In the EU Commission’s view a tax ruling (hereinafter “APA”) confers on the beneficiary a selective advantage under Art. 107(1) of the Treaty on the Functioning of the European Union (hereinafter “TFEU”) insofar as it leads to a lowering of the tax burden by deviating from the tax that the beneficiary would otherwise be obliged to pay under the general corporate tax system.
The legal analysis of the EU Commission does not refer to the arm’s length principle as elaborated by the “OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations” (hereinafter the “OECD Guidelines”).
The starting point of the EU Commission’s legal analysis is Art. 107 TFEU of which the arm’s length principle “necessarily forms part” (see Commission decision of 21.10.2015 on state aid SA.38374 (2014/C ex 2014/NN) implemented by the Netherlands to Starbucks, § 264 hereinafter “Starbucks”).
Accordingly, the arm’s length principle that the Commission applies in its State aid assessment is not that derived from Article 9 of the OECD Model Tax Convention, which is a non-binding instrument, but is a general principle of equal treatment in taxation falling within the application of Article 107(1) of the TFEU, which binds the Member States and from whose scope the national tax rules are not excluded.
Despite the lack of reference to the OECD arm’s length principle, in all recent decisions on tax rulings the EU Commission has made thorough reference to the OECD Guidelines.
The EU Commission does so because it believes “OECD guidelines are an existing manual in the area of transfer pricing that are the result of expert discussions in the context of the OECD and elaborate on techniques aimed to address common challenges of the application of the arm’s length principle to concrete situations” (Starbucks, § 66). The OECD Guidelines also capture the international consensus on transfer pricing and provide useful guidance to tax administrations and multinational enterprises on how to ensure that a transfer pricing methodology produces an outcome in line with market conditions.
It follows that whenever the application of the transfer pricing methodology endorsed in APA is compliant with OECD standards, i.e. it does not depart from the OECD Guidelines, the APA itself does not amount to State aid under Art. 107 TFEU, i.e. the APA beneficiary is not treated more favorably compared to non-integrated companies whose taxable profit is determined by the market.
The present contribution does not comment on the legal analysis substantiating the various EU Commission’s decisions. It is intended to provide an illustration of the manner in which the EU Commission interprets the OECD Guidelines, by pinpointing some remarkable illustrations of such interpretation.
1. Comparability Analysis
According to the EU Commission, the taxable basis accepted in the APA must be substantiated by reference to comparable transactions. In particular, if direct observations can be identified in respect of the related transactions, such observations should serve to determine the remuneration of the company engaging in comparable transactions (Starbucks, § 368). In particular, in the assessment of the arm’s length nature of commercial conditions applicable between related parties, the first step to be taken is to look for and analyse potential internal comparables, if any (Starbucks, § 272).
a. Selection of Comparables
In the EU Commission’s view, the comparability analysis (inherent to the choice of transfer pricing method) should be determined on the basis of the functional analysis of the company for which the APA is requested (Starbucks, § 379). In particular, the Commission endorses the use of comparables database search to estimate arm’s length returns, provided that selected comparables result in a reliable approximation of a market-based outcome. For instance, according to the Commission, companies which do not have a sustainable business model cannot, in principle, constitute reliable comparators when establishing an appropriate level of remuneration (Commission decision of 30.8.2016 on state aid SA.38373 (2014/C) (ex 2014/NN) (ex 2014/CP) implemented by Ireland to Apple, hereinafter “Apple”, § 352).
Moreover, according to the Commission, the approximate nature of the arm’s length principle cannot be invoked by the taxpayer to justify a transfer pricing analysis that is either methodologically inconsistent or based on an inadequate comparables selection (Commission decision of 21.10.2015 on state aid SA.38375 which Luxembourg granted to Fiat, hereinafter “FTT”, § 230).
Finally, it is worth mentioning that the Commission has also referred to industry average determinations (FFT, § 311). However, such approach may be arguable since the OCED Guidelines explicitly affirms that in no event can unadjusted industry average returns themselves establish arm’s length conditions (OECD Guidelines § 1.35).
b. Comparability Analysis under the Transactional Net Margin Method (TNMM)
According to the EU Commission, in order to appropriately estimate the arm’s length remuneration of functions, the taxpayer should carry out a comparison of the functions performed by each party to the related transactions (Starbucks, § 364). As regard the TNMM, the EU Commission maintains that the analysis should take into account the complexity of the functions of all group companies involved in controlled transactions so as to identify the entity to be regarded as the “least complex function” (i.e. the “tested party”) (Starbucks, § 273). On this regard the EU Commission, referring to the OECD Guidelines, agrees that the “tested party” is the one to which a transfer pricing method can be applied in the most reliable manner and for which the most reliable comparables can be found (i.e. it will most often be the one that has the less complex functional analysis). According to the EU Commission, complexity is to be assessed in relative terms or, in other words, by comparison to the other parties involved in the transactions. Thus, for the purpose of choosing the tested party, reference should be made to the “less” complex function rather than in absolute terms to a function which would not be complex (Starbucks, § 366 and § 367).
2. Selection of the Transfer Pricing Method
According to the EU Commission, the OECD Guidelines set certain requirements for the choice of the appropriate transfer pricing method in order to comply with the arm’s length principle (State aid SA.38944 (2014/C), Luxembourg, Alleged aid to Amazon by way of a tax ruling, hereinafter “Amazon”, § 64). Within the framework of an APA procedure, taxpayers and tax authorities should always be able to justify the reasons behind the selection of the most appropriate method. This conclusion has been elaborated by the EU Commission in various recent decisions.
a. Most Appropriate Method vs. Second-Best Method
The EU Commission has stated that the use of the most appropriate transfer pricing method does not rule out per se the existence of a State aid. The choice of appropriate transfer pricing method and of the parameters which support its application must still be tested against the “market-based outcome” standard. Accordingly, the EU Commission has explicitly maintained that the choice of the method and the choice of parameters cannot be an arbitrary choice (FFT, § 242).
By contrast, the choice of a second-best method does not give rise per se to a State aid. For example, where such a method is chosen, but that method is used in combination with an overly conservative set of parameters, the remuneration arrived at might nevertheless result in a market-based outcome or in an overestimated tax burden, in which case a tax ruling accepting that second-best method would not give rise to an advantage for the purposes of Article 107(1) of the TFEU (FFT, § 243).
b. Two-Sided Methods vs. One-Sided Methods
According to the EU Commission, methodologies based on a two-sided approach, i.e. where both companies to the intra-group transaction are analyzed, leave less room to deviate from a market outcome (DG Competition – Internal Working Paper – Background to the High Level Forum on State Aid of 3 June 2016, hereinafter “IWP”, § 20). Thus, theoretically, they are less likely to give rise to State Aid.
This being said, amongst the two-sided methodologies, the Commission’s decisional practice, in line with OECD Guidelines, set a preference for the CUP method. Where applicable, such method is considered the best way for approximating conditions close to normal competition (Amazon, § 73 and FTT, § 245). However, the EU Commission also seems to accept the use of “transactional profit methods”, such as the profit split method. Provided that such method is applied consistently in all jurisdictions involved, the profit split leads to an allocation of the full amount of profits between the companies participating to the intra-group transaction (IWP § 20).
This being stated, the apparent preference of the EU Commission vìs à vìs two-sided methods does not imply that one-sided methods may not be considered state aid compliant. In a number of decisions, the EU Commission has also referred to the TNMM which determines the remuneration of a tested party based on its activity or functions performed. As regards this methodology, the criticism of the EU Commission originated from the fact that it was applied in a wrong manner only. For example, in the Commission decision of 11.1.2016 on the excess profit exemption state aid scheme SA.37667 (2015/C) implemented by Belgium, the use of the TNMM was challenged since its application resulted in the automatic allocation of the remaining profit (i.e. the residual profit) to another group company in a foreign jurisdiction, regardless of any precise information about the activities carried on by such foreign group company.
3. Positioning in the Arm’s Length Range
The EU Commission has taken the view that, in order to avoid the granting of an advantage, the point in the range closest to the most likely market outcome should be used for the purposes of pricing controlled transactions (Commission decision of 21.10.2015 on state aid SA.38374 (2014/C ex 2014/NN) implemented by the Netherlands to Starbucks, hereinafter “Starbucks”, § 396). With regard to the positioning in the range, the EU Commission has also acknowledged that, according to paragraph 3.57 of the OECD Guidelines, the use of the central tendency of the sample minimizes the risk of error due to unknown or unquantifiable comparability defects (FFT § 295).
4. Profit Level Indicators
According to the EU Commission, a profit level indicator is deemed to be appropriate insofar as it is consistent with the taxpayer’s main functions (Starbucks, § 400). Typically and subject to the facts of the case, sales or distribution operating expenses might be an appropriate base for distribution activities when using the TNMM (Starbucks, § 387). In case of profit generated and recorded through a margin on distributed products, the Commission considers sales as a more adequate indicator of profit generating reselling function (Starbucks, § 388).
As regard cost-based indicators, the Commission may challenge both the choice of the taxpayer to include/exclude certain costs from the cost-base and the calculation of the mark-up applied (Apple, § 149). As regards the mark-up, the Commission affirms that it may not be reverse engineered so as to arrive at a target taxable income (State aid SA.38373 (2014/C) (ex 2014/NN) (ex 2014/CP), Ireland, Alleged aid to Apple, § 62).
5. Royalty Payments
According to the EU Commission, should an entity not derive any benefit from the use of an intellectual property licensed, no royalty should be due (Starbucks, § 339). However, the EU Commission maintains that the calculation of the royalty’s amount should always be commensurate to the user’s output, sales or, in some rare circumstances, profits (Starbucks, § 287). Thus, the EU Commission would raise doubts on the arm’s length level of a royalty payment disconnected from the economic value of any underlying IP such as, for example, when the royalty is merely calculated as a residual in the taxpayer’s profit and loss account (Starbucks, § 287).
6. APA and Future Changes
According to the EU Commission, the method accepted by the tax authorities should take into account future changes, if any, in the economic environment and/or in the remuneration levels required which may occur in the years following the ruling application. In the EU Commission’s view, indeed, an agreement between a tax administration and a taxpayer that has no end date makes less accurate predictions as to future conditions on which that agreement is based, thereby casting doubt on the reliability of the method endorsed by the APA (Apple, 364). This applies a fortiori for open-ended rulings.
7. Transfer Pricing Report
The EU Commission acknowledges that the Section V.C of the OECD Guidelines lists the type of information that may be useful when determining transfer pricing for tax purposes in accordance with the arm’s length principle (Amazon, § 64). The APA request must be accompanied by a transfer pricing report to substantiate the choice of a transfer pricing method and the arm’s length nature (IWP, § 12).