Getting Ready for the EU’s Anti-Tax-Avoidance Directive

With less than six months to go before EU member states are required to have transposed new EU anti-tax-avoidance rules into their domestic legislation, the pressure is on. The anti-tax-avoidance directive, or ATAD (Council Directive (EU) 2016/1164), is the EU’s answer to part of the OECD’s base erosion and profit-shifting action plan. While it may have been well intentioned, there are concerns that it will at best have little impact and at worst lead to the kind of chaos that it is intended to prevent. This article looks at some of the things that could go wrong, the choices now facing EU member states in implementing the EU rules, and if and when taxpayers can expect to notice the changes.

What Is the ATAD?

The ATAD is the EU’s response to three of the 15 BEPS actions: interest deduction limitation, controlled foreign corporations, and anti-hybrid rules. None of these is an OECD minimum standard. The directive also introduces an EU general antiabuse rule (GAAR) and exit tax rules. Neither of these last two features is in the OECD’s 15 action points.

The first draft of the ATAD was released shortly after the BEPS final reports were published, on January 28, 2016, as part of a larger package of antiavoidance measures. After some nine interim drafts, the final version was approved in the summer of 2016. EU member states are required to have implemented its provisions generally by January 1, 2019, but there are a number of important exceptions that will be explained below. This directive, commonly referred to as ATAD 1, was followed during the course of 2017 by a second directive (Council Directive (EU) 2017/952), ATAD 2) that radically revised the anti-hybrid rules of ATAD 1.

What Happened to the Other 12 BEPS Actions?

The ATAD is effectively a subset of the proposals for an EU corporate tax system (common consolidated corporate tax base, or CCCTB). However, rather than wait for those proposals, which were relaunched in 2016,¹ to be approved (a wise choice given the bumpy ride CCCTB is having), the European Commission extracted “some international aspects of the common base which are linked to the BEPS project”² and turned them into the ATAD, including the three that feature in the OECD’s BEPS project.³ That does not mean the EU has neglected other BEPS actions. In fact, EU-level activity has been taken on almost all the BEPS actions, albeit in a somewhat piecemeal fashion and using various instruments, including:

• monitoring of intellectual property box regimes and transfer pricing by the EU Code of Conduct Group (BEPS action 5 and actions 8-10);

• European Commission recommendations on tax treaty permanent establishments and principal purpose tests (BEPS action 6 and action 7);

• a directive on disclosure of aggressive tax planning (BEPS action 12);

• a directive on country-by-country reporting (BEPS action 13);

• a directive on dispute resolution (BEPS action 14); and

• (proposed) directives on digital activities (BEPS action 1).

Although the ATAD’s exit tax and GAAR provisions are not directly associated with the BEPS reports, it has been suggested that they are part of the broader aims of the OECD’s program.⁴ At any rate, both have been on the EU’s policy radar for a number of years, so it comes as no surprise that they have been included in the ATAD.⁵

Why EU-Level Action?

The overarching objective of the ATAD is to provide for a coordinated EU approach to aggressive tax planning. This means, at least in principle, that EU member states must introduce the new rules in broadly the same way. That is not to say that no such rules exist today, but not all member states have them, and — if they do — the rules are generally not aligned with each other. While it may seem somewhat counterintuitive, governments may not always be keen on introducing antiavoidance rules. The reasons vary but can include:

• they don’t need to (for example, they already have such rules, or are not an “at risk” — high-tax — jurisdiction);

• they do not want to put off potential investors; or

• they do not have the resources to deal with complex tax administration rules.

The European Commission put it like this in its communication on the 2016 package: “In some instances member states are reluctant to act, being fearful of the competitive disadvantage this might bring.” Furthermore, it said, “Divergent national approaches to tackling this cross-border problem can create loopholes for aggressive tax planners.”⁶ Hence the need for action at the EU level.

Why National Implementation?

If the ATAD’s aim is to ensure a coordinated set of measures across the EU, a fairly obvious question that needs addressing is why those measures must be separately implemented in the 28 EU member states. The simple answer is that this is required by EU law: A directive is binding on member states regarding the result to be achieved but leaves the form and methods for doing so to each member state.⁷ Transposition into domestic law has to be done within the time limits set by the directive. The European Commission can take action against a member state for non-implementation or incorrect implementation. In practice, the degree to which implementing action is required will depend on the extent to which domestic law already complies with the aims of the directive.

The withholding tax exemption for dividends under the parent-subsidiary directive (Directive 2011/96/EU) is an obvious example, with those member states that did not impose withholding tax not having to take any action at all on that aspect. The ATAD also illustrates this, with many member states already having domestic rules that to a greater or lesser extent reflect those in the directive. Because of their results-based approach, directives tend to be short on detail, leaving a fair degree of discretion to the member states in the implementation process. The explanatory memorandum to the original draft of the ATAD points out that it reflects a balance between the search for uniformity and the need to “accommodate the special features of [member states’] tax systems within these new rules.” This partly explains why EU antiavoidance rules took the form of a directive, rather than one of the other possible EU instruments.

Another consequence of choosing the directive route is that a directive has to respect member states’ autonomy. In particular it should not include common EU rules that could better be left to each member state to determine case by case (the subsidiarity principle). This is particularly important in the politically sensitive area of taxation. Notwithstanding the discretion they are permitted, member states’ practice in implementing directives tends to vary, with some more or less copying and pasting the text into their domestic legislation and others making more of an effort to tailor the provisions. While this approach to EU legislation may make sense from a principles or pragmatic standpoint, the downside is that taxpayers can be faced with inconsistent rules across 28 member states, resulting in extra compliance burdens, uncertainty, and possibly double taxation.

Minimum Standards: How Low Can You Go?

Although none of the ATAD rules constitutes a “minimum standard” under the OECD’s BEPS project, they do represent a kind of EU minimum standard. In the BEPS context, a minimum standard is one of the four actions (treaty abuse, harmful tax practices, CbC reporting, and dispute resolution) that members of the OECD’s inclusive framework have committed to implement and that will be monitored by peer review. The BEPS actions that correspond to the ATAD provisions are no more than recommendations for best practices or common approaches.

Although members of the inclusive framework have committed to implementing the BEPS package as a whole, it seems inevitable that the pressure to implement these other measures will be weaker. The ATAD provisions, on the other hand, constitute a legally binding “minimum level of protection for national corporate tax systems against tax avoidance practices across the Union.”⁸ More specifically, article 3 provides that: “This Directive shall not preclude the application of domestic or agreement-based provisions aimed at safeguarding a higher level of protection for domestic corporate tax bases.” That means member states may apply even stricter standards or even additional antiavoidance rules, provided they observe EU law, in particular the EU fundamental freedoms. Conversely, member states may not apply more generous rules for taxpayers. In practice there will be gray areas.⁹ Even in implementing the ATAD as it stands, member states have a certain amount of discretion, in addition to numerous options that would in general soften the impact for taxpayers. Again, this may lead to inconsistent and unclear rules once they are implemented across the 28 member states.

Where’s the Abuse?

Notwithstanding the reference to tax avoidance in its title, the ATAD’s interest deduction limitation, exit, and anti-hybrid provisions do not contain explicit references to tax avoidance. In contrast, the CFC provisions include two antiavoidance conditions, one for each of the two options given to member states to formulate their CFC rules, and the general antiabuse rule is, by definition, an antiavoidance provision. It is settled EU law that cross-border barriers put up by member states to prevent tax avoidance are only acceptable if their effects are limited to abusive situations.¹⁰ That principle is reflected in the ATAD’s CFC provisions in two different ways. The approach that targets passive CFC income is conditional on there being no “substantive economic activity” carried on. The “transfer pricing” approach that targets income that is properly attributable to the parent’s significant people functions is conditional on the “essential purpose” being to obtain a tax advantage. While the question of how these tests should be applied is anything but clear, they do at least take EU case law into account.

The absence of a reference to avoidance in the other three ATAD provisions may not seem strange if one simply assumes that the situations they address are, by definition, abusive. Given the legislative and political context of the ATAD, such an assumption seems more than reasonable. From an EU law perspective, the absence of some kind of tax avoidance condition or rebuttal option for the taxpayer may seem less obvious.¹¹ For example, the original draft of the ATAD explained that the exit tax targeted “Taxpayers [that] may try to reduce their tax bill by moving their tax residence and/or assets to a low-tax jurisdiction.” This may still be the underlying aim, but the ATAD’s recitals now adopt a more neutral formulation — that is, to tax “the economic value of any capital gain created in [a member state’s] territory even though that gain has not yet been realised at the time of the exit.” The change most likely reflects the fact that it has long been established by the Court of Justice of the European Union that taking advantage of another member state’s more favorable (tax) rules is not in and of itself abusive under EU law.¹² Its prevention is therefore not something that could be openly put forward as a primary target of the ATAD. The absence of an avoidance condition or taxpayer rebuttal option is most likely due to the fact that existing exit taxes have generally been justified on grounds other than tax avoidance.¹³ In the case of the ATAD’s interest deduction limitation provisions, an example is given of a loan from a no-tax third country to an EU group member to illustrate the aim of the provision.¹⁴ In this case, the absence of an avoidance condition or taxpayer rebuttal option may be explained by the fact that the rules do not, as such, distinguish between domestic and cross-border situations, so no EU justification, whether on avoidance or other grounds, is called for. This cannot be the case for the anti-hybrid rules, which do clearly distinguish between domestic and cross-border situations. It is not surprising that questions have been raised regarding their compatibility with EU law.¹⁵

How Fast Is the Clock Ticking?

All the ATAD provisions have hard deadlines by which EU member states should have transposed them into their domestic legislation. Failing this, the European Commission can take legal action against the member state in question. By contrast, the only (somewhat) concrete timelines under the BEPS project relate to the four minimum standards, and these are only “enforced” through peer pressure, that is, monitoring under the auspices of the inclusive framework. Given the approaching EU deadlines, it is a little surprising that there has not been more legislative activity in EU member states. A number of member states have announced public consultations or even draft legislation, but progress is generally slow or in some cases appears entirely absent.

There are a number of possible reasons for the inactivity. Not all the ATAD provisions must be implemented by January 2019: The exit tax and anti-hybrid rules only need to be in place by January 2020 (with an extension for one anti-hybrid provision to 2022). Member states can also defer introducing the interest limitation rules until 2024 (unless the OECD turns action 4 into a minimum standard in the meantime) on the grounds that their existing rules are “equally effective.” Some member states, including Austria, France, and Ireland, are understood to have already done this. Other member states have acted on the corresponding BEPS actions and have already introduced rules that are also ATAD compliant, while others already had such rules in place so need take no further action. All these are possible reasons for inactivity on some ATAD provisions but are unlikely to mean that no action at all is required. In certain cases (for example, Germany and the Netherlands), a general delay was caused by government elections during 2017, and in some cases, it may well be that they are waiting to see how other member states implement the provisions. At any rate, we can expect to see a flurry of activity once everyone is back from their summer break.

Nightmare or Business as Usual?

Whatever action member states do or do not take, businesses will need to know what has or has not changed and in which jurisdictions. Knowing what is in the ATAD itself will not be sufficient. In particular, they will need to know what choices have been made, such as the choice between a “passive income” and a “transfer pricing” approach under the CFC provisions. They should be aware of which options have been exercised, such as grandfathering or carryover rules under the interest limitation provisions. In addition, they will need to be aware of detailed implementation choices that may not be immediately apparent from the text of the ATAD, such as the application of the low-tax CFC test to just “tainted” income as put forward in the Dutch consultation draft¹⁶; the possibility under discussion in Austria of applying the exit tax provisions to assets transferred to “taxable” as well as exempt permanent establishments;¹⁷ or the Polish proposal to tax all CFC income rather than just “tainted” income.¹⁸ Most importantly they will need to identify where structural changes may be needed if member states do not provide solutions to problems left unresolved by the ATAD, such as the double taxation that can arise under some holding arrangements. While EU governments and tax administrations may welcome the prospect of increased revenues from tighter antiavoidance rules, it seems likely that it will take time and many sleepless nights before the new rules are fit for purpose.