Aggressive Tax Planning in the Post-BEPS Era: Systematic Approach with Analysis and Examples¹

ATP is often equated with tax avoidance.¹¹ As already stated above, ATP does not have a legal definition, while tax avoidance or preventing it is regulated in the legal order – in general anti-avoidance rules (GAARs). Examples include section 42 of Abgabenordnung¹² in Germany; section 2 of lag mot skatteflykt¹³ in Sweden; and section 28 of the Act on Assessment Procedure (laki verotusmenettelystä)¹⁴ in Finland. While some jurisdictions have not necessarily enacted tax avoidance defining GAARs, in these cases, it is not uncommon for a similar anti-abuse doctrine to have been established in the case law.¹⁵ In addition, many tax systems also include targeted anti-avoidance rules (TAARs), which grant discretion for tax-avoidance assessment in specific limited situations.¹⁶

If one of the concepts (tax avoidance) is included in the legal order, but the other (ATP) is excluded, then how can these concepts be regarded as synonymous or parallel, considering the fact that they refer to a similar kind of activity or behavior?

First of all, it should be noted that in various legal contexts, the general and legal-technical definitions of tax avoidance can be distinguished.¹⁷ As a general definition (or rather, a description), tax avoidance refers to tax-minimizing practices that the legislator did not originally anticipate or intend to be accepted. These are actions that could also possibly be evaluated ex post (later), in light of (technical) tax-avoidance norms.

Furthermore, one can use the term “tax avoidance” in a narrower and more precise manner. In the legal-technical context, specifically and retrospectively, an action is considered to be tax avoidance if there is law – or a legal claim asserted by tax authorities – claiming that an action should be considered as such. This might be a statute-based GAAR, a TAAR, or some other similar tax-avoidance norm (e.g., an anti-abuse doctrine developed in the case law).

In the latter legal and technical sense, tax avoidance is an activity that has authoritatively defined ex post, based on the GAAR or the TAAR. Actions that are dealt with on the basis of the GAAR or the TAAR are thus technically defined as tax avoidance, because they meet the characteristics described in the law.¹⁸ Meanwhile, actions that are not addressed by these measures fall within the realm of (legally speaking) permitted tax planning.

In international (cross-border) contexts especially, there have been situations involving various activities that would clearly look like tax avoidance from a national point of view in many respects, but where the tax benefit sought is based on interface differences between tax systems that are beyond the reach of national GAARs, or the applications of other national or taxtreaty-based tax-avoidance norms. In such cross-border cases, it is not uncommon to refer to “tax avoidance” in the general sense of the term when there has not been a legal or technical question concerning the issue. As such, it has not been possible to interfere with the tax planning in these cases, even if the actions of a taxpayer have resulted in double non-taxation or another kind of “undertaxation”. The actual result of such forms of tax planning have thus been similar or identical to activity tackled by the GAARs or corresponding norms on the national level. However, the underinclusiveness of these measures creates a fundamental legal-technical difference between national and cross-border tax planning. The term “aggressive tax planning” has often been used to refer specifically to such cross-border activities, which appear to be nearly identical to tax avoidance in the term’s broader conceptual sense.¹⁹

The term “aggressive tax planning” can thus be used loosely, in a similar manner to how “tax avoidance” is used in the general sense. It is also worth noting how these concepts relate to the timeline of actions – tax planning activities, the actions of tax authorities, and court decisions. As stated earlier, the term “aggressive tax planning” can also be used in an ex-ante review to describe actions whose acceptability under tax law is ex post assessed from the perspective of the applicability of the GAAR or other tax avoidance norms, which bring it into either the area of either legitimate tax planning, or of illegitimate tax avoidance.

When a taxpayer at time t carries out business or financial measures that also have, at least partially, fiscal goals, that taxpayer cannot always be sure whether the actions will be treated fiscally in the way that was originally intended. Depending on the point of view, such actions could be described as innovative, creative or aggressive tax planning.

From the standpoint of civil law, these actions are usually considered to be valid transactions. However, their tax-legal significance or acceptability may be examined later. The tax authorities may take into account the fact that a fiscal outcome was reached through aggressive tax behavior that may ultimately be regarded as tax avoidance. There are differences between states in how much power tax authorities have. In some countries – such as Finland – they have direct authority under the GAAR to recharacterize actions as tax avoidance and to make decisions accordingly.²⁰ In some other countries, such as Sweden, there is no such direct right; the tax authorities can only challenge the arrangement and take the case to court.²¹ In either of these situations, though, a court will ultimately decide the fiscal acceptability of the disputed actions. Most importantly, in all these evaluations, both tax authorities and courts must rely on the law in their actions and decisions.²²

The logical use of the terms is linked to whether the level of fiscal acceptability of actions is examined ex ante or ex post. In the ex-ante review, aggressive tax planning and tax avoidance can generally be equated to each other; in advance, there is often no certainty as to whether a tax position will be ultimately accepted. As such, the ex-ante review focuses on the likelihood of a certain action’s acceptance from a tax perspective.²³ As further explained in Chapter 2.3, this review may also include various social, economic and policy considerations that cannot be regarded relevant from an ex-post perspective.

Conversely, in an ex-post review, the court deliberates on whether a GAAR or a TAAR is applicable. In this sense, the question concerns judicial assessments and the validity of the action from a tax perspective – that is, tax avoidance in the legal-technical sense. The transaction is deemed either unapproved tax avoidance or as approved tax planning. From a strictly legal point of view, there is no such thing as aggressive tax planning.²⁴ However, if we deviate from the strict legal evaluation, the social evaluation of the matter is not tied to purely legal criteria; regardless of the court’s decision, the company’s stakeholders may still decide the activity is aggressive tax planning.

As mentioned earlier, there are differences between states regarding the powers of tax authorities. The tax authorities and the courts must always follow the law in their actions and decisions, but broader social evaluation is not tied to legal norms or formal legal processes (which can have harmful consequences for a company’s legal security). This kind of social assessment can be done by any or all of the company’s stakeholders – for example, investors within the ESG framework of responsible investment, the media, or non-governmental organizations – each from their own perspectives.²⁵ Even if an activity is considered to be acceptable tax planning by the courts, stakeholders may still view it as aggressive tax planning.²⁶ As a matter of fact, anyone can conduct such an assessment, since it is not tied to any legal powers or formal criteria.

Therefore, it is important to distinguish three types of assessments. The first is a preliminary assessment of a tax-planning measure or its fiscal outcome, which may take into account economic logic, the internal systematics of the legal tax system, or the equality or horizontal fairness of taxation.²⁷ The second type of assessment concerns legal aspects – that is, assessments and decisions made by courts or other law-enforcement authority. Such assessments are bound by legal norms, the doctrine of legal sources, and other conditions of legal decision making.²⁸ The third type of assessment is when the company’s stakeholders determine taxplanning structures or actions to be aggressive tax planning, regardless of the formal legal approval of the matter. In this case, the assessment can be based on non-legal criteria – for example, taking into account economic or social perspectives.²⁹

Although the court cannot render a judgement concerning ATP, the legislator can either implicitly or explicitly consider whether certain formally accepted actions represent ATP, and address them with, for example, a new special anti-avoidance rule (“SAAR”).³⁰ Though ATP may be hardly mentioned in the statute itself, the concept can be found either directly or indirectly in the travaux préparatoires of the provision. In this way, the legislator can link the concept of ATP to a specific set of characteristics by holding up the case as an example, without defining the concept itself.³¹ Here, the assessment is legal in the sense that the legislator must weigh many issues when formulating the legislation, such as the impact of EU law, the applicability of GAARs, and the clarity of tax legislation. Accordingly, such legislative and tax policy review approximates a legal assessment of ATP.³²

It is worth emphasizing, however, that in these instances the legislator still does not formulate a general definition of ATP. On the contrary, the legislator only describes a certain action or measure and connects it with ATP by removing its tax advantages. In many cases this means enacting a new SAAR or amending an existing SAAR.³³ These norms are more specific than GAARs and TAARs, and they do not authorize the tax administrations or courts to determine certain actions as “tax avoidance” or “aggressive tax planning.” Instead, negative tax consequences of a SAAR are applied based on specific, objective criteria set in its wording.³⁴ As such, SAARs can be viewed as legislator’s way of establishing certain actions as ATP.

The BEPS project can generally be characterized as an extensive “repair package” of the international tax system, specifically aimed at eliminating or reducing the possibilities of ATP in international situations. In the initial report on the project in 2013, the problems of international taxation were described, in part, as follows:

BEPS issues may arise directly from the existence of loopholes, as well as gaps, frictions or mismatches in the interaction of countries’ domestic tax laws. These types of issues generally have not been dealt with by OECD standards or bilateral treaty provisions. There is a need to complement existing standards that are designed to prevent double taxation with instruments that prevent double non-taxation in areas previously not covered by international standards and that address cases of no or low taxation associated with practices that artificially segregate taxable income from the activities that generate it. Moreover, governments must continue to work together to tackle harmful tax practices and aggressive tax planning.³⁵

Indeed, many researchers have referred to ATP specifically as international tax planning that utilizes and ultimately relies on the gaps between different national tax systems, in which the aim is to make use of the differences between these tax systems and the lack of coordination between them.³⁶ Some authors even consider the whole concept as referring solely to international situations.³⁷ In the general tax debate, the focus has especially been on ATP in large MNEs.

Within the national tax system, many undesirable phenomena can be tackled either by introducing new tax-avoidance provisions (especially SAARs or TAARs), or by amending existing tax legislation as necessary. Meanwhile, in cross-border situations, there have been some limitations placed on the application of taxavoidance provisions.³⁸ Amending relevant tax legislation or tax treaties is usually more complicated than introducing new rules. Overall, it can be argued that international situations have not been as well controlled by the national legislatures.

As stated earlier in this article, we do not aim to exhaustively define the concept of ATP. Instead, we aim to identify and locate the occurrence of the phenomenon in relation to the national tax system, and especially in relation to the BEPS project. Accordingly, we take the position that ATP can occur in the tax behavior of MNEs, even after the BEPS project. Consequently, we utilize a systematic approach in assessing in what kind of forms ATP can still occur or for what reasons this can occur. Our approach is general in nature, and suitable for examining the tax systems of different states, but the concrete examples discussed later in Chapter 4 will be more focused on the tax system in Finland and how BEPS measures have been implemented there.

Assuming that we accept the foregoing proposition that the concept of ATP is tied to an ex-ante perspective and can only be defined through legislative and tax policy review of certain taxpayer actions, we can also deduce that ATP can still occur after the BEPS project. This is for a number of reasons. First of all, there are scenarios and activities that have been left out of the BEPS project, which focuses on limiting the tax planning of MNEs; certain specific types of tax subjects fall outside that focus (class 1a in the figure presented below), although some of the BEPS regulatory models may, as a result of their implementation, also apply in themselves to private individuals. However, certain situations or actions might also be included in the main themes of the BEPS project while, for one reason or another, falling outside the scope of the project (class 1b in the figure).

Secondly, taxpayer actions may be of such a type that the BEPS project clearly intends to address the tax benefits obtained through them, but the proposed regulation (or regulatory model) has, for one reason or another, remained “underinclusive” and insufficient in relation to the explicitly stated or assumed goals (class 2 in the figure).

Thirdly, some taxpayer options that have yielded benefits have been the target of interference as part of the BEPS project, but, for one reason or the other, BEPS-inspired regulation has been possible to successfully circumvent (class 3 in the figure).

Fourthly, a given state BEPS regulation model may have the aim of regulating certain taxpayer actions, and be capable of effectively tackling any underlying tax benefits, but suffer from inadequate implementation (class 4 in the figure).

The above situations are represented in the following figure:

This classification is used as a reference framework for this paper, and it can be relied upon more generally when trying to describe the aggressive “post-BEPS” tax planning opportunities of MNEs. Next, we will discuss each numbered “class” or “region” of the proposed approach in more detail, both generally and using concrete examples. As some concrete examples show, a certain practical phenomenon (or means for ATP) cannot necessarily be placed in a single category, because many phenomena may overlap with two or even more categories.

Classes 1a and 1b in the figure above describe ATP situations or actions that are not relevant to the BEPS project from a thematic perspective or, alternatively, have intentionally been left outside its scope.

From an MNE’s perspective, the BEPS project has focused on tax-planning activities that aim to reduce MNEs’ overall income-tax exposure. Accordingly, the overall income-tax exposure pertaining to, for example, the ownership structure of MNEs has not been the focus of the project. In principle, it might be argued that MNEs also utilize tax planning practices that were not identified during the preparatory phase of the BEPS project. However, given the extent of the BEPS project, it can be argued that the likelihood of such unidentified practices’ existence is insignificant.

Further, it may be argued that class 1a of the proposed framework includes, for example, aggressive tax planning concerning private wealth management carried out, inter alia, through unit-linked insurance products, such as unit-linked life insurance contracts and capitalization bonds.

ATP may also be considered to encompass actions or phenomena that have been intentionally or unintentionally left out of the scope of the recommendations included in the BEPS project. One possible rationale might be the fact that even if certain ATP actions had been properly identified, they were either not prioritized or considered too complicated to address within any of the proposed recommendations.

Indeed, numerous phenomena have been intentionally left outside the scope of the BEPS project. The original intention of the BEPS project was to tackle various inconsistencies between different national tax systems to prevent the corresponding tax planning opportunities utilized by MNEs; as a result, various questions regarding subjects other than corporate income tax were excluded from the project.³⁹ Accordingly, questions regarding the international aspects of transfer taxation, property and wealth taxation, and gift and inheritance taxation are outside the scope of BEPS, even if these areas may bear similarities to items within its scope. It is self-evident that corporate income taxation is not the sole area where the co-existence of numerous national tax systems creates inconsistencies.⁴⁰

When the BEPS project’s recommendations were published in 2015, concrete proposals regarding the various problems associated with the taxation of the digital economy were left unaddressed.⁴¹ The final report explicitly stated that “because the digital economy is increasingly becoming the economy itself, it would be difficult, if not impossible, to ring-fence the digital economy from the rest of the economy for tax purposes.”⁴² However, the OECD has recently developed concrete solutions to these problems. Hence, various tax complications surrounding the digital economy are now being tackled, albeit with a rather complicated and extensive legislative framework (including OECD Pillar II and global minimum taxation) that affects numerous areas outside the digital economy.

Outside of OECD Pillar II concerning global minimum taxation, however, the BEPS project has not changed the basic underlying principles of the international tax system. As such, the international tax system is fundamentally based on profit-allocation rules arising out of the benefit principle⁴³ and the separate-entity approach,⁴⁴ both of which still drive international income taxation.⁴⁵ Because of this, MNEs are still able to utilize differences and inconsistencies in national tax systems through, inter alia, tax planning related to a group entity’s choice of tax residency. For example, the check-the-box rules provided by the U.S. tax legislation are still possible in the taxation of MNEs.⁴⁶

As such, the BEPS project has also been criticized for neglecting the creation of truly innovative and new means of reform for the international tax framework. The proposed reasons for this lack of innovation vary to some extent. Certain authors believe that it is caused by the unambitious approach of the OECD. Others argue that the opportunistic behavior of taxpayers is the root cause of the ineffectiveness of international tax rules, rather than the changes in the international business environment that especially occurred during the later stages of the project.⁴⁷

One of the clear limits of the BEPS project is that it did not address – and could not address by any plausible means – the differences in income tax rates used by different countries. As such, MNEs can still effectively gain benefits from the variance in tax rates, especially considering that they are relatively free to choose the tax residence for different group entities for the reasons mentioned above. However, the possibilities of utilizing tax planning opportunities in connection with tax-rate-focused competition have been effectively reduced with the introduction of the rules concerning global minimum taxation.⁴⁸ One of the clear signs of the significance and extensiveness of these rules is that numerous low-tax countries and jurisdictions have been forced into introducing legislation regarding global minimum taxation.⁴⁹

It is somewhat surprising that the BEPS project did not focus further on the utilization of crossborder corporate losses, even though this is a particular feature of international corporate taxation and tax planning. In 2011, the OECD had already prepared an extensive report discussing various aggressive tax planning techniques related to the utilization of crossborder corporate losses. The report stated that the differences regarding the restrictions of cross-border loss utilization, and transfer of such losses, amounted to a considerable tax risk for member countries. Because of this, it recommended implementing various legal measures restricting the transfer of corporate losses and double deductions.⁵⁰

Such rules were not proposed in the recommendations of the BEPS project, except for those regarding the prevention of double deductions of corporate losses using hybrid entities.⁵¹ This can be attributed to the fact that most countries had already implemented effective legal remedies regarding the possible misuse of corporate losses prior to the BEPS project – at the latest, after implementing the anti-avoidance rules proposed in the project. As such, the international consensus seems to be that although the BEPS project did not focus on the regulation of cross-border corporate tax loss utilizations and transfers thereof, these behaviors do not pose any meaningful tax risks from the perspective of profit shifting and base erosion under the current circumstances.⁵²

Class 2 in the figure above describes ATP situations or activities that have been the focus of the BEPS project and also are within the scope of its regulatory proposals, but for some reason the regulatory proposal has been insufficient in relation to its ultimate objectives and leaves room for opportunities for aggressive tax planning.

One of the classic challenges of legal regulation concerns the scope of a legal norm.⁵³ This is also a common challenge in the context of preventing various tax avoidance practices. An anti-avoidance norm should not be underinclusive, in order to ensure its effectiveness in relation to the avoidance practices it is intended to counter. However, anti-avoidance regulation should not be overinclusive, either, as this may lead to general ambiguity in the tax law and potentially impede legitimate transactions, thus hampering bona fide business activities.⁵⁴ Within the context of this framework, it is of special interest whether the BEPS actions are underinclusive in relation to the project’s original goals. Even if a certain situation or action has been covered by the BEPS actions, the OECD may not have been able or inclined to extend these measures to such lengths that any kind of aggressive or other tax planning activity would have been excluded.

In practice, the weaknesses in the BEPS project materialize in two distinct and inherently interconnected ways.

Firstly, none of the four minimum standards included in the 15 action items of the BEPS project concern actual material tax legislation. Instead, all action items concerning the material tax legislation are, by nature, either recommendations or descriptions of best practices.⁵⁵ This is reflected especially in the implementation of Actions 2–4; concrete measures regarding those items’ implementation have taken off rather slowly outside the European Economic Area (EEA). For instance, according to the progress reports published by the OECD, the majority of the member states in the Inclusive Framework of the BEPS project have yet to implement controlled foreign corporation (CFC) rules based on Action 3 of the project.⁵⁶

Secondly – and partly as a result of the foregoing – the implementation of the BEPS project’s recommendations and practices concerning substantive tax legislation has not been uniform across different countries. However, since issues related to implementation do not directly fall within the project’s actual scope – which is the theme of this subsection – challenges pertaining to the project’s implementation will be addressed more specifically in Chapter 4.4.

On a more concrete level, addressing the use of so-called “shell companies” can be seen as one of the gaps in the BEPS project. It is uncertain whether the CFC rules recommended in Action 3 can effectively target the use of such entities. This has become particularly evident in the EU, where CFC legislation based on the Anti-Tax Avoidance Directive (ATAD)⁵⁷ has had a supplement proposed called ATAD 3 (also known as the “Shell Company Directive”)⁵⁸ that would impose various reporting obligations and tax disadvantages tied to the substance of companies.⁵⁹ While the CFC regulation in the ATAD largely aligns with Action 3 of the BEPS project, ATAD 3 goes beyond the scope of the BEPS project. In particular, amendments proposed by the European Parliament to the proposed directive reveal that the EU legislator does not consider the current CFC legislation a sufficiently effective means to prevent tax planning through conduit entities and base companies.⁶⁰ The CFC legislation can thus be considered an example of a regulatory phenomenon that has not been deemed sufficiently effective – at least in the view of the European Parliament and the Commission.⁶¹ However, time will tell the extent to which the legislator is willing to go in preventing aggressive tax planning, even if it simultaneously significantly complicates the position of companies established for conducting operative business functions on a legitimate business rationale.⁶²

Furthermore, the interest deduction limitations proposed in Action 4 do not address the fundamental issue of the tax treatment between equity and debt-financed investments, but rather aim solely to restrict the deductibility of interest. Thus, the original purpose of the BEPS project was not to entirely eliminate deduction of interest expenses, but rather to create an environment for businesses where the benefits of profitshifting through debt and related interest expenses would merely be reduced.⁶³ Since the deduction of interest expenses remains possible within the framework of interest deduction limitations, profit shifting can still be conducted to a limited extent. Additionally, mechanisms have been established to allow deductions of interest expenses that were not possible in previous tax years due to interest deduction limitations. Such mechanisms can be leveraged through various means, analogous to certain tax depreciation methods.⁶⁴

Even if a specific regulation was implemented within the framework of the BEPS project (preparation of model rules and their implementation), it is possible for it to be avoided or circumvented through certain means. In the figure above, these situations are described by class 3.

Possible reasons for these scenarios are (i)

a certain phenomenon has not been effectively regulated, or

For instance, neither the BEPS project nor its actions included any explicit recommendation for a GAAR. Nevertheless, a GAAR aimed at preventing corporate tax avoidance is included in Art. 6 of the ATAD.

However, in the context of the BEPS project, issues or instances of abusive tax behavior have been highlighted, especially in international situations. BEPS Action 6 pertains to the denial of tax treaty benefits in cases of abuse⁶⁵ and provides minimum standards for preventing tax treaty abuse. Compliance with the minimum standards set forth in Action 6 requires either (1) the addition of explicit language in the tax treaties that the treaty is not intended to create situations of untaxed income or (2) a principal purpose test (PPT) or limitation on benefits (LOB) provision.

Action 6 has been implemented with the multilateral instrument (MLI). The first signing ceremony of the MLI occurred on June 7, 2017, and 68 states were present, including all the Nordic countries. The entry into force of the MLI in different countries has depended on the ratification schedules of individual states.⁶⁶ Each signatory state of the MLI is required to declare which tax treaties it wants to bring within the scope of the instrument. For the instrument to have an impact on a specific tax treaty, both contracting parties of a bilateral tax treaty must also sign and ratify the instrument. Through the MLI, a kind of tax treaty GAAR, based on the principal purpose of a transaction or arrangement (i.e., PPT provision), has been added to most of the tax treaties concluded by Finland, for example.

As already noted, within the scope of the BEPS project, no explicit recommendation for a GAAR against general tax avoidance was included, apart from addressing the aforementioned instances of tax treaty abuse. However, such a provision is included in Art. 6 ATAD pertaining to prevention of corporate tax avoidance.⁶⁷ From the perspective of legal decisionmaking, this involves the previously-discussed question of the relationship between a specific rule and a general rule. If there is a specific rule on a certain matter, the assessment and testing of a particular phenomenon should be conducted primarily in relation to the specific norm and not the general rules. This is due to the lex specialis derogat legi generali principle (i.e., a specific norm prevails in case of conflict with a general norm).

Therefore, it is crucial that a specific rule be properly calibrated if the non-applicability of a specific rule prevents the possible secondary application of a general rule concerning tax avoidance. According to the lex specialis principle, the application of specific rules takes precedence over general provisions; thus, norms specifically enacted in anti-tax avoidance rules should be primarily considered. In this way, the legislator has indicated – in the relevant context – where the boundaries between acceptable and unacceptable tax planning lie.⁶⁸ The immediate downside of this is that the precise definitions or the use of specific threshold values in special anti-avoidance rules simultaneously create new distinctions that can be exploited in aggressive tax planning.

There is currently little jurisprudence or concrete evidence on how Art. 6 ATAD affects the scope of member states’ norms regarding tax avoidance and, consequently, the framework around the prevention of aggressive tax planning. Based on the foregoing, the possibilities of circumventing special anti-avoidance regulations largely depend on how the specific regulatory recommendations within the BEPS project have been implemented, and the tolerance of an individual state or jurisdiction in applying its domestic GAAR. For example, in Finland, regulatory solutions derived from the BEPS project and ATAD have been implemented rather comprehensively, with special anti-avoidance regulations that exceed the minimum standards. Finland has also applied its domestic GAAR quite effectively since the 2010s. Therefore, circumventing the rules within the scope of the BEPS project may be difficult from a Finnish perspective.

As stated earlier, the BEPS project has not established consistent standards or practices regarding the application of GAARs or similar legal principles (outside of abuses of tax treaty benefits). Thus, the BEPS project does not provide answers to how situations falling outside the scope of the action proposals’ special anti-avoidance rules should be assessed – for example, in light of GAARs. In practice, the interplay between GAARs and anti-abuse doctrines on one hand, and precise SAARs on the other, may still lead to significant inconsistencies between different countries and jurisdictions. For example, in Finland, “correcting” the scope of a SAAR through the application of a GAAR has been accepted, and even the simultaneous application of a GAAR and a SAAR has been positively regarded in Finnish case law.⁶⁹ There are no specific reservations against the interaction between general clauses and SAARs found in the reasoning of the court in the relevant decisions. By contrast, in Germany, the existence of SAARs has effectively prevented the application of a GAAR in several instances, and the possibility of expanding the scope of special anti-avoidance regulation by applying a GAAR has been approached with more reservation.⁷⁰

Regarding the BEPS project, a potential issue with the interaction between SAARs and general clauses can be exemplified by Action 4, which addresses interest deduction limitations. Action 4 proposes a uniform regulation on interest deduction limitations, without taking a position on the relationship between interest deduction limitations and general clauses. The problematic aspect of this proposal lies in the deductibility of interest expenses per se; when the deductibility of interest is restricted, can the deduction of interest expenses be allowed in all situations within the prescribed thresholds, or can the deductibility of interest expenses be denied entirely by, for example, applying a general clause? The outcome may vary significantly in different countries, leading to distortions between different tax systems.

For example, in Finland, it might be inferred from the provisions on interest deduction limitations that the benefits of profit-shifting would be permissible outside the scope of the limitations, as the interest limitation rules “aim at reducing the possibilities to carry out profit shifting through debt financing”.⁷¹ However, the travaux préparatoires explicitly state that in Finnish tax practice, the deductibility of interest expenses related to intra-group transactions has been occasionally denied by applying the GAAR.⁷² This position has recently been since confirmed in Finnish case law, where a strict line has been drawn between the deductibility of interest expenses generated in intra-group and in third-party transactions.⁷³

According to the authors’ understanding, intragroup interest expenses are deductible certain states – for example in Sweden and Norway. This assumption is based on the CJEU judgment in Lexel,⁷⁴ the ETFA Court’s judgment in PRA Group Europe,⁷⁵ and the resulting decisions by the Swedish SAC⁷⁶ and the Norwegian Court of Appeal (Borgarting lagmannsrett),⁷⁷ where the deductibility of intra-group interest expenses in a cross-border situation was eventually confirmed. It is worth mentioning that none of the cases addressed the question of applying the GAAR or the prevention-of-abuse principle, as the cases solely concerned rules targeting deductibility of interest expenses in crossborder situations. These targeted rules were considered contrary to the freedom of establishment in both CJEU cases and, as such, could not be applied.

However, recently AG Emiliou concluded in case XBV⁷⁸ that intra-group loans may be considered artificial if they are principally motivated by tax reasons. The AG proposed for the Court to (partly) overrule the judgment in this sense in Lexel, where the Court had held that intra-group loans cannot be deemed artificial if the terms comply with the arm’s-length principle.⁷⁹ CJEU ended up following AG Emiliou’s proposition, albeit while being a bit more careful in its approach. Essentially, the Court agreed that intragroup loans complying with the arm’s-length principle may also be considered artificial.⁸⁰ Even so, the Court emphasized that the tax authorities need to provide prima facie evidence of abuse⁸¹ and the denial of interest deductions must be based on sufficiently clear criteria.⁸²

Class 4 in the figure above describes ATP situations or activities that have been the subject of the BEPS project and its legislative proposals. Despite this, there are still opportunities for aggressive tax planning; in one or more countries, the implementation of the proposed regulation is completely lacking, incomplete, ineffective or inconsistent.

As mentioned above, the BEPS project does not include minimum standards for substantive tax legislation, but rather recommendations and best practices to develop national tax legislation. Hence, the legislative content of the BEPS project and the exact scope of the proposed regulation are largely subject to the discretion of domestic legislatures. However, this only applies to those states that have adopted the legislative recommendations contained in the BEPS project. Even for these states, it must be noted that the actual implementation of the legislation may ultimately be quite limited.⁸³

The implementation of the BEPS project has not been consistent, even within the EU, though the BEPS project’s action proposals have been extensively adopted by member states through the implementation of the ATAD.⁸⁴ While the directive establishes a minimum standard⁸⁵ for the SAARs included therein, it also provides member states with numerous regulatory alternatives. Member states also have the option of enacting special anti-avoidance regulations that go beyond the minimum standards, as long as the SAARs are compatible with EU primary law. In this way, the ATAD ensures a broadly similar type of regulation across all member states, but does not guarantee that the regulation is precisely identical in content in each member state.⁸⁶ Within the EU, there are even member states that have implemented SAARs solely to fulfil the formal requirements of the ATAD. In these states, it has been debated whether the application of these rules will be effective in practice.⁸⁷

The fragmentation of specific anti-avoidance regulation is particularly evident in the case of the CFC legislation included in the ATAD. The ATAD provides member states with two alternative methods for determining CFC income,⁸⁸ resulting in slightly different approaches to determining CFC income in different member states. There are also significant differences in the exact scope of the rules. Some member states have implemented significantly stricter CFC rules than the minimum standard set by the directive, while others have sought to make the CFC rules as permissive as the minimum standard allows. These differences become apparent when comparing certain details of the CFC rules in, for example, Luxembourg and Finland:

Firstly, according to the Finnish CFC legislation, the formation of a CFC requires a minimum ownership stake of 25% in the foreign entity’s voting rights, capital, or rights to the entity’s profits or proceeds of the entity’s wealth. Instead, Luxembourg’s CFC legislation requires a 50% ownership stake, which, like Finnish legislation, is linked to voting rights, capital, and the foreign entity’s profits. However, first, Luxembourg’s CFC legislation does not include rights linked to the proceeds of the entity’s wealth. Secondly, the condition related to the level of taxation of the foreign entity is lower in Luxembourg, partly due to a lower corporate tax rate.⁸⁹ Thirdly, Luxembourg’s CFC legislation has more lenient substance requirements compared to Finland and applies to non-EEA countries.⁹⁰

Additionally, outside the EU and the EEA, there are numerous countries that have not adopted the substance-based tax legislation recommended by the BEPS project. In many jurisdictions, the focus has been solely on implementing the minimum standards of the BEPS project. In many countries, the BEPS project has even posed challenges due to existing tax advantages offered by those country’s domestic tax legislation.⁹¹ Traditionally, the anti-avoidance regulation has not been comprehensive in these countries – sometimes, it has been practically non-existent. The BEPS project has not brought significant change to this situation, and as the fundamental principles of international taxation remain unchanged, it is still possible to benefit from the tax advantages offered by these countries. For instance, one of the primary objectives for the United States in relation to the BEPS project was to ensure that the CFC regulation recommended by Action 3 would be universally adopted, but the BEPS project ultimately proved to be a significant disappointment for the United States in this regard.⁹²