence to other Member State is connected to a tax advantage. Although Daily Mail depicts the interplay between the EU freedoms and taxation, the Court reformulated the questions referred in terms of Corporate Law:

‘The answer to the first part of the first question must therefore be that in the present state of Community Law Articles 52 and 58 of the Treaty, properly construed, confer no right on a company incorporated under the legislation of a Member State and having its registered office there to transfer its central management and control to another Member State.’

In the then current status of Community law (in 1988), the Court did not rule on whether the national measure requiring authorization for the purpose of changing the tax residence in the cases in which the undertaking would obtain a tax benefit was contrary to the Treaty. Notice the wording ‘in the present state of Community Law’ reflects that the timing of the case didn’t allow the Court to solve the case as it was framed by the national Court. It should be observed that the first landmark case in direct taxation, Avoir FiscalCase C- 270/83, Commission/France [1986] ECR 00273. was handed down quite recently, so there was not a solid line of cases to appraise national direct tax measures that could pose as a hurdle to the achievement of the Common Market aim.

The same issue was disputed in 2010. In National Grid, an undertaking incorporated under the Netherlands transferred its place of effective management or control (tax residence) to the United Kingdom. This transaction triggered a capital gain tax in the Netherlands which was levied on the unrealized capital gain at the time of the transfer of the company’s place of management. The national court raised the following question: If a Member State imposes on a company incorporated under the law of that Member State which transfers its place of effective management from that Member State to another Member State a final settlement tax in respect of that transfer, can that company, in the present state of Community law, rely on Article 43 EC (now Article 49 TFEU) against that Member State? Contrary to what it was upheld in Daily Mail, the maturity of EU Law had reached a certain level, after landmark decisions, that entitled the Court to frame the case in precise terms of taxation: Does the ‘exit tax’ accrued in the transfer of the centre of management of an undertaking to another Member State oppose the EU freedom of establishment? However, in 1988, to answer the same question on an exit tax, the Court opted to reformulate it in terms of corporate law. Both cases demonstrate that the reformulation of preliminary rulings is not a neutral technique, but unveils a political decision of the Court.

This suffices to conclude that the technique of reformulation could hypothetically play an important role in an area like direct taxation where the unanimity rule thwarts future developments in this field. This occurs when the reframing of the questions referred by the national court allows the Court to abandon a consolidated precedent in the light of new interpretative criteria. For instance, in Lasertec,Case C-492/04, Lasertec [2007] ECR I-03775. regarding the applicability of thin capitalization rules involving third-country nationals, the CJEU recast the questions referred on the basis of freedom of establishment, instead of the free movement of capital. Since the freedom of establishment does not extend the scope of its provisions to situations involving nationals of non-member countries who are established outside the European Union, the reference was rejected. In Lasertec, although the Court had never before provided a decision on the priority of these two fundamental freedoms in a third-country situation, the recasting of the question took the form of an order of article 99 of the Rules of Procedure of the Court.V. Englmair, ‘The Relevance of the Fundamental Freedoms for Direct Taxation’, in:
M. Lang, Introduction to European Tax Law on Direct Taxation (Vienna 2013), p. 53. Giving an order instead of a judgment reveals how reframing entails executive powers to foster the developments of EU direct tax law. Notwithstanding the above, the CJEU has recently overruled this criterion in Test Claimants in the FII Group Litigation and Itelcar.Case C-35/11, Test Claimants in the FII Group Litigation 2 [not yet reported] dated 13 November 2012; Case C-282/12, Itelcar [not yet reported] dated 3 October 2013, paragraphs 23-24. The Court has widened the scope of the free circulation of capitals concerning third countries nationals, regardless of the participation threshold.

Nevertheless, taking into account that direct taxation is still under the sovereignty of the Member States, the reformulation in this particular area could also serve the Court to narrow the scope of the question, thereby preventing resistance and criticism from Member States. Rather than a general answer to be extrapolated to other contexts, the Court, instead, decides to give a concrete answer tailored to the factual settings provided by the national court. This cautious approach is followed in the recent Verest.Case C‑489/13, Ronny Verest, Gaby Gerards [not yet reported] dated 11 September 2014, at 13-17. If the referring court asked the Court to rule on the compatibility with EU law of national law which levies a tax on immovable property which is not rented out on a different basis that its local cadastral income, the Court reformulated the question to limit the answer to the particular context in which the questions were raised (interpretation of the progressivity clause contained in the Convention for the prevention of double taxation between France and Belgium).

Pursuant to the case law in the field of direct taxation, reframing the questions referred by the national court requires the use of discretionary powers to place the question into the suitable context. Consequently, as Daily Mail reveals, the CJEU has the right to alter the original framing of the question suggested by the referring national court. Although the Court disciplines the dialogue with the national court by reframing the questions posed into the most suitable EU law context (setting aside a clear precedent like in Lasertec), in other cases the reformulation limits the scope of the initial question and therefore, shows the fears of the Court to penetrate in the field of direct taxation which is reserved to Member States.

2.3Solving the Question: the Upraising of Normative Criteria

The separation of functions means that the CJEU is constrained to give an interpretation of EU law that enable the referring court to solve the case at stake.P. Boria, Diritto Tributario Europeo, Giuffré (Milano 2010), p. 116: ‘La Corte di Giustizia è chiamata a pronunciarsi sulla quaestio juris, definendo il significato della norma comunitaria rilevante, mentre il giudice nazionale è tenuto a pronunciarsi sulla quaestio facti, così da prevenire alla decisione sulla fattispecie concreta applicando le norme pertinenti (compresa quella comunitaria)’. Whilst interpretation corresponds to the CJEU, application comes under the scope of the national court. However, in the daily practice of the Court, drawing a line between interpretation and application becomes almost impossible. The facts and national law aspects provided by the national court are integrated within the preliminary ruling.K. Lenaerts, ‘Form and Substance of the Preliminary Rulings Procedure’, op. cit., p. 379: For the first time, the Foglia v. Novello saga – Case C-104/79, Foglia v. Novello [1980] ECR 00745; Case C-244/80, Foglia v. Novello [1981] ECR 03045 – implies reviewing the facts provided by the national court. See G. Vandersanden, Renvoi préjudiciel en droit européen (Bruxelles 2013), p. 84: ‘Par ces deux arrêts, la Cour se réserve désormais le droit – tout au moins en cas de litiges dits artificiels – de contrôler les faits qui sont à la base du litige de droit interne et qui ont conduit à sa saisine et, dès lors, elle oblige le juge national à lui fournir le dossier de l’affaire pour lui permettre d’examiner le contexte dans lequel elle est saisie. Le juge national doit donc, en toute hypothèse, définir le cadre factuel et réglementaire dans lequel s’insèrent les questions qu’il pose ou, à tout le moins, expliquer les hypothèses factuelles sur lesquelles ces questions sont posées’. Direct taxation is not an exception to this general blurring of lines between interpretation and application. The Court is basically prompted to solve the clash between the EU freedoms of circulation and the principles like non-discrimination, abuse of rights, etc., and the national provisions. Consequently, not only an abstract interpretation of EU law is needed, but also the ruling of the Court needs to appraise national law provisions in the light of EU law. As TRIDIMAS has put it:T. Tridimas, ‘Constitutional review of member state action: The virtues and vices of an incomplete jurisdiction’ [2011/9, nº 3-4] I•CON, p. 738. See also L. Azoulai & R. Dehousse, ‘The European Court of Justice and the legal dynamics of integration’, in: E. Jones, A. Menon &
S. Weatherill, The Oxford Handbook of the European Union (Oxford 2012), p. 356: ‘the Court of Justice accepted the national courts’ using the channel of the preliminary ruling (article 267 TFEU) to ask it – in terms often devoid of all ambiguity – about the compatibility of national provisions with EU law, thus bringing about a true decentralization of judicial review of the way EU law was implemented’. ‘The national court requests a ruling on the interpretation of EU law but does so with a view to establishing, in one form or another, the compatibility of national law with EU law’. These are defined as outcomes cases inasmuch as the referring court does not have margin for manoeuver: dis-apply national law in breach of EU law.T. Tridimas, ‘Constitutional review of member state action: The virtues and vices of an incomplete jurisdiction’, op. cit., p. 739. The list of rulings related to deductions and allowances with respect to family circumstances – Schumacker, Gilly, Gschwind, De Groot, Wallentin, GielenCase C-279/93, Schumacker [1995] ECR I-00225; Case C-336/96, Gilly [1998] ECR I-02793; Case C-391/97, Gschwind [1999] ECR I-5451; Case C-385/00, De Groot [2002] ECR I-11819; Case C-169/03, Wallentin [2004] ECR I-6443; Case C-440/08, Gielen [2010] ECR I-02323. – reveals how the Court must dig into the facts provided by the referring court in order to verify whether a discrimination between residents and non-residents takes place. In other words, the assessment of the factual settings yields precise outcomes concerning the compatibility of the national provisions at stake with EU law.Vid. G. Melis, ‘Motivazione e argomentazione nelle sentenze interpretative della Corte di Giustizia in materia tributaria: alcuni spunti di riflessione’ [2005/2] Rassegna Tributaria,
p. 401: ‘...che la questione sollevata raramente si limita alla richiesta di mera interpretazione della norma comunitaria, coinvolgendo sovente anche la compatibilitá con quest’ ultima della norma interna, e, dall’altro, che la sentenza della Corte puó, per il contenuto che la caratterizza nello specifico caso, porsi como immediatamente decisoria della controversia pendente dinanzi ai giudici nazionali, oppure essere fonte di ulteriori accertamenti di fatto in capo al giudice nazionale’; R. Van Brederode, ‘Judicial cooperation and legal interpretation in European Union Tax Law’ [2009/1-nº 1], Faulkner Law Review, p. 5: ‘The functions of the ECJ and the national courts are distinct. The first is limited to the interpretation of Community Law and to deciding whether actions by the Community or its institutions are valid. The latter has to apply Community Law (as interpreted and explained by the ECJ) to the facts of the case. This functional separation between interpretation and application is not as strict in practice as it may be in theory, because the ECJ renders its ruling within the factual context of the referred case. The facts and circumstances may be so specific that the ECJ’s preliminary ruling may not be generally applicable’; V. Nucera, Sentenze pregiudiziali della corte di giustizia e ordinamento tributario interno (Milano 2010), p. 62; M. Scuffi, ‘I rapporti tra la giurisdizione tributaria e l’ordinamento comunitario: i poteri del giudice tributario nell’ interpretazione ed applicazione del diritto comunitario’ [2005] Rivista Il Fisco, p. 6529.

Nevertheless, in exceptional cases whereby the evidence undertakes a decisive role in solving the case, the CJEU provides the referring court guidelines as to how to resolve the dispute.These cases are quoted by professor Tridimas as guidance cases. See T. Tridimas, ‘Constitutional review of member state action: The virtues and vices of an incomplete jurisdiction’, op. cit., p. 739. Unlike the outcome cases, in these guidance cases, to the extent that the referring court is the only competent to assess the evidences collected, the Court elaborates certain interpretative rules that would assist the referring court to solve the trial.. The guidance cases are frequent in areas such as tax fraud or avoidance. For instance, in FoggiaCase C-126/10, Foggia – Sociedade Gestora de Participações Sociais [2011] ECR I-10923. the national court is the only competent authority to assess the existence of ‘valid commercial reasons’ in a merger operation between two companies of the same group, but in the light of the interpretative criteria outlined in the ruling:

‘the fact that the acquired company does not carry out any activity, does not have any financial holdings and transfers to the acquiring company only substantial tax losses of undetermined origin, even though that operation has a positive effect in terms of cost structure savings for that group, may constitute a presumption that the operation has not been carried out for “valid commercial reasons”.’Ibid., at 53.

Turning to the outcome of the cases, the interplay between the facts of the case and the final outcome reached by the Court can be clearly appreciated in the reasoning in Schumacker. In this case, Mr Schumacker (a resident in Belgium with his wife and children) worked in Germany for a certain period of time. In the host country, Germany, the tax authorities refused to grant him the same treatment of married employed persons residing in Germany (i.e. tax splitting). Whilst in Germany this allowance was denied, in Belgium neither Mr Schumacker, nor his wife could benefit from family and personal deductions due to the absence of sufficient income to be taxed. Although, a classical principle of international tax law sets forth that it is primarily a matter of the state of residence of the taxpayer to grant personal and family deductions, the Court held that in the cases wherein the non-resident receives the major part of his income in a Member State other than that of his residence, discrimination arises from the fact that his personal and family circumstances are taken into account neither in the State of residence nor in the State of employment. Thus, The Court urged Germany to take into account the personal and family circumstances of Mr Schumacker. Summarizing, the particular factual settings of the case – Mr Schumacker obtains the major part of his income in the host country and the residence country cannot take into account the personal and family allowances – lead to the outcome of the case: the host state must take into account his personal and family circumstances.

The line of reasoning in Schumacker can be clearly perceived in the following cases. For example, in Wallentin the Court employs the Schumacker’s yardstick to appraise the discriminatory effects of the Swedish law which did not grant a non-resident certain tax benefits. Bearing in mind that Mr Wallentin obtained the major part of his income in Sweden and his resident sate could not take into account his personal circumstances, The Court finally determined the existence of a discrimination prohibited by the Treaty. With regard to the personal and family allowances in personal income tax, the cases handed down since Schumacker constitute a clear ‘jurisprudence’, a series of precedents that the Court itself is constrained to follow.M. Poiares Maduro, ‘Interpreting European Law: Judicial Adjudication in a Context of Constitutional Pluralism’ [2007/1-nº2], European Journal of Legal Studies, p. 14: ‘Finally, the Court is also constrained by a ‘precedent-oriented’ approach. Independently of determining whether or not its decisions have the nature of a classical legal precedent, the Court has consistently stated that deference is due to a well-established line of case law. The authority which the Court itself recognises to its previous decisions is a consequence of the need to guarantee the values of coherence, uniformity and legal certainty inherent to any legal system’; M. Poiares Maduro, ‘Preface’, in: A.P. Dourado & R. Da Palma Borges, The Acte Clair in EC Direct Tax Law (Amsterdam 2008), p. 3. In response to the lack of positive harmonization in direct taxes, these precedents are regarded as the normative criteria that set up the Corpus Iuris of the European direct taxation. The rise of these normative criteria can be traced across the case law. With regard to the inheritance and gift taxes, the Court has challenged the national provisions which led to different treatment to non-residents depending on the place of residence of the deceased or the recipient of the gift.Case C-181/12, Yvon Welte [not yet reported] dated 17 October 2013, paragraph 25: Case C‑11/07, Eckelkamp [2008] ECR I‑6845; Case C‑43/07, Arens-Sikken [2008] EC I‑6887; Case C-510/08, Vera Mattner [2010] ECR I-03553; Case C-132/10, Halley [2011] ECR I-08353; Case C-31/11,
Scheunemann [not yet reported] dated 19 July 2012. These normative criteria have a clear radiation effectP. Pescatore, The Law of integration (Leiden 1974), p.100: ‘But beyond the legal authority of the Court’s judgements – which, like all judicial decisions, are strictly binding only in relation to the particular subject matter – preliminary rulings have a radiation effect which makes them directives of interpretation observed throughout the Community’; P. Pescatore, ‘Il rinvio pregiudiziale di cui all’art.177 del trattato CEE e la cooperazione tra la corte ed i giudici nazionali’ [1986/109], Il foro Italiano, p. 26-47. in the Member States.

In the field of direct taxation, assessing the compatibility of national law with EU law enables the Court to establish precedents.G. Itzcovich, ‘The European Court of Justice as a Constitutional Court. Legal Reasoning in a Comparative Perspective’, Sant’Anna Legal Studies Research Paper 4/2014, 2014, p. 11: ‘The Court has the authority to question, of its own motion, the admissibility of the action and, in so doing, it can establish precedents that gradually constitute a more or less consistent case law on the criteria according to which the Court can be called to adjudicate on certain matters. Nonetheless, the ECJ does not enjoy discretion in the strong sense that it can refuse to review a case without providing any argument, on a groundless basis or for reasons of opportunity and/or political necessity’. The constructive cooperation lies in these precedents that guarantee a correct application of the acte clair doctrine. To achieve a decentralized jurisdictional model,P. Craig, ‘The Classics of EU Law Revisited: CILFIT and Foto-Frost’, in: L. Azoulai & M. Poiares Maduro, The Past and Future of EU Law: the Classics of EU Law Revisited on the 50th Anniversary of the Rome Treaty (Portland 2010), p. 188: ‘The result of CILFIT y De Costa was that national courts became “enrolled” as part of a network of courts adjudicating on Community Law, with the ECJ at the apex of that network. They become “delegates” in the enforcement of EC Law, and part of a broader Community judicial hierarchy’; A. Stone Sweet, ‘The juridical Coup d’État and the problem of authority: CILFIT and Foto-Frost’, in: L. Azoulai & M. Poiaes Maduro, The Past and Future of EU Law: the Classics of EU Law Revisited on the 50th Anniversary of the Rome Treaty, op. cit. , p. 188: ‘The Court expects national judges to operate as agents of the Community order: when they adjudicate disputes in domains governed by EU Law, they are obliged to take decisions with reference, and deference, to that Law. As European integration has deepened, the list of duties the ECJ has assigned to national judges, as the facto Community judges, has lengthened’.} the Court needs to settle clear precedents that would allow national courts to apply EU law correctly without requesting a reference for preliminary ruling, and therefore in the long term, resulting in a decrease of the Court’s workload.

3The Rationale of the Constructive Cooperation

The constructive cooperation outlines a profound breach of the separation of functions. Regardless of the fact that the Court continues to allude to the Costa v. Enel formula,See Case C-370/12, Pringle [not yet reported], dated 27 November 2012, paragraph 83 ‘It should first be recalled that, in accordance with settled case‑law of the Court, the procedure provided for by Article 267 TFEU is an instrument for cooperation between the Court and national courts by means of which the Court provides national courts with the criteria for the interpretation of European Union law which they need in order to decide the disputes before them’. it pursues to discipline the dialogue with the national courts as well as to push forward the development of EU direct taxation. Therefore, article 267 TFEU can no longer be depicted as placing the CJEU and the national courts at the same hierarchical level (‘l’horizontalité’). The tools employed by the Court under the constructive cooperation are not unique for the field of direct taxation as, for example, the reformulation technique shows in the area of fundamental rights. Nonetheless, in a field devoid of positive integration, the role of the Court to eliminate the hurdles that hinder the internal market becomes crucial and the preliminary reference procedure matches this goal. What rationale is behind the constructive cooperation?

Since Van Gend en Loos,Asunto 26/62, Van Gend en Loos [1963] ECR 00003. EU law grants individuals with subjective rights opposable to the Member States:

‘Community law therefore not only imposes obligations on individuals but is also intended to confer upon them rights which become part of their legal heritage. These rights arise not only where they are expressly granted by the Treaty, but also by reason of obligations which the Treaty imposes in a clearly defined way upon individuals as well as upon the Member States and upon the institutions of the Community.’

Due to the individuals’ lack of standing under the CJEU by means of the infringement procedure (articles 258, 259 and 260 TFEU), the preliminary reference procedure becomes the procedural mechanism to safeguard the rights granted by the EU law to the individuals: the citizens’ infringement procedure.P. Pescatore, ‘Van Gend en Loos, 3 February 1963 – A view from within’, in: M. Poiares Maduro & L. Azoulai, The Past and the Future of EU Law – The Classics of EU Law revisited on the 50th Anniversary of the Rome Treaty, op. cit., p. 7: ‘This formulation is prophetic, since we would see later on that the preliminary reference – managed with great care by the Court (always emphasising that in the framework of the preliminary ruling it is not its task, but rather the task of the national judge to solve the conflict between national law and Community law) – will effectively become the infringement procedure for the European citizen’; B. de Witte, ‘The Impact of Van Gend en Loos on Judicial Protection at European and National Level: Three Types of Preliminary Questions’, Conference Proceedings of 50th Anniversary of the judgment in Van Gend en Loos, Court of Justice of European Union, 13 May 2013, p. 93-102. They are vested with the powers to circulate freely within the European territory and to receive the same treatment as nationals of the host country (i.e. non-discrimination principle). The emergence of this catalogue of rights conferred on individuals by the Treaties place the CJEU as the paramount institution entitled to protect them against the infringements of the Member States, as long as they convince the national courts to lodge a reference for preliminary ruling.

Turning to Schumacker’s case, the Court granted Mr Schumacker the right to take into account his personal and family circumstances in his personal income tax in Germany, although his residence was in Belgium. By means of a request for a preliminary ruling, the CJEU could challenge German provisions that jeopardized the right of Mr Schumacker to move freely in the Union. In this respect, the principle of effective judicial protection of the rights granted by EU law guides the Court’s task within the preliminary reference procedure.A further analysis of this principle in A. Von Bogdandy, ‘Founding Principles’, in: A. Von Bogdandy & J. Bast, Principles of European Constitutional Law (Oxford 2009), p. 32.

Accordingly, the constructive cooperation responds to this rationale. Disciplining the dialogue with the national court is justified only if the preliminary reference procedure becomes the citizens’ infringement procedure. The blurring line between interpretation and application affects the interplay between the national courts and the CJEU. In order to safeguard the rights conferred upon the individuals by EU law, the powers granted by the Court by means of article 267 TFEU must contribute to achieve this mission. For instance, the reformulation of the questions referred could play an important role to overcome the gridlock created by the unanimity rule in direct taxation.

4The Perils of the Constructive Cooperation

As shown above, the procedural framework envisaged by article 267 TFEU endows the CJEU with discretionary powers to admit, to frame and to solve the questions posed by the national courts in the field of direct taxation. The procedural framework provided the Court enough room for manoeuvre to push forward EU direct taxation. Although, the solutions reached by the Court in this field are not always well accepted by either scholars or the referring court,See for instance the critics to Case C-337/08, X Holding [2010] ECR I-01215 in D. Weber, ‘Eight points as to why the X holding judgment is incorrect’, en D. Weber & B. Da Silva, The future of cross-border group taxation (Alphen aan de Rijn 2011), p. 25-49; S. van Thiel, ‘X Holding: A denial of justice’, in: D. Weber & B. Da Silva, The future of cross-border group taxation, op. cit., p. 51-70. it is worth mentioning that the number of references has not decreased.See Annual Reports of the CJEU for the periods 2012, 2013 and 2014.
http://curia.europa.eu/jcms/jcms/Jo2_7000/. With regard to period 2013, a historical record was set: ‘In 2013 the number of references for a preliminary ruling, which rose to 450, was the highest ever’. Quite the opposite, as the CJEU is ‘fed’ by recurring requests raised by national courts acting as repeat-playersThis term comes from M. Galanter, ‘Why the haves always come out ahead: speculation on the limits of legal change?’ [1974/9-nº1], Law and Justice Review, p. 95-160. The repeat players are the institutional actors who persistently intervene in the adjudication process with the aim to influence the judicial outcomes. Instead, one shotters players occasionally pop up in the adjudication process. – Hoge Raad der Nederlanden (Netherlands), Finanzgericht and Bundesfinanzhof (Germany), Korkein Hallinto-oikeus (Finland) – on the same constant tax matters: cross-border losses, personal allowances and deductions, dividends, etc.

Whereas in certain countries such as Belgium, the largest number of questions are posed by lower courts (Tribunal de première instance, Rechtbank van eerste aanleg, Hof van Beroep) or UK (High Court of Justice), in other Member States, due to the complexity of direct taxation, supreme courts are the ones which address references to the CJEU: Conseil d’État in France, Hoge Raad der Nederlanden in the Netherlands, Korkein Hallinto – oikeus in Finland. In Germany, the number of questions referred by the Finanzgericht is nearly equivalent in number to those raised by the Bundesfinanzhof. Although these figures display the national courts’ engagement in article 267 TFEU and the growing trust in the CJEU, in direct taxation it is frequent to find rulings of the Court which avoid giving a precise answer to the referring court. These judgements, dubbed “open-ended cases” by professor Pistone, are hazardous to the constructive cooperation.P.J. Wattel, ‘Throwing back some curves – some comments on the presentation and proceedings of the 7th GREIT International Conference, Madrid, 13 September 2012’, in: D. Sarmiento & D.J. Jiménez-Valladolid, Litigating EU Tax Law in International, National and Non-EU National Courts (Amsterdam 2014), p. 304: ‘Pasquale Pistone, in particular, got under the collar as regards “open-ended” judgments of the ECJ. He referred to the FII saga, which in his view has turned into a Brazilian soap opera telenovela and a bonanza for lawyers: in one and the same group litigation on the former UK franked investment income tax system, preliminary questions have now for the third time been referred to the ECJ (Cases C-446/04, C-35/11 and C-362/12) because the referring national judges are unable to decide the cases on the basis of the answers provided by the ECJ’. Since the CJEU replies to the national court in such vague terms, it principally defers to the referring court the final solutionSee the ‘deference cases’ within the classification of cases elaborated by T. Tridimas, ‘Constitutional review of member state action: The virtues and vices of an incomplete jurisdiction’, op. cit., p. 739. and therefore, it renders useless the reference to the Luxembourg court.

With regard to the treatment of dividends received by a resident company from a non-resident company (inbound dividends), the Test Claimants in the FII Group sagaCase C-446/04, Test Claimants in the FII Group (I) [2006] ECR I-11753; Case C-35/11, Test Claimants in the FII Group (II) [not yet reported] dated 13 November 2012 and Case C-362/12, Test Claimants in the FII Group (III) [not yet reported] dated 12 December 2013. is a good example. The High Court of Justice of England and Wales asked whether or not it is contrary to the freedom of circulation of capital to apply for foreign-source dividends – a credit method to eliminate double taxation – whereas for nationally-sourced dividends the exemption method applies. In the second reference,See Case C-35/11, Test Claimants in the FII Group (II) [not yet reported] dated 13 November 2012. the Court reached a complex solution that begets practical implementation problems on the grounds that national courts are urged to investigate in any particular case where the following requirements meet:

‘Articles 49 TFEU and 63 TFEU must be interpreted as precluding legislation of a Member State which applies the exemption method to nationally-sourced dividends and the imputation method to foreign-sourced dividends if it is established, first, that the tax credit to which the company receiving the dividends is entitled under the imputation method is equivalent to the amount of tax actually paid on the profits underlying the distributed dividends and, second, that the effective level of taxation of company profits in the Member State concerned is generally lower than the prescribed nominal rate of tax.’

Perhaps a better solution that could have prevented the saga would be to have asserted that only one method to eliminate double taxation is permitted for both types of dividends, either the exemption or the credit method.A. Martin Jimenez, F.M. Carrasco Gonzalez & A. Garcia Heredia, ‘Jurisprudencia del Tribunal de Justicia de la Unión Europea’ [2013/ 157] REDF, p. 329-383. Unlike this proposed solution, the Test Claimants in the FII Group saga is not closed yet and surely, the Court will receive new requests for preliminary rulings to sort out the implementation of the adopted solution. In Marks & Spencer,Case C-446/03, Marks & Spencer [2005] ECR I-10837. the Court refers to the concept of ‘final losses’ – ‘where there are no possibilities for those losses to be taken into account in its State of residence for future periods’ – that leaves unsolved questions like: Do final losses require that the subsidiary company undergo a winding-up process? Which regulation applies to account for the final losses? These open questions laid down in the Marks & Spencer judgment triggered a complex procedural iter upon its reception in UK.P. Farmer, ‘The English Courts and the application of EU Tax Law’, in: D. Sarmiento &
D.J. Jiménez-Valladolid, Litigating EU Tax Law in International, National and Non-EU National Courts (Amsterdam 2014), op. cit., p. 49-59.

The Court’s reluctance to give a precise and concrete answer that hold clear normative criteria (outcome and guidance cases, in Tridimas’s terms) stems from the fact that in direct taxation there is a need to accommodate sensitive national interests. However, these ‘open-ended’ cases which do not satisfy the expectations of the referring court jeopardize the coherence of the EU legal system, threatening the constructive cooperation designed by article 267 TFEU.

5Conclusion

This contribution seeks to explore the interplay between the CJEU and the national courts in the field of direct taxation. The traditional legislative blocking in this field enables the Court to become the institutional actor in charge of safeguarding the rights granted by EU law to individuals. Therefore, the procedural framework laid down in article 267 TFEU responds to this relevant aim.

Challenging the separation of functions envisaged by the Costa v. Enel formula, I made the claim that the Court holds discretionary powers to pursue the developments of EU direct taxation by undermining the initial monopoly of the referring court to pose the question. The constructive cooperation places the Court becomes the peak of the European Judicial Power.The concept of European Judicial Power is coined in P. Pescatore, The Law of integration, op. cit., p. 91.

Insofar as the constructive cooperation is not only present at this particular field of direct taxation, this contribution intends to encourage EU law scholars to look into the interaction between their substantive fields and the preliminary reference procedure. In my opinion, the dialogue between the national courts and the CJEU still meets the test for a fruitful area in which to carry out research.