Bloomberg Tax
Feb. 26, 2020, 8:01 AM

The CJEU Judgments in the Danish Beneficial Ownership Cases—To Be, Or Not To Be

Pascal Faes
Pascal Faes
Bloomberg Tax International Forum

For decades, the Bloomberg Tax International Forum has convened international tax experts from all over the globe. In this series, Forum members will regularly share their views on current and important topics. Read the full archive of the Bloomberg Tax International Forum articles here.

The Grand Chamber of the Court of Justice of the European Union (the CJEU or the Court) rendered two groundbreaking decisions in the so-called Danish beneficial ownership cases (Cases C-115/16, C-118/16, C-119/16 and C-299/16, N Luxembourg 1 et al., on the Interest and Royalties Directive (IRD) and Cases C-116/16 and C-117/17, T Danmark et al., on the Parent-Subsidiary Directive (PSD)), on February 26, 2019.

At issue in these cases is the source taxation of interest and dividends paid by various Danish companies to their EU parent companies. Those EU companies were themselves held by third-country funds, partnerships or corporations, and had obviously been interposed following legislative changes in Denmark (introducing withholding taxation of cross-border interest payments) and in the U.S. (permitting tax-favorable repatriation of foreign profits), respectively.

Four of the six cases involved back-to-back financing transactions, under which a Danish resident subsidiary was financed by its nonresident parent company via a series of loans granted to intermediary EU holding companies in Luxembourg and Sweden. The other two cases concerned dividend distributions by Danish companies to intermediate EU holding companies in Cyprus and Luxembourg. In all applicable (bi- and multilateral) tax treaties between Denmark on the one hand, and Sweden and Luxembourg on the other, there was no source tax on interest, while the treaties with Cyprus and Luxembourg provided for a reduced 10% and 5% withholding tax on dividends, respectively, in each case under the condition that the recipient be the “beneficial owner” of that income.

In its decisions, the CJEU, inter alia:

(i) gives its interpretation of the concept of “beneficial owner” and the effect of (the Commentaries on) the Organization for Economic Co-operation and Development (OECD) Model Tax Convention thereon;

(ii) expands the general anti-abuse principle enshrined in EU law to areas of tax law that are subject to minimum harmonization; and

(iii) attempts to shed light on the constituent elements of “abuse.”

This Insight is intended to provide a critical analysis of these points. In addition, it will focus on what the author perceives as the Court’s conflation of the beneficial owner concept with the general anti-abuse principle.

CJEU Concept of Beneficial Owner—Effect of (the Commentaries on) the OECD Model Tax Convention

In the Danish (interest) cases, the CJEU emphasizes, at the outset, that the term “beneficial owner” “cannot refer to concepts of national law which vary in scope.” A comparative analysis of (EU and non-EU) national court decisions on beneficial ownership makes this variance in scope overly plain: the concept of beneficial owner is interpreted by certain national courts as reminiscent of, or close to, a “substance over form” approach (see e.g. the U.K in the Indofood International Finance Ltd. V. JP Morgan Chase Bank case; Switzerland in the Swiss total return swaps, SMI index futures and dividend stripping cases; Spain in the Real Madrid FC vs Oficina Nacional de Inspección case—see also the recent resolution adopted by Spain’s Central Economic Administrative Court (CEAC) of October 8, 2019 on the basis of criteria set out by the CJEU in the Danish beneficial owner cases; France in the Ministre de l’Economie, des Finances et de Industrie v Société Bank of Scotland case); while at the other end of the spectrum certain other national courts tend to adopt a strict legalistic (“form over substance”) approach (e.g. Canada in the Prévost and Velcro cases; the Netherlands in the Royal Dutch Shell case).

Rather, therefore, in the CJEU’s analysis, the “beneficial owner” concept (in the IRD) must be given an autonomous EU law meaning. Clearly, this has ramifications for the member states’ tax treaties, namely where such treaties contain a provision equivalent to Article 3(2) of the OECD Model Tax Convention (the OECD Model) requiring a reference to a domestic law meaning to interpret a given concept. In such instances, the autonomous EU law meaning referred to by the CJEU should in principle have precedence.

Whether this conclusion is valid as a practical c.q. pragmatic matter is doubtful, especially in cases where the national tax legislation of a member state leaves an option to the recipient of the (interest) income to rely on the IRD exemption or on a treaty-based exemption (whereby typically the application conditions of the treaty-based exemption (e.g. in terms of minimum participation requirement and minimum holding period) tend to be formulated in a less restrictive manner).

As regards the “autonomous EU law meaning” of the concept of “beneficial owner,” the CJEU acknowledges that the OECD Model and (the most current version of the) OECD Model Tax Convention Commentary (the OECD Commentary) “are relevant when interpreting the IRD,” thereby implying that these OECD materials serve as a source for interpretation. In that interpretation, a conduit company cannot be a beneficial owner. This is the case in instances where the recipient of the (interest) income, though the formal owner of such income, has, as a practical matter, very narrow powers which render such owner, in relation to the income concerned, a mere fiduciary or administrator acting on account of the interested parties.

The Court justifies the relevance of the OECD Model and the OECD Commentary when interpreting the IRD, specifically having regard to the context of the IRD’s adoption and the use in the IRD of the OECD Model’s terminology.

Specifically, the Court emphasizes that it is apparent from the March 6, 1998 Proposal (COM (1998) 67 final) that formed the basis for the IRD, that the directive draws upon Article 11 of the OECD Model and pursues the same objective, namely avoiding international double taxation.

The Court further adds, undoubtedly building and expanding on its traditional approach of evolutionary or dynamic interpretation of (EU) legal norms, that the most current version of the OECD guidance should be used to interpret the IRD, even where (and notwithstanding that) the directive was proposed in 1998 and adopted in 2003.

The Court’s “adherence” to the OECD Model and the (most current version of the) OECD Commentary as source of interpretation of the “beneficial owner” concept deviates from the analysis that Advocate General Kokott had developed in her opinion. According to the Advocate General, the concept of beneficial owner must be interpreted under EU law autonomously and independently of Article 11 (Interest Article) of the OECD Model (in its 1977 version or subsequent versions).

The reason is that OECD Models are not legally binding, multilateral conventions under international law but rather the unilateral acts of an international organization in the form of recommendations to its member countries. Even the OECD does not consider these recommendations to be binding; rather, under the OECD Rules of Procedure, the member countries must consider whether their implementation is opportune. This applies a fortiori to the commentaries published by the OECD, which ultimately only contain legal opinions.

Admittedly, the Advocate General nuances the above-referenced statements by emphasizing that it is not inappropriate for the member states to derive guidance for the balanced allocation of their fiscal competence from international practice, as reflected in the OECD Model. The same applies to guidance from any prevailing international legal opinion, which may be reflected in the Commentaries on the OECD Model.

Having said that, however, in the Advocate General’s opinion, the Commentaries on the OECD Model cannot have a direct effect on the interpretation of an EU directive, even if the terms used are identical. In that sense, those Commentaries simply reflect the opinion of the persons who worked on the OECD Model, not the views of a parliamentary legislature or indeed of the EU legislature. At most, should it transpire from the wording and history of the directive that the EU legislature was guided by the wording of an OECD Model and the Commentaries (available at the time) on that OECD Model, a similar interpretation might be appropriate.

Therefore, in the Advocate General’s analysis, if the OECD Commentaries have no direct binding effect and if the criterion applied in Article 1(4) of the IRD is whether the recipient receives the payments for its own benefit rather than as a trustee, then that is the decisive criterion (under EU law) by which to determine a beneficial owner within the meaning of Article 1(1) of the IRD. If no (possibly hidden) trust exists, then the person with the civil law claim is also the beneficial owner for the purposes of the IRD.

Against this backdrop, relying on non-EU law sources for the interpretation of EU law remains debatable. The sole fact that the Commission stated in an old proposal that some aspects of the IRD are inspired by the then applicable OECD Model (and its Commentaries), does not mean, in our view, that the (relevant) EU law definitions should lose their autonomous nature.

Extension of General Principle of Abuse of EU Law to Areas Subject to Minimum Harmonization

The referring Danish court had asked whether the beneficial ownership requirement in a treaty between member states is an agreement-based anti-abuse provision within the meaning of Article 5 of the IRD and Article 1(2) of the (pre-2015) PSD. The CJEU does not as such answer this question. Rather, the CJEU refers to the general legal principle pursuant to which EU law cannot be relied on for abusive or fraudulent ends.

To this end, the Court expressly states that “(…) even if it were to transpire, in the main proceedings, that national law does not contain rules which may be interpreted in compliance with [Article 1(2) of the PSD or Art 5 of the IRD], this — notwithstanding what the Court held in [Kofoed, see infra] — could not be taken to mean that the national authorities and courts would be prevented from refusing to grant the advantage derived from the right of exemption provided for in [Article 4 of the PSD or Art 1(1) of the IRD] in the event of fraud or abuse of rights.”

As a result, in the Court’s opinion, it is unnecessary for a domestic or agreement based anti-abuse provision to exist in abusive cases for the PSD or the IRD to be disapplied. The Court thus expands the general anti-abuse principle enshrined in EU (primary) law to areas of tax law that are subject to minimum harmonization, implying that the PSD and IRD are “controlled” by the general principle of EU law according to which EU law cannot be relied upon for abusive purposes.

The Court’s position undoubtedly builds further on its holding in the Mangold (Case C-144/04) (and subsequent case law e.g. Case C-555/07, Kücükdeveci) following which, while a principle derived from a directive cannot be invoked against an individual, an identically formulated principle can be used in this way if it constitutes a “general principle of EU law,” thereby extending the doctrine of direct effect to “general principles.” (Note: General principles of EU law constitute a genuine, autonomous source of EU law; these principles have a constitutional status and are equal, in terms of hierarchy, to the EU Treaties and are binding on the EU institutions and on the member states.)

One may question the doctrinal foundation of the expansion of the general anti-abuse principle enshrined in EU law to areas of tax law that are subject to minimum harmonization (such as the PSD and IRD). First, the proposition per se that EU law contains a general principle laying down a prohibition of abusive practices continues to be highly debated among legal authors.

Second, assuming that such general principle is accepted to exist, the question arises as to how the Danish beneficial ownership cases fit in the context of the earlier case law of the CJEU in the Kofoed case (Case C-321/05), where the Court held, with respect to direct taxation subject to minimum harmonization (in casu the Tax Merger Directive), that “a Member State which has failed to transpose the provisions of a directive into national law cannot rely, as against Community citizens, upon limitations that might have been laid down on the basis of those provisions.”

At the heart of that dictum is the consideration that as directives are unable by themselves to create obligations for individuals and cannot therefore be relied on per se by the member state as against individuals, each of the member states to which a directive is addressed is obliged to adopt, within the framework of its national legal system, all the measures necessary to ensure that the directive is fully effective, in accordance with the objective that it pursues.

It follows that if a member state has not incorporated the directive-based anti-abuse clause into its legal order, whether formally and specifically or by transposition achieved through a general legal context that does not go beyond the directive-based anti-abuse rule as interpreted by the CJEU (so that a formal and express re-enactment of the provisions of the directive in specific national provisions is not necessary), that member state cannot reduce the effect of the rest of the provisions of the directive at issue on that basis (and, hence, in such, has to accord the benefits of the directive).

It follows that, in such circumstances, recourse to the general principle of EU law prohibiting the misuse c.q. abuse of (EU) law should be barred. Applied to the case at hand, the Kofoed dictum would imply that if the legislator of a member state decides not to implement rules permitted by a directive’s anti-abuse reservation such as Article 1(2) of the (pre-2015) PSD or Article 5 of the IRD, which in the current state of affairs should in essence (still) be seen as a sovereign domestic tax policy decision, the tax authorities would not be entitled to rely on an unwritten, general EU principle to counter perceived abuse, as Article 1(2) of the (pre-2015) PSD and Article 5 of the IRD precisely constitute a concrete expression of such principle.

See also, to that effect, the Advocate General Kokott who in her opinions in the cases at hand appreciated the direct connection between a general principle prohibiting abusive practices enforceable against an individual and the limits of direct effect of directives in inversely vertical situations. According to the Advocate General, the former cannot be used to circumvent the latter. This means that a national law creating a power for national courts to combat abuse has to exist and cannot be substituted by an alleged EU law “general principle.”

In the extreme, the Court’s bypassing of the lack of horizontal direct effect of a directive by giving direct effect to the general principle of abuse of EU law might, at least conceptually, even raise the question whether, precisely because such general principle is independent from the relevant directive, the insertion of an anti-abuse reservation in a directive has (still) any merit attached to it as, in any case, absence of exercise of such a reservation by a member state would/could be overrun by the alleged (direct effect of the) general principle at issue.

Constituent Elements of Abuse

Proof of an abusive practice requires, first, a combination of objective circumstances in which, despite formal observance of the conditions laid down by the EU rules at issue, the purpose of those rules has not been achieved, and second, a subjective element consisting in the intention to obtain an advantage from the EU rules by artificially creating the conditions laid down for obtaining it. Examination of a set of facts is therefore needed to establish whether the constituent elements of an abusive practice are present, and in particular whether economic operators have carried out purely formal or artificial transactions devoid of any economic and commercial justification, with the essential aim of benefiting from an improper advantage.

While it is not for the CJEU to assess the facts in a case brought before it, the Court may however, where appropriate, specify indicia to guide the member states’ national courts in the assessment of the cases that they have to decide. It is nevertheless for the referring courts to establish whether such indicia are objective and consistent, and whether the taxpayers concerned have had the opportunity to adduce evidence to the contrary.

In the cases at hand, the CJEU enumerates a number of indicia that, in its view, may be considered to be constituent elements of abuse of rights.

  • First, a group of companies may be regarded as being an artificial arrangement where it is not set up for reasons that reflect economic reality, its structure is purely one of form and its principal objective or one of its principal objectives is to obtain a tax advantage running counter to the aim or purpose of the applicable tax law. That is so inter alia where, on account of a conduit entity interposed in the structure of the group between the company that pays interest c.q. dividends and the entity which is its beneficial owner, payment of the tax on the interest c.q. the dividends is avoided. Thus, it is an indication of the existence of an arrangement intended to obtain improper entitlement to the exemption provided for in the IRD/PSD that all or almost all of the interest c.q. the dividends is, very soon after its receipt, passed on by the company that has received it to entities which do not fulfil the conditions for the application of the IRD/PSD, either because those entities are not established in any member state, or because they are not incorporated in one of the forms referred to in the annex to the directives, or because they are not subject to one of the taxes without being exempt, or because they do not have the status of associated company within the meaning of the directives.
  • Second, the artificiality of an arrangement is capable of being borne out by the fact that the relevant group of companies is structured in such a way that the company which receives the interest c.q. the dividends paid by the debtor/distributing company must itself pass such interest c.q. dividends on to a third company which does not fulfill the conditions for the application of the IRD/PSD, with the consequence that it makes only an insignificant taxable profit when it acts as a conduit company in order to enable the flow of funds from the debtor/distributing company to the entity which is the beneficial owner of the sums paid.
  • Third, the fact that a company acts as a conduit company may be established where its sole activity is the receipt of interest c.q. dividends and the transmission thereof to the beneficial owner or to other conduit companies. The absence of actual economic activity must, in the light of the specific features of the economic activity in question, be inferred from an analysis of all the relevant factors relating, in particular, to the management of the company, to its balance sheet, to the structure of its costs and to expenditure actually incurred, to the staff that it employs and to the premises and equipment that it has.
  • Fourth, indications of an artificial arrangement may also be constituted by the various contracts existing between the companies involved in the financial transactions at issue, giving rise to intragroup flows of funds which may have the aim of transferring profits from a profit-making commercial company to shareholding entities in order to avoid the tax burden or reduce it as much as possible.
  • Fifth, the way in which the transactions are financed, the valuation of the intermediary companies’ equity and the conduit companies’ inability to have economic use of the interest c.q. the dividends received may also be used as indications of such an arrangement. In this connection, such indications are capable of being constituted not only by a contractual or legal obligation of the company receiving interest/dividends to pass it/them on to a third party but also by the fact that, “in substance”, that company, without being bound by such a contractual or legal obligation, does not have the right to use and enjoy those sums.
  • Sixth, the foregoing indications may be reinforced by the simultaneity or closeness in time of, on the one hand, the entry into force of major new tax legislation, which some of the groups of companies strive to circumvent and, on the other hand, the setting up of complex financial transactions and the grant of intragroup loans.
  • Finally, when examining the structure of the group it is immaterial that some of the beneficial owners of the interest c.q. the dividends paid by the conduit company are resident for tax purposes in a third state which has concluded a double taxation convention with the source member state. The existence of such a convention cannot in itself rule out an abuse of rights. Thus, a convention of that kind cannot call into question that there is an abuse of rights where its existence is duly established on the basis of a set of facts showing that economic operators have carried out purely formal or artificial transactions devoid of any economic and commercial justification, with the essential aim of benefiting improperly from the exemption from withholding tax, provided for in the IRD/PSD.

In setting out these indicia, the CJEU seems to depart from the interpretation of “abuse” in its earlier case law, where reference was made to “wholly artificial arrangements” to define abusive situations. In the Danish beneficial owner cases, it seems to be sufficient for an arrangement to be considered as being abusive if the principal objective or one of the principal objectives is to obtain a tax benefit under the relevant directives. This is very similar to the principal purpose test that has recently been introduced in the OECD Model.

Conflation of Beneficial Owner and Abuse of Law Concepts

Absence of beneficial ownership does not equate to abuse in the EU law sense, i.e. artificial transactions devoid of any economic and commercial justification, with the essential aim of improperly benefiting from a tax advantage. As Advocate General Kokott pointed out in her opinion (para. 60 of C-115/16), the concerns addressed by the abuse of law concept and the beneficial ownership concept are fundamentally different. The CJEU seems to be aware of this where it recognizes the difference between both concepts at certain stages of its analysis, making it clear that denial of a benefit based on a lack of beneficial ownership does not require tax authorities to prove abuse of law.

To a certain extent, however, the concepts of “beneficial owner” and “abuse of law” seem (nonetheless) to be intertwined (or even conflated) in the CJEU’s analysis. More likely than not the (primary) reason for this is that the PSD —in contrast to the IRD—does not contain an express “beneficial owner” requirement, although the CJEU seems to read an implied beneficial owner requirement in the PSD especially where it states, in largely the same language as in the IRD cases, that “where the beneficial owner of dividends paid is resident for tax purposes in a third State, refusal of the exemption provided for in Article 5 of [the PSD] is not in any way subject to fraud or an abuse of rights being found.”

While in the interest cases the Court appears to keep the two concepts more clearly separate, in the dividend cases the line of distinction arguably becomes more blurred, in that the interpretation of the PSD’s scope by reference to the general principle of abuse of EU law essentially appears to be induced with a view to denying the directive’s benefits to companies that are not beneficial owners. In other words, the abuse of law “detour” seems to be (deliberately) taken to (nonetheless) render the beneficial owner criterion, either because it is lacking in the relevant directive (PSD) or because it has not been implemented in the domestic legislation of a member state (IRD), pertinent as a substantive matter within the context of (or even arguably beyond) these directives.

Admittedly, where the concept of beneficial owner is construed to have an economic (“substance over form”) meaning, the concept no longer refers to a mere “attribution of income rule” and assimilation with the concept of anti-abuse is not far away. This becomes ostensible, for example, where the CJEU in its analysis regarding the constituent elements of abuse of rights refers to the situation of a recipient company that does not “in substance” have the right to use and enjoy the sum it received even “without being bound by (…) a contractual or legal obligation [to pass it on to a third party].” This clearly goes beyond the OECD Commentary’s guidance on beneficial ownership, which confines the denial of treaty benefits to situations where such contractual or legal obligation exists (OECD Commentary para. 12.4 on Article 10).

This may be best understood as clarifying the relationship between beneficial ownership and abuse of law in that an entity may be the beneficial owner (as interpreted in conformity with the OECD materials), yet still be denied the IRD/PSD’s benefits due to the artificiality of the underlying legal structure.

Going Forward

The CJEU’s rulings in the Danish beneficial owner cases are likely to have a significant impact on most international group structures and the flow of funds from EU subsidiaries to parent companies where the ultimate parent is resident in a third country. Multinational groups operating within the EU are (henceforth) to secure substance at the level of the income recipient whereby such recipient is not factually, legally or contractually obliged to pass on the income to another person.

In addition, by intertwining the beneficial owner concept with that of “abuse of EU law” or “abuse of rights,” taxpayers can no longer avail themselves of the benefits of the IRD/PSD directives if the underlying arrangement may be considered to be artificial. While the CJEU has attempted to set out a number of indicia that are considered to be constituent elements of abuse of rights, this guidance remains abstract and raises the question as to how the member states’ national courts will effectively translate this guidance in concreto.

Pascal Faes is Of Counsel with Antaxius Advocaten

This column does not necessarily reflect the opinion of The Bureau of National Affairs, Inc. or its owners.

Read the full archive of the Bloomberg Tax International Forum articles here.