This is the first Op-Ed of a Symposium on the Apple State Aid case ( C‑465/20 P). More Op-Eds will follow soon in EU Law Live.
The recent final ruling on the Apple State-Aid case surprised many commentators, including this author, who trusted the Court of Justice of the European Union (‘the Court’) would be a competent technical interpreter of international tax law. The proper interpretation of Articles 7 and 9 of the OECD Model Convention (‘MC’) is what must underlie the proper understanding of these infamous cases where the EU Commission found Member States granted illegal state aid primarily via an alleged misapplication of transfer pricing rules in convoluted international structures involving U.S. multinationals. Trust in the technical competence of the Court means trust in the rule of law within Europe, in the quality of EU institutions that ought to reflect the low sovereign-political risk implied in the assurance of fundamental freedoms inherent to the functioning of the internal EU market.
If the final decision of the Court of Justice reflects legal misinterpretation, whether purposeful, in a quest for broader justice under tax or competition laws, or whether accidental, as a showing of the technical inability of the Court to address and interpret very complex transfer pricing notions, both, that would signal an incremental degree of political (sovereign) risk in Europe. And in that sense, increased political risk signals not only a potential increase to the cost of capital for Europe, but, most worryingly, a potential erosion of the quality of EU legal institutions and of the rule of law within Europe.
In the Amazon, Starbucks and Apple cases, as this author has demonstrated in 2016, the main tax effect of the intricate legal structures that were put in place was the deferral of residual U.S. taxation on profits that, under the arm’s length principle (ALP) as interpreted by OECD Transfer Pricing Guidelines, and the Authorized OECD Approach (AOA) for the Attribution of Profits to Permanent Establishments (APPE), were not and should not be attributable to any other country. Prior to the U.S. Tax Reform of 2017 (Tax Cut and Jobs Act (TCJA)), that is, prior to the U.S. reduction of its federal corporate tax rate from 35% to 21%, and prior to the U.S. adoption of its own version of an ‘innovation box’ (Foreign Derived Intangibles Income or FDII) at par with multiple EU regimes, U.S. multinationals that did not adopt complex international structures seeking deferral of the residual U.S. tax burden would be under a competitive disadvantage vis-à-vis their non-US competitors which operated at lower corporate tax rates (and/or benefiting from incentives that are more potent than those explicitly available under U.S. law) and under participation exemption (rather than deferral) regimes.
On the other hand, the structures implemented by U.S. multinationals were so efficient in preventing the triggers of U.S. taxes on residual profits whilst not attributing incremental profits to market countries (for instance, here and here), that such structures could hypothetically have been perceived as distortionary from a competition law or international trade law perspective (perhaps or allegedly income tax subsidies favoring U.S. exports), albeit never materialising any state-aid granted by any non-U.S. jurisdiction. Perhaps U.S. transfer pricing rules and jurisprudence, by not incorporating the AOA/APPE among other features of the OECD Guidelines, would be viewed as too laxed or too formalistic; or perhaps U.S. controlled foreign company (CFC) and entity classification (‘check-the-box’) rules would have too many loopholes. Perhaps not.
Tackling that hypothetical distortion and seeking competitive or trade justice may have been a motivator for the Court of Justice to find against Ireland and Apple. However, the route of illegal state-aid findings is grounded on an apparent misinterpretation of transfer pricing rules which may itself be a poisoned tree – from which the Euro 13 billion (plus) fruits might also be poisoned, infecting the European Internal Market via increased political risk.
Crucially, the lingering mistake and misinterpretation that started with the EU Commission, that was momentarily corrected by the General Court, and that ultimately tainted the Court of Justice’s final decision concerns a gross misunderstanding of the facts underlying the Cost Sharing Agreement (CSA) between the non-resident Irish entities (that have Irish branches) and Apple, Inc., and the relevance of those facts in the interpretation of Articles 7 and 9 of the OECD MC, in the application of the ALP and AOA/APPE. The Court of Justice recites
‘ The Commission recognises in its decision that key functions in relation to the Apple Group’s IP were performed by Apple Inc., either as parent company of the Apple Group or under the cost-sharing agreement, but it explains that that is not relevant for the purposes of the allocation of ASI’s and AOE’s profits among their respective head offices and branches, and relevant only in the light of the reference framework applicable (recitals 308 to 318 of that decision)’ (para. 90).
And it summarises the General Court finding, noting that:
‘More specifically, the Court held: that, in finding that the Apple Group’s IP licences had to be allocated to the branches by default because ASI and AOE had neither employees nor any physical presence outside the Irish branches, the Commission had allocated profits using an ‘exclusion’ approach, that it had not correctly assessed the activities of those companies in Ireland and that it had based its reasoning on an incorrect assessment of normal taxation under Irish law (paragraphs 166 to 249 of the judgment under appeal); that ASI’s and AOE’s branches in Ireland did not control the Apple Group’s IP licences and did not generate the profits which the Commission claimed they achieved (paragraphs 251 to 295 of the judgment under appeal); and that the agreements and activities of ASI and AOE outside Ireland showed that those companies were in a position to develop and manage the Apple Group’s IP and to generate profits outside Ireland and that those profits were, consequently, not subject to tax in Ireland (paragraphs 296 to 311 of the judgment under appeal)’ (para. 92).
Furthermore, the CJEU recalls that
‘ (…) the Court made detailed factual findings, in paragraphs 251 to 310 of the judgment under appeal, about the branches and ASI’s and AOE’s decision-making in the United States and found that the Commission’s claims about the actual activities of the Irish branches and the head offices of those companies were inaccurate’ (para. 100).
However, the Court of Justice delved into a procedural argument and technicality to sustain that the additional evidence concerning Apple Inc.’s functions and activities performed within the U.S. and affecting the intangibles licensed to the Irish non-resident entities was inadmissible in the proceeding (Apple judgment at pars. 142, 146, 147, 180-182, 186, 234, 257) and, embarking in the same (very material) factual and legal error incurred by the EU Commission in its arguments (Apple judgment at pars. 159, 185, 220-222, 235, 236, 254-256, 258), to rule that the U.S. functions and activities were irrelevant for the case. Note, that evidence was supplementary, and thus presented in addition to and reaffirming the CSA itself.
From a transfer pricing perspective, under Articles 7 and 9 of the OECD MC, and in conformity with the OECD Guidelines and AOA/APPE, the CSA should serve to evidence that U.S. employees located in Cupertino performed significant people functions (SPFs) and managed economically significant risks pertaining to the operation of the Apple global value chain (including its global outsourcing of hardware manufacturing to Foxconn) which yields residual profits that should not be attributable to the Irish branches, as it should serve to demonstrate that U.S. employees managed all critical functions, activities and risks pertaining to the intangibles that benefit the entire group and each entity in the value-chain, including those intangibles licensed to the Irish non-resident entities and not managed or controlled by their (taxable) Irish branches which merely performed routine functions. By excluding (or not properly regarding) the functions and activities performed by Apple Inc. through its U.S.-based employees from the transfer pricing analysis, especially where a CSA involving Apple Ind and the Irish entities is in place and in evidence, and therefore by ignoring the Apple value chain within which the Irish entities and branches are inserted, the Court of Justice made a grave error in its interpretation of Articles 7 and 9 of the OECD MC, without which no illegal state aid could have been found as being granted by Ireland.
Oddly, the rulings on Amazon and Starbucks were contrary to the EU Commission, and as such were not surprising at all. What might have been so different about the Apple case, that would have prompted the Court of Justice to embark on what should be perceived as a misinterpretation of the ALP as per the OECD Guidelines and the AOA/APPE?
Is it because it is Apple, the Apple fact pattern and structure? Is it because of Apple’s perceived monopolistic profile (perhaps enabled by the U.S. tax system) and unique combination of hardware sales to consumers and digital services revenues? Is it there where the Court of Justice is drawing the line on what it perceives as unlawful international tax competition or avoidance fueled by the U.S. tax system and perpetrated by U.S. multinationals? Or is it because it is Ireland, and its overall attraction of Foreign Direct Investment (FDI), international trade, foreign reserves, and economic growth, allegedly through tax competition within the EU? Most likely, it is both, Apple and Ireland, while the EU Commission and the Court would have jurisdiction over ‘Ireland, the enabler’. If any of this is hinted in the Court’s ruling, then the State aid rules may not have been the most adequate and legitimate approach to pursue justice under EU competition laws or international trade law. If none of it is there, the only hypothesis left would be the simple misunderstanding of the intricacies of transfer pricing norms. Either way, this judgment is not a testament to the quality of institutions and of the rule of law within Europe.
Any shortcomings, digressions and leaps of thought inevitably present in this Op-Ed remain the responsibility of the author. Note that the viewpoints in this article are not necessarily the viewpoint of the author’s firm.
Romero J.S. Tavares, PhD (Wirtschaftsuniversität Wien) is a Partner at an international Firm and a Professor of International Tax Law.
