Last month, Dennis Weber started a debate on recent BEPS-related changes to European tax directives with his post on the General Anti-Abuse Rule in the Parent-Subsidiary. I would like to follow up on this with a short discussion of the second anti-avoidance measure introduced last year through Directive 2014/86/EU, which specifically targets hybrid financial instruments and the issue of double non-taxation following incoherent qualification of payments under such instruments in two EU Member States. The amendment has to be implemented by 31 December 2015 and most Member States have already introduced legislation to that end.
Two recent posts by Ramon Tomazela have extensively covered the OECD proposals under BEPS Action 2, which I will thus not discuss at length here. The principles are by now well known: the OECD relies on a combination of “primary” and “defensive” rules in both source state and residence state in order to achieve effective coordination of unilateral responses much in the way place of supply rules under national indirect taxation systems avoid double taxation without the need for a tax treaty covering such legislation. Although the amendments made at the EU level appear closely connected to the OECD proposals, the initiatives go back further than the BEPS project to various reports and recommendations made by the European Code of Conduct group. However, due to the very limited scope of harmonization of direct tax law in the EU, the actual legislative response of the EU is also much more limited. EU law contains no rules to “link” the qualification of financial instruments in different Member States; nor is there any harmonisation in this respect. EU directives on direct taxation generally address obstacles resulting from double taxation, but they do not concern the corporate tax base in the source state, as the CJEU has most recently clarified for the Interest and Royalties Directive in Scheuten Solar Technology (C-397/09). It is thus unsurprising that the amendment in Art. 4(1) Parent-Subsidiary Directive only includes the equivalent of the OECD’s “defensive” rule by imposing an obligation on the recipient Member State to
“tax such profits to the extent that such profits are deductible by the subsidiary”.
The wording leaves no doubt about the compulsory nature of this newly added phrase: where distributed profits are deductible for the subsidiary, the residence state has an obligation to tax these profits. However, there are a number of open questions regarding the amendment, either directly to that phrase or more generally to the justification of including that rule:
- When is the profit “deductible”? (e.g. does a deduction for purposes of a net wealth tax trigger the obligation to tax?)
- Why a rule for “deductible”, but not for “exempt” profits, although both are the same economically?
- What does it mean to “tax” the profit? (e.g. what about tax base inclusion at a 0% rate? What about a 0.1% rate? Does inclusion in a net wealth tax or subscription tax suffice? (the Parent-Subsidiary Directive defines “taxes” only for purposes of companies’ eligibility))
- Why avoid double non-taxation for capital returns from equity, but not from debt investment? (as there is no equivalent obligation to tax in the Interest and Royalties Directive or any other directive)
For some of these questions, there are more or less clear answers that can be identified by looking at the legislative history and purpose, while some others are much more difficult to resolve and may yet attract divergent views of national legislators and courts. Many more issues are certain to come up in practice over the next few years.
For this post, however, I would like to focus on a different issue: Considering that this is the first time for a direct tax directive to impose an obligation to tax on Member States in this way, did the EU even have the competence to make this amendment? First of all, the situation is quite different from the obligation to tax as a substitute for exchanging information under the Savings Directive. Not only was that obligation an exception and meant to be merely transitory, it also was meant as an effective means of collecting tax for another Member State which that Member State was unable to collect for practical purposes. The Parent-Subsidiary Directive amendment, by contrast, aims to force Member States to collect a tax for themselves that they (apparently) would not otherwise want to collect. Note that, if they desired to collect that tax, they could already do so under the directive’s rules prior to the amendment, simply by applying the credit method for any previously untaxed (or low-taxed) profits.
The anti-hybrid measure thus has a purpose that is very different from that of the original Parent-Subsidiary Directive itself and is also distinct in its obligation to tax from the newly introduced general anti-abuse provision in that directive. The latter prohibits Member States to “grant the benefits of the directive” to abusive arrangements. However, this should not be seen to amount to an obligation to tax, not least because there is no guidance as to how such tax should be calculated. The directive does not require a Member State to “tax” abusive arrangements by way of substituting an arrangement that is not genuine with another one. It primarily ensures that a Member State can apply its own GAAR if it complies with the EU standard.
The EU Commission relied on Art. 115 TFEU in its proposal to amend the directive, which is usually the basis for EU legislation on direct taxation. This provision does not allow for harmonization (only “approximation”) and it has important conditions: a legislative act can only be based on that norm if it is necessary and proportionate for the functioning of the internal market. The Commission starts to justify its proposal only indirectly with regard to that objective: rather than directly improving functioning of the internal market, it sees the change as necessary to “protect the functioning of this directive”. More directly addressing the internal market, it states that the differences in national legislation allow “cross-border planning strategies which lead to distortions of capital flows and of competition in the Internal Market” and the differences in approaches against such planning in turn would undermine the aim of the Parent-Subsidiary Directive to abolish tax obstacles to cross-border grouping of companies. The proposal goes on to state that “single uncoordinated initiatives may result in additional mismatching or in the creation of new tax obstacles” and thus could not effectively address the issue, making the directive’s amendment the only feasible solution.
At first sight, this appears sensible. Uncoordinated tax regimes create market distortions; therefore coordination is necessary to remove distortions. However, I would like to make some observations: first, taxes by their very nature distort markets. Different taxes/rates therefore inevitably create distortions for the internal market. If that is sufficient to give the EU the competence to mandate what to tax and what not, there are hardly any limits to that power and the Member State’s “sole competence” in direct tax matters is a mere illusion. Second, while the line between approximation and harmonisation is not easy to draw, it is arguable that an obligation to tax goes beyond mere approximation: considering the Commission’s concern about uncoordinated responses against double non-taxation, it would have been sufficient to make it explicit for Member States that they are not forced to give relief under the directive for previously deducted profits. An obligation to tax goes beyond that, raising questions about the proportionality of the measure. Third, due to the open questions raised above (e.g. concerning the rate at which a tax would have to be imposed), it is not at all clear that the proposal will achieve the intended abolition of double non-taxation; it may well be rather ineffective, especially when compared to the alternative where Member States are allowed, but not forced, to tax. Fourth, it is also not obvious that the coordination will result in a reduction of existing distortions, since it is not known to what extent such distortions exist. It is also a question of the right benchmark: if the ideal to be achieved is single taxation in either of two Member States, the amendment is a step forward; if the ideal to be achieved is the removal of barriers to cross-border investment, this is less obvious. Fifth, it is a compelling thought that within a regime that prevents double taxation (as an unacceptable distortion to the Internal Market), its counterpart should equally be viewed as a distortion that should be avoided. But this does not in itself give rise to a conclusive argument for forced taxation: the avoidance of double taxation required coordination because each Member State has an inherent interest to collect taxes; if that is the case, however, no obligation to tax is necessary to fight double non-taxation.
The limited scope of the amendment to tackle hybrid mismatches indeed shows that the Commission is also strongly driven by the idea of symmetry. It is otherwise difficult to understand why the counter-measure would be confined to situations where the Parent Subsidiary Directive applies. Hybrid mismatches are just as powerful and distortive when a company holds less than 10% of shares of a subsidiary. The Commission may have considered its mandate limited to closing “loopholes” in the existing directive and avoiding unintended tax benefits to groups of companies (see Directive Recital 2 of 2014/86/EU). But the directive would be just as effective to achieve its objective of avoiding double taxation without it; the amendment is thus hardly necessary to ensure the positive impact of the directive on the Internal Market. If double non-taxation were itself an obstacle to the Internal Market, why limit the scope of the solution? In contrast to the abolition of double taxation, it would not impose revenue losses on Member States, so presumably there would be no losers?
In the end, the seemingly minor change in the directive creates perverse incentives: on the one hand, it may now be beneficial to avoid coming within the scope of the directive by staying below the 10% holding threshold or investing in companies not covered by the Parent Subsidiary Directive (making the CJEU’s Gaz de France (C-247/08) decision – which contrasts with case law on company law directives – rather more significant than previously thought!); on the other, it may even result in situations where it is preferable to invest in third countries, as Member States remain free to provide more beneficial treatment to, i.e., continue to exempt, dividends received from third countries.
If these observations – which in the end all relate to the necessity and proportionality of the measure in light of the internal market objective – raise doubts concerning the Commission’s decision to propose (and the Council’s to adopt) the amendment under Art 115 TFEU, what are the potential consequences? It is unthinkable for a Member State or EU institution to challenge the validity of the directive. A taxpayer, on the other hand, would only ever be affected by a domestic rule that implements it. Unless the implementation itself violates another higher-ranking rule, the taxpayer will have no standing as the implementing law remains valid in the absence of the directive. If implementing legislation violated the fundamental freedoms, the taxpayer could challenge it, but the question of competence would not arise. But if it violated a constitutional rule (which might in some countries include double tax conventions), it may well be relevant: in the absence of the overriding binding effect of the directive, the domestic legislator would then not have been able to impose the tax charge. In such a case, the CJEU would have to rule on the question following a preliminary ruling request from domestic courts.