Since the press releases confirming negative State Aid decisions in the Starbucks and Fiat tax ruling cases late October, the EU tax law community is still eagerly awaiting the publication of the actual decisions (to be published here and here). Most likely, the decisions will hinge on the concrete transfer pricing calculations made (or accepted) by domestic tax administrations and whether these are in line with the transfer pricing rules applicable for other multinational enterprises (MNEs) in the respective jurisdictions. In previous contributions to this blog, several issues have already been discussed (see here for a first reaction to the press releases, here for the relationship of State aid and other BEPS tools and here for a US perspective). I want to highlight three additional inter-related questions that are raised by these cases:

  • Are Member States obliged to apply the arm’s length principle to allocate taxing rights?
  • Are Member States required to coordinate their tax systems to avoid double non-taxation?
  • Are Member States free to choose any double taxation relief mechanism?

Obligation to apply the Arm’s Length Principle?

The first question appears to be answered affirmatively by the Commission in its preliminary decisions in the recent “tax rulings” cases. It reaches that result by way of a comparison between multinational companies and stand-alone companies. While the latter are taxed on the market outcome of their activities, the former report profits that are affected by their close relationship to other group companies. In the Commission’s view, however, they should be taxed in the same way, which would only be achieved if the reported profits of multinationals are subject to adjustments in line with the arm’s length principle. The OECD transfer pricing guidelines may act as “guidance” for that purpose.

It appears that the Commission is ignoring the concrete rules applicable in the Member State concerning the taxation of MNEs, and instead uses the corporate tax system as a whole as reference system. This raises the question whether stand-alone companies and group companies are really comparable. Economically, the answer must be in the negative: it is precisely the economic benefits from being part of a group that lead to their existence. If taxation in line with ability to pay is the guiding principle of corporate taxation, such economic differences should be relevant to assess comparability (“in light of the objectives of the system in question”). The Commission’s position seems wanting from this perspective.

It also has another weakness, which is best discussed using a simple example:

A Member State makes a corresponding adjustment to prevent double taxation following a non-arm’s-length upward adjustment made by another country. Both adjustments match, but are not in line with the Member State’s view of the ALP (so that it corresponding adjustment would not have been required under the applicable tax treaty).

In this case, the Member State does not follow the ALP and thus, in the Commission’s view, provides a benefit on the taxpayer. Regardless of the questions surrounding comparability in these cases (which may justify a deviation from the ALP, but would still require consistency in adjustments made for all group companies), more fundamental doubts of the correctness of that approach are warranted: from a State aid perspective, the relevant “aid recipient” is not the corporate entity, but the MNE as a whole (the “undertaking”). But if both countries’ tax take is coordinated so that there is no gap between the two, the MNE as such does not reap an advantage, at least if we ignore tax rate differentials and other disparities. If such disparities are to be considered, we arrive at the same problem the ECJ found insoluble in its jurisprudence considering the fundamental freedoms: which country is to blame?

The impact of double non-taxation?

This last point leads to the second question: what if a Member State neither follows the ALP nor ensures a coordinated result pursuant its DTC? Again, the answer seems to depend on the perspective one takes in the analysis. If the MNE as a whole is taken as relevant aid recipient, an advantage is clearly visible, but it is not clear which country should take the blame for this benefit over independently operating taxpayers. If we look, instead, at the corporate entity (the taxpayer), an adjustment leading to non-taxation should not as such be considered to confer an advantage. The following example illustrates why:

Example: Company A and Company B are associated enterprises residing in MS A and non-MS B, respectively. Each company reports profits of 100. Following the application of the arm’s length standard, MS A finds that company A has sold products at inflated prices to company B and adjusts profits of company A downward to 70. Non-MS B, however, fails to make a corresponding upward adjustment of company B’s profits to 130. As a consequence, a portion of the overall profits remains untaxed in either jurisdiction.

If double non-taxation were an issue in itself, the Commission would have to declare the application of the arm’s length principle itself to be wrong in this instance. But it seems difficult to argue that MS A deviated from its “normal tax regime” or that a regime that does not take into account another country’s reaction to a transfer pricing adjustment is inherently providing a benefit to MNEs, as it might just as often be disadvantageous as advantageous. Although this might be the case in the investigation into the Belgium “excess profits” rulings, the Commission appears to be challenging the interpretation of the ALP rather than the principle of a downward adjustment.

In the end, although double non-taxation and double taxation are unwelcome results, and unilateral deviations from the arm’s length standard are a common source for them, State aid law is not the right tool to address either. The only situation the Commission can legitimately challenge is that where an advantage is provided to a selected group of taxpayers.

Freedom to choose virtual double tax relief?

This leads finally to the third question: whether Member States are free to provide relief from mere virtual double taxation, in particular where this is not done in all cases.

It is clear that neither the regulatory technique nor the source of regulation should make a difference for the assessment under State aid law. Prima facie, it is thus immaterial whether relief is provided by way of national law or through a tax treaty. One might argue that the provision of double taxation relief can never be State aid, because it does not provide an “advantage”, but merely negates the disadvantage of double taxation. That is not a foregone conclusion, however. It depends on the definition of the proper reference framework or, as it were, the proper comparison: Are taxpayers who benefit from unconditional exemption of their foreign income (relief from virtual double taxation) comparable to taxpayers who benefit from a tax credit (and thus have to prove actual double taxation)? If the reference framework is domestic law applicable to all taxpayers with foreign income, that will normally be the case, potentially leading to the identification of an advantage for taxpayers benefiting from exemption.

This particular difference in treatment might then still be justifiable by reference to the “nature and general scheme of the tax system”, taking into account both the underlying principle of single taxation (as derived from the ability-to-pay principle) and the Member States’ authority to conclude tax treaties. Such justification would be less convincing, however, if the regulatory technique used to provide relief is not “standard practice” (and thus more easily identifiable as part of the “general scheme”), but rather unusual, such as the availability of sparing credits.

One can escape that conclusion by postulating that a relief provision that is available to all undertakings protected by a particular tax treaty is not selective in nature, as it applies generally and does not make a distinction based on inherent characteristics of these undertakings. The ECJ does not appear to have formed a final view on how closely defined a group of taxpayers has to be targeted for the selectivity criterion to be fulfilled, leaving a possible opening for the Commission to investigate situations where benefits are only available in relation to investments in a selected number of countries.

The Commission seems to be doing exactly that in the most recently launched formal investigation into the tax arrangements of McDonald’s. The case involves the use of an establishment in the US that qualified as “permanent establishment” under the tax treaty between Luxembourg and the US, but was not recognised as having sufficient commercial presence in the US to be taxed there. Despite the fact that the US did therefore not exercise its taxing right, Luxembourg exempted the income that would be allocated to the PE. On the one hand, the Commission seems to take issue at the exemption of profits “despite knowing that they in fact were not subject to tax in the US”. Such double non-taxation is, however, a common (if not intended) consequence of applying exemption without a subject-to-tax clause. The Commission announced its intention to assess whether Luxembourg’s authorities “selectively derogated from the provisions of their national tax law and the Luxembourg-US Double Taxation Treaty” without calling into question the general tax regime in Luxembourg, which indicates, on the other hand, that it is not concerned with general double non-taxation effects flowing from the indiscriminate application of the tax treaty rules.

It seems unlikely, then, that the Commission will broaden its view to a comparison between taxpayers benefiting from a beneficial tax treaty and those that fall under a different one (or none at all). This at least is good news for the Member States’ sovereignty in international tax matters. It remains to be seen, however, whether the Commission will not find a way to argue that certain tax treaty benefits are inherently problematic from a State aid perspective.


Leave a Reply

Your email address will not be published. Required fields are marked *