EC law and direct taxation: l’Europe des juges et des contribuables Prof. Dr. Peter J. Wattel Advocate-General in the Netherlands Supreme Court (Hoge Raad der Nederlanden) and professor of EC Tax Law at the University of Amsterdam (UvA). Legal basis for fiscal integration Indirect taxes (trade taxes, transaction taxes) are specifically addressed in the EC Treaty. The draughtsmen of the EEC Treaty realized that (i) taxes on the border crossing of goods and services, and (ii) uncoordinated transaction taxes were major impediments to the free movement of goods and services. Indirect taxes therefore had to be either uniformized (customs duties; Articles 23-28) or harmonized (turnover taxes, excises, and energy taxation; Articles 90-93). Thus, far-reaching harmonization or uniformity has been reached at EC level in respect of indirect taxes: the Community Customs Code, the Recast VAT Directive, and the Horizontal Excises Directive, leaving Member States little national sovereignty. Direct taxes (taxes on income and wealth) are not addressed in the EC Treaty, except in Article 293, inviting the member States to negotiate, but only as far as necessary, inter alia, bilateral treaties to prevent double taxation. This provision has no direct legal effect. Either it never entered the minds of the draughtsmen of the EEC Treaty that national direct tax measures might be relevant for the functioning of a common market, or it did, but in the latter case, either their perception of a common market was quite different from what is viewed as an internal market nowadays, or they did not want the EEC to interfere with direct taxes. In principle, therefore, competence as regards direct taxation remained with the Member States. In fact, however, not so much competence seems to be left as a result of extensive interpretation of the EC Treaty Freedoms by the ECJ. When the Member States signed up to the free movement rights, they never realized what they got themselves into in that respect; when they later did, they did not, however, change the Treaty (though several attempts were made, especially to exclude direct taxes from the competence of the ECJ), so one must presume the Member States are not totally unhappy, al least not all of them (yet), with the case law of the ECJ on direct tax matters. Indirect taxation; positive integration; implementation problems; tax law rather than EC law Because of the marked differences in legal basis for, and degree of integration, the ECJ's case law in indirect tax matters is different in character from its case law in direct tax matters. For indirect taxation, a large body of very technical, I. II. 17 very detailed secondary EC has been enacted (positive integration). The indirect tax cases brought before the ECJ therefore mainly concern implementation problems, i.e. the interpretation of these very detailed, very technical EC rules on indirect taxation. They concern the question of whether the national detailed implementing measures fully correspond with the detailed indirect tax directives and regulations at EC level. They are hardly ever on the significance of the Treaty Freedoms for indirect taxation. This part of EC tax law is really tax law: the rules to be interpreted and applied are detailed rules of taxation. Direct taxation; negative integration; EC Treaty law rather than tax law In the field of direct taxes, it is the other way around. Because of the lack of legal basis for positive integration and because of the member States’ reluctance to give up sovereignty as regards direct taxes, there is hardly any secondary EC law on direct taxation. As a consequence, there are no detailed EC rules to be interpreted in that field. Direct tax cases before the ECJ thus rarely concern the implementation of an EC Directive. Rather, they concern the clash between the EC Treaty Freedoms and principles on the one hand, and unharmonized national tax laws on the other. They lead to negative integration: national measures are declared incompatible with the Treaty Freedoms. Consequently, direct tax issues before the Court do not so much concern tax law as they concern Community law. The rules to be interpreted and applied are not rules on taxation, let alone detailed rules. Rather, they are the very general, sweeping EC Treaty rules of principle, such as market access, market equality, proportionality, abuse of rights, nondiscrimination, nonrestriction, free competition, effectiveness of EC law, etc. By contrast, the national direct tax rules to be tested against these principles are often excessively technical and detailed. This extreme difference in abstraction level of the two bodies of law having to cohabit, makes negative tax integration extra complex and chaotic, more so because there is also a third, sophisticated set of rules involved: the bilateral tax treaty network between the Member States, based on the extremely successful OECD Model Taxation Convention and its official commentary. Before being able to say whether the Treaty Freedoms oppose a national direct tax measure, it is necessary to first understand the mechanics and effects of, and the principles behind, that measure, and of possible applicable bilateral tax treaty provisions. This may entail having to study and interpret very technical and detailed parts of national and international tax law, both of which the ECJ is not competent to interpret. In that respect, the ECJ is dependent on the accuracy of the briefing by the referring national Court, by the parties in the main proceeding, by the government involved, and by the intervening other governments. III. 18 Peter J. Wattel Limits to negative integration in tax matters The EC Treaty being silent on direct taxes, the only EC law instruments available are the general Treaty rules, especially the free movement and free competition rights. Their impact on direct taxation depends on how extensively they are interpreted. In that respect, a cyclical pattern is visible in the Court’s case law:1 hesitance at first, until 1994 (e.g. Cases 204/90 Bachmann, and 270/83 Avoir Fiscal), followed by a period of outright tax activism (1995-2004; e.g. Cases C-319/02, Manninen, C-385/00, De Groot, and C-168/01, Bosal Holding), and lately, since 2005, judicial restraint and a certain deference to the fiscal sovereignty of the Member States (e.g. Cases C-376/03, D. v Inspecteur, C-513/04, Kerckhaert-Morres, C-446/04, Test Claimants in the FII Group Litigation, and C-470/04, N. v Inspecteur). The recent restraint may be explained by the fact that the Court has been faced – especially by British Courts dealing with Group litigations and asking long lists of extremely technical questions – with the consequences of its brave, but fiscally not always robustly reasoned previous case law in direct tax matters. I consider the Court’s recent judicial restraint to be appropriate, for two reasons: (i) the EC Treaty offers no clue as to a choice between, e.g., the source principle of taxation and the residence principle of taxation, or between methods of prevention of double taxation, nor to the question of which of two States must take away any double taxation or mismatch if unilaterally neither of them is discriminating against the cross-border position as compared to the domestic position. Neither does the Treaty provide the Court with any lead to choose a ‘best practice’; (ii) the Court should not get itself stuck in a legal quagmire to a point where it will be unable to withdraw without either having to overstep its competence, thus jeopardizing its position, or having to leave international tax law in desperate disorganization. The Court’s possibilities in direct tax matters are not to create, but only to destruct. It can only tell us whether the EC Treaty opposes a certain national measure. It cannot tell us which measure should replace it. Where no legislation at EC level exists, there is no choice but to leave legislation to the Member States. A restrictive national measure may be unavoidable to protect an overriding reason of public interest enumerated in the EC Treaty or developed by the Court (such as the preservation of a balanced interjurisdictional allocation of taxing rights, or fiscal symmetry between deduction and taxation within the same taxing jurisdiction, or between profits and corresponding losses). The judiciary cannot solve political problems. It cannot and should not make policy choices and budget division choices which have been entrusted to chosen political decision-making bodies. The Court is not competent to create, allocate or divide taxing power, nor to prescribe which of IV. 1 See Servaas van Thiel: The Direct Income Tax Case Law of the European Court of Justice: Past Trends and Future Developments, and Peter J. Wattel: Judicial Restraint and Three Trends in the ECJ’s Direct Tax case Law, in: 62 Tax Law Review 1 (Fall 2008), New York University School of Law, pages 143 and 205, respectively. EC law and direct taxation: l’Europe des juges et des contribuables 19 two concurring jurisdictions should take away any double taxation – let alone in which detailed manner. Two-jurisdiction problems, such as differences (mismatches) in income characterization, in transfer pricing rules, in rules on income attribution to persons, etc. (disparities), and problems resulting from the division of the tax base over more than one jurisdiction (dislocations), can only be remedied by making political choices. The Court cannot do away with the negative tax effects of the 'exercise in parallel' of two taxing powers (see Case C-513/04, Kerckhaert-Morres). I submit the Court was wrong in holding (in case C-279/93, Schumacker) that “discrimination arises from the fact that (..) personal and family circumstances are taken into account neither in the State of residence nor in the State of employment.” This was incorrect, as the fact that something does not happen in either State, is not a discrimination in itself, and in any case the Court has no competence to pick and choose the State which in its view should make it happen. In Schumacker and De Groot, it even made the wrong choice, for that matter. Likewise, the fact that something does happen twice, in both States (taxation), is not a discrimination either, as the Court acknowledged in Kerckhaert-Morres. Sometimes, the Court cannot withstand the temptation to blame a convenient State, although that State did not exercise its taxing jurisdiction (as in the Cases C-385/00, De Groot, and C-446/03, Marks & Spencer II). Moreover, the tax law in one Member State is not necessarily comparable to the tax law in another Member State. Decisions such as De Groot, Schumacker, and Marks & Spencer II, allocating deduction of allowances or losses, presuppose that both States involved have more or less the same tax system (e.g. both allowing deduction of personal allowances to individuals, or both allowing comparable loss relief roll over, etc). But that may not to be the case at all: one Member State may decide not to allow any personal deductions in return for a low flat tax rate; or it may have no loss roll-over in exchange for a lower corporation tax rate. In Case C-527/06, Renneberg, one of the States involved allowed mortgage interest deduction and the other State did not; in Case C-157/07, Krankenheim Ruhesitz am Wannsee, one of the States involved allowed loss relief roll over to other tax years; the other State did not. The Treaty Freedoms do not make one Member State responsible for a lack of tax relief in another Member State, as the Court correctly held in Case C-403/03 Schempp. Fiscally, a cross-border position (a two-jurisdiction position) is not the same as a purely domestic (one-jurisdiction) position: being exposed to another jurisdiction or to two jurisdictions at the same time is not comparable to being exposed to just one jurisdiction, much as this is difficult to accept in a (fiscally nonexistent) single internal market. As the Court recognizes in recent case law: it is solely the Member States’ competence to assume and define their jurisdiction in direct tax matters. Also in the field of parafiscal levies, the Court’s recent case law shows a return from earlier activism: In Case C-438/04, Mobistar, concerning taxation of telecommunication services, the Court apparently realized that its earlier Cases 20 Peter J. Wattel C-430 and 431/99, Sea Land Services, implied that every tax is a restriction of free trade and needs to be justified by the Member State levying it, resulting in the Court having to judge on the reasonableness of any national tax. But that is the exclusive competence of the Member States. In Mobistar, the Court therefore returned to a purely discrimination-based approach of such parafiscal levies. In direct tax cases, however, the Court is still reluctant to give up competence already at the first question (is there a restriction?), because if it does not find a restriction, it is not in a position to require any justifications, nor to judge the proportionality of the measure to safeguard that justification. Although no discrimination was present in cases like C-414/06, Lidl Belgium, C-446/03, Marks & Spencer II, and C-231/05, Oy AA, the Court nevertheless insisted on finding one, but subsequently considered it justified. This judicial policy, especially in Lidl Belgium, where no difference whatsoever exsisted between the cross-border position and the comparable domestic position, may be explained by the Court’s apparent wish to keep the Member States on their toes, i.e. under a more or less constant obligation to justificaty their tax measures, and to judge the proportionality of national tax measures. This approach led the Court, in Case C-446/03, Marks & Spencer II, to force the UK to extend its taxing jurisdiction to foreign losses of nonresident subsidiaries without any presence within the UK taxing jurisdiction. With this policy of first finding a “restriction” or “discrimination” and subsequently accepting justifications for it, instead of immediately finding subject-to-tax and not-subject-to-tax incomparable, the Court forces itself to accept justifications (in cases such as C-414/06, Lidl, C-446/03, Marks & Spencer II and C-231/05, OY AA), which make no sense in the cases decided, such as the need to prevent double dips or to prevent abuse. There was no question of double dips or abuse in these cases. Only rarely the Court accepts that outside a taxing jurisdiction and within a taxing jurisdiction are incomparable positions, as it did in Case C-403/03, Schempp. Einzelfallgerechtigkeit; apparent inconsistency Not even the most basic of fiscal choices being made in the EC Treaty, such as the choices between source or residence, credit or exemption, territoriality or worldwide taxation, capital import neutrality or capital export neutrality, and virtually no secondary EC law being available, the case law of the ECJ in direct tax cases necessarily mainly follows EC Treaty law, which is unsuitable for deciding international tax cases. The case law consists of Einzelfallgerechtigheit, lacking focus on implications for other cases and for other tax systems; it is often inconsistent. There are by now over one hundred direct tax cases, handed down in a period of about 22 years, and some 40 more cases pending. At the current rate, 30 or more direct tax cases will be decided each year in the V. EC law and direct taxation: l’Europe des juges et des contribuables 21 next few years. Obviously, under these circumstances, it is difficult to keep up stare decisis. The Court makes it extra difficult to read its cases because it is unexplicit about repeal or modification of previous decisions. Two examples out of many: as regards the deductibility of cross-border pension and annuity contributions, Case C-204/90, Bachmann, was clearly overturned by Case C-150/04, Commission v. Denmark, but the Court did not acknowledge its apparent change of mind in any way. In Case C-385/00, De Groot, the Court was explicitly satisfied by the unilateral neutralization, in one State, of a (perceived) fiscal discrimination in the other State; in Case C-379/05, Amurta, by contrast, the Court, without providing any reason for this marked contrast, explicitly required a bilateral (tax treaty) provision to neutralize the restriction found. It held explicitly that unilateral neutralization in the other State was not enough. This is all the more puzzling as to the addressee of the Treaty Freedoms (the taxpayer), it is irrelevant whether he gets a tax credit on the basis of the national law of his home State, or on the basis of a tax treaty between his home State and the source State. To illustrate the difficulty for tax policy makers, taxpayers and tax administrations to fathom the Court’s direction in direct tax cases, some apparent inconsistencies are listed below: – sometimes the Court recognizes that being subject to tax in one jurisdiction is not comparable to being subject in another jurisdiction or in two jurisdictions at the same time (as in the Cases C-403/03, Schempp, C-513/04, Kerckhaert-Morres, C-231/05, Oy AA and C-157/07 Krankenheim Ruhesitz am Wannsee). In these cases, the Court held that one State cannot be required to compensate for the fact that the other State applies a different system of taxation (e.g. does not allow certain deductions or loss relief). But in other instances, the Court squarely equates subject-to-tax and not-subject-to-tax within the same jurisdiction, as in the Cases C-168/01, Bosal Holding, C-446/03, Marks & Spencer II, C-385/00, De Groot, and C-527/06, Renneberg, forcing the Member States involved to take into account expenses, losses or allowances which were clearly not attributable to their taxing jurisdictions. In De Groot, the Court even required the residence State to compensate for the fact that the source State was overtly discriminating against Mr De Groot by refusing him the same personal deductions as residents solely because he was a nonresident; – for incorporated undertakings, the Court allows immediate taxation upon emigration (Case 81/87, Daily Mail), for unincorporated undertakings, on the other hand, it does not even allow a merely conservatory assessment with security for payment upon realization (Cases C-9/02, De Lasteyrie du Saillant, and C-470/04, N. v Inspecteur); – usually the Court allows restrictive anti-abuse measures only if they are specifically targeted at ‘wholly artificial arrangements’ and actual abuse is demonstrated by the tax administration (e.g. Cases C-196/04, Cadbury 22 Peter J. Wattel Schweppes, and C-264/96, ICI); but sometimes even a remote and abstract possibility of abuse is already enough to justify a restrictive measure in a case in which manifestly no actual abuse was present (e.g. Cases C-446/03, Marks & Spencer II, C-231/05, Oy AA, and C-414/06, Lidl Belgium). – sometimes branches and subsidiaries must be treated alike (e.g. Cases C-307/97, Saint-Gobain, C-330/91, Commerzbank, C-311/97, Royal Bank of Scotland); but sometimes Member States may – or even must – treat them differently (Cases C-168/01, Bosal Holding, and C-298/05, Columbus Container); – in Case C-168/01, Bosal Holding, the Court did not allow an exclusion of deduction of financing expenses for non-subjected (and therefore mostly foreign) subsidiaries, even though such exclusion was fiscally quite coherent, and Article 4 of the EC Parent-Subsidiary Directive 90/435/EEC allowed such exclusion; in Case C-446/04, FII GLO, however, the Court did allow, on the basis of the very same Article 4, an inconsistent and overtly discriminatory measure which exempted domestic subsidiary dividends, but only provided indirect credit for foreign subsidiary dividends; – where Member States put forward administrative difficulties in order to justify a distinction between domestic and cross-border positions, the Court hissed them down (e.g. Cases 204/90, Bachmann, and C-18/95, Terhoeve), but where the Court itself needed a justification for distinguishing between residents and nonresidents, administrative difficulties were posited by the Court as a valid argument (Schumacker, De Groot); – usually the Court requires the source State of the income to allocate necessary expenses to its jurisdiction (Cases C-324/01, Gerritse, and C-345/04, Centro Equestre da Leziria Grande), but sometimes the Court requires such expenses to be allocated to the non-source State (Case C-527/06, Renneberg). – sometimes foreign dividends may not be discriminated against (Cases C-319/02, Manninen, and C-315/02, Lenz), sometimes they may be discriminated against (Cases C-446/04, FII GLO, and C-513/04, Kerckhaert- Morres), apparently because the Court views international juridical double taxation and international economical double taxation as fundamentally different (see also the before-last indent below); – sometimes the Court accepts that a Member State cannot be required to abandon its claim on income sourced within its territory, as in Case C-374/04, Class IV ACT (“a Member State cannot be obliged to abandon its right to tax a profit generated through an economic activity undertaken on its territory”) and in Case C-336/96, Gilly (no refund of foreign tax at the cost of tax on domestic income), but sometimes the Court forces Member States to give up tax revenue on domestically sourced income to compensate for losses that are not connected to their territory or taxing jurisdiction, as in C-446/03, Marks & Spencer II and C-293/06, Deutsche Shell). EC law and direct taxation: l’Europe des juges et des contribuables 23 – sometimes the Court accepts territorially consistent taxation (as in the Cases C-414/06, Lidl Belgium, C-231/05, Oy AA, and C-157/07, Krankenheim Ruhesitz); sometimes it does not (as in the Cases C-168/01, Bosal, and C-446/03, Marks & Spencer II); – normally, the Court does not, under its proportionality test, accept a restrictive measure if there is a less restrictive measure which is capable of protecting the mandatory requirement of public interest which justifies the restriction of free movement. In Case C-414/06, Lidl Belgium, however, the Court surprisingly and contrary to the opinion of the advocate-general, allowed Germany to choose the most restrictive measure (base exemption) out of two possible measures (base exemption or tax exemption), even though it was common ground that a less restrictive measure existed (tax exemption) and that six other Member States (and even Germany itself, in the past) actually applied such less restrictive measure without any difficulty; – a prepayment in the form of foreign corporation tax must be credited as if it were a domestic prepayment (C-319/02, Manninen), but a prepayment in the form of a foreign dividend withholding tax does not need to be credited as if it were a domestic withholding tax (C-513/04, Kerckhaert-Morres and C-194/06, Orange European Small Cap Fund). This raises the question why the Court accepts international juridical double taxation (Orange Small cap; Kerckhaert-Morres), where it does not accept economical double taxation (Manninen, Lenz), except when caused by the source State (see Case C-374/04, Class IV ACT). The economic difference between the two is negligeable: as the Advocate General Geelhoed observed in his Opinion in Case C-446/04, FII GLO (paragraph 5): in an imputation system such as in Manninen, "corporation tax serves as a prepayment for (…) income tax.” Moreover, to the taxpayer it is irrelevant whether the international double burden he suffers as a result of the border-crossing of his dividends is called “economical” or “juridical.” To add to the confusion, in Case C-446/04, FII GLO, the Court required the UK to credit both the foreign corporation tax (economic double taxation) and the foreign withholding tax (juridical double taxation). – In Case C-446/04, FII GLO, the Court’s first rejected discriminatory taxation of foreign dividends, but subsequently accepted a system that overtly discriminates against foreign dividends: (para 46): “It is thus clear from case-law that, whatever the mechanism adopted for preventing or mitigating the imposition of a series of charges to tax or economic double taxation, the freedoms of movement guaranteed by the Treaty preclude a Member State from treating foreign-sourced dividends less favourably than nationally-sourced dividends, unless such a difference in treatment concerns situations which are not objectively comparable or is justified by overriding reasons in the general interest (...).” 24 Peter J. Wattel (para 53): “(…) [T]he mere fact that, compared with an exemption system, an imputation system imposes additional administrative burdens on taxpayers, with evidence being required as to the amount of tax actually paid in the State in which the company making the distribution is resident, cannot be regarded as a difference in treatment which is contrary to freedom of establishment, since particular administrative burdens imposed on resident companies receiving foreign-sourced dividends are an intrinsic part of the operation of a tax credit system.” It would seem that this last paragraph says that the mere fact that foreign dividends are discriminated against, cannot be regarded as discriminatory, as such discrimination is intrinsic to the operation of a discriminatory system. Confusing is also the fact that the Court in its decisions regularly refers, as authority, to previous decisions which, however, do not visibly support the referring decision. The impalpability of the ECJ’s direct tax case law is further exacerbated by the introduction, abandonment and reintroduction of elusive concepts that do not exist in international tax law, such as “exercise in parallel of taxing power” (disparity?), “fiscal cohesion”, and “balanced allocation of taxing power” (legitimate tax base protection? Territoriality? Source country entitlement?). This phenomenon may be illustrated by the full circle made by the ‘fiscal cohesion’ concept, which apparently means that causally linked positive and negative elements of the same tax base (like profits and losses, or income and necessary expenses) should be matched within the same taxing jurisdiction. First, this idea was called “fiscal cohesion” (204/90, Bachmann); in later cases, it was referred to as “the fiscal principle of territoriality” (C-250/95, Futura; C-446/03, Marks & Spencer II, C-470/04, N. v Inspecteur); it went on to become the “preservation of a balanced allocation of taxing rights’ (C-446/03, Marks & Spencer II; C-231/05, Oy AA), and the principle of “symmetry” (C-446/03, Marks & Spencer II, C-157/07, Krankenheim Ruhesitz am Wannsee), to finally return to the characterization “cohesion of the tax system” (C-157/07, Krankenheim Ruhesitz am Wannsee). All of these terms apparently mean the same thing: tax base integrity for the source State. Why not call it that? It should be observed that in the process, the court apparently, but implicitly, has made a choice for prevalence of the source principle of taxation over the residence principle. This may be welcomed, and even applauded, as it generally corresponds with (OECD) international tax law, but it is difficult to see the legal basis in the EC Treaty. EC law and direct taxation: l’Europe des juges et des contribuables 25 But the Member States are to blame just as much; not providing guidance produces a loss of sovereignty and an Europe des juges et des contribuables The above has been critical of the apparent inconsistencies in the Court’s direct tax case law. To the Court’s defence, it should be observed that the Court has little choice. It must decide cases referred to it. It does not have the power to say, like one of the brethren in the US Supreme Court has been reported to say, looking at a petition for a writ of certiorari: “This is a tax case; deny!” I imagine the ECJ sometimes might dearly have wanted the power to deny a writ of cert for tax cases, especially British group litigations. Moreover, the Member States (the Council) are just as responsible for the lack of consistency as the Court. One should be just as critical of the Council’s lack of guidance as of the Court’s wandering. Apparently, the Member States still consider the political price of necessary harmonization of direct tax law too high, or at least higher than the political price of not harmonizing direct tax law. That is remarkable, as the drawbacks of sitting duck in the negative integration theatre are many and obvious: – the lack of harmonization and coordination invites bad tax competition (tax dumping; tax degradation), tax avoidance and tax fraud; – individuals and companies do not know their tax position upon emigrating or taking up a job or a business across the border; – national tax administrations do not know their positions either; – EU industry must comply with 27 substantively and procedurally different tax systems; in terms of administrative burden and risk of double taxation, this is a depressing competitive disadvantage as compared to Europe’s competitors with home markets in the US, Japan or China; – the lack of secondary EC law on direct tax matters is asking for trouble: the ECJ is forced to decide on extremely technical questions, often involving application of bilateral tax treaties and detailed anti-avoidance rules, with no other hold than sweeping statements like free movement, non-discrimination, level playing field, mutual recognition, effectiveness of EC law, etc., which are inadequate (too vague) to decide on the question, e.g., whether currency exchange losses on the Italian lire, suffered in an Italian permanent establishment of a German company, the profits and losses of which establishment are exempt in Germany by the tax treaty between Germany and Italy, must be attributed to the permanent establishment, or to the head office, or to the entrepreneurial decision not to hedge the currency risk (which also entailed the good chance of an exempt currency profit); – lack of harmonization and uncertainty on the impact of the Treaty freedoms gives national tax legislation the character of budgetary gambling: having no other hold than the same vague and general EC Treaty statements, national governments are forced to gamble on the outcome of preliminary questions proceedings during the period it takes the ECJ to answer preliminary questions from their courts: they must decide on whether or not to VI. 26 Peter J. Wattel enact legislation in anticipation of an unfavourable answer, or to do nothing. In both cases, serious budgetary consequences may ensue if they get it wrong; – with over one hundred of direct tax cases already handed down, some 40 more pending, and an increasing number per year to be expected, it is becoming increasingly difficult to predict the compatibility of new tax legislation with the Court’s interpretation of the Treaty Freedoms; – Because of the unpredictable and uncontrollable budgetary effects of ECJ decisions, national treasuries may be forced (i) to economize elsewhere, and/or (ii) to increase taxes elsewhere, and/or (iii) to increase the budget deficit (with possibly stability pact based sanctions, which will increase the deficit, which will lead to new stability pact sanctions, which … (etc.)); – The EC Commission is unable to develop any consistent positive integration policy, as it is forced to confine itself to following up on haphazard ECJ decisions on coincidental issues put forward by taxpayers in national courts. As Directive proposals do not stand a chance of acquiring the required unanimity, the Commission can only issue ‘communications’ (soft law), send letters to individual Member States and possibly subsequently refer them to the ECJ in infringement proceedings against them if they do not comply. The remarkable effect of all this is exactly what the Member States do not want to happen: a loss of fiscal sovereignty. The cherishing of national sovereignty produces an Europe des juges et des contribuables. The Member States’ policy of not having any direct tax policy gives their sovereignty away to the market: to the taxpayer and the judiciary. More often, more aggressively and more creatively, taxpayers off late rely on EC law for purposes that may have very little to do with European ideals. This occurs in national courts, which have their own policy of asking or not asking preliminary questions, and of framing these questions. The ECJ in its turn has its own way of answering them, without any parliamentary control on budgetary consequences, without any appeal and, most importantly: without the possibility for the EC legislator to legislate otherwise, as the Court’s interpretation of primary EC Law (the Treaty Freedoms) is primary EC law as well, and cannot therefore be superseded by secondary EC law such as a Directive. Consequently, politics may only be able to reclaim the areas lost to the market and to the judiciary by amending the EC Treaty, which needs 27 ratifications. We know how well that works. At the beginning of the chain of events that leads to a preliminary referral, or to a Commission infringement action, is a taxpayer. Therefore, all progress depends upon the taxpayer: European direct tax policy has effectively been privatized to a large extent. The remainder of direct tax policy is in the hands of the judiciary as long as the Member States do not act. EC law and direct taxation: l’Europe des juges et des contribuables 27 Upshot: harmonization is becoming increasingly difficult As observed, even if the EC-legislator would act, he could not divert from existing direct tax case law (the acquis), as it forms primary EC law. Unlike at national level, at EC level there is no political body that may undo (the consequences of) case law. That would mean that the only option – disregarding Treaty renegotiations - at EC level is codifying the ECJ’s case law in direct tax matters. But that is very, very difficult, as it is inconsistent. The legislator can only enact one ECJ line of cases by disavowing another line of cases. Which line then to choose? This may, however, be less of a problem than it seems, as the Court appears to apply a double standard: the Court is rather strict and unforgiving where it is testing restrictive national law against primary EC law, but it is less strict where Member States have taken the trouble of agreeing on a bilateral tax treaty (see, e.g., Case C-379/05, Amurta), and it is simply lenient where secondary EC law has been enacted. Even a discriminatory system such as the UK system of relief for economical double taxation (exemption for domestic group dividends, but credit for underlying tax for foreign group dividends) was in principle accepted by the ECJ (in C-446/04, FII GLO), apparently because secondary EC law (the EC Parent-subsidiary Directive 90/435/EEC) allowed Member States a choice between these two methods of prevention of double taxation. In fact, it would seem that in FII GLO the Treaty Freedoms were interpreted by the Court to conform to Article 4 of the Directive, instead of the other way around. Nevertheless, however benevolent the Court may view brave attempts at codification in secondary EC law of its case law, it will not be easy, and at any rate, neither the Member States nor European industry seem to entertain an urgent need for it. VII. 28 Peter J. Wattel