Revealing a disguise tax harmonization through State aid Control:
the need for a selectivity analysis in line with EU Law
Adv LLM thesis
submitted by
Marta Climent I González
in fulfilment of the requirements of the
'Advanced Master of Laws in International Tax Law'
degree at the University of Amsterdam
supervised by
Sjoerd Douma
co-supervised by
Rutger Hafkenscheid
PERSONAL STATEMENT
Regarding the Adv LLM Thesis submitted to satisfy the requirements of the 'Advanced Master of Laws in International Tax Law' degree:
1. I hereby certify (a) that this is an original work that has been entirely prepared and written by myself
without any assistance, (b) that this thesis does not contain any materials from other sources unless these sources have been clearly identified in footnotes, and (c) that all quotations and paraphrases have been properly marked as such while full attribution has been made to the authors thereof. I accept that any violation of this certification will result in my expulsion from the Adv LLM Program or in a revocation of my Adv LLM degree. I also accept that in case of such a violation professional organizations in my home country and in countries where I may work as a tax professional, are informed of this violation.
2. I hereby authorize the University of Amsterdam and IBFD to place my thesis, of which I retain the
copyright, in its library or other repository for the use of visitors to and/or staff of said library or other repository. Access shall include, but not be limited to, the hard copy of the thesis and its digital format.
3. In articles that I may publish on the basis of my Adv LLM Thesis, I will include the following statement
in a footnote to the article’s title or to the author’s name:
“This article is based on the Adv LLM thesis the author submitted in fulfilment of the requirements of the 'Advanced Master of Laws in International Tax Law' degree at the University of Amsterdam.”
4. I hereby certify that any material in this thesis which has been accepted for a degree or diploma by
any other university or institution is identified in the text. I accept that any violation of this certification will result in my expulsion from the Adv LLM Program or in a revocation of my Adv LLM degree.
signature:
name: Marta Climent I González
Table of Contents
Table of Contents ... III
List of Abbreviations used ... V
Executive Summary ... VI
Main Findings ... VII
1. Introduction ... 1
1.1. Background and Motivation of the Study ... 1
1.2. Research question and scope of the Study ... 2
2. The Normative Framework of Fiscal State Aid ... 3
2.1. Prohibition of State Aid in Competition Law ... 3
2.1.1. The Commission notice on the notion of State Aid as referred to in art.107 (1) TFEU . 3 2.1.1.1. An economic advantage ... 4
2.1.1.2. Selectivity Analysis ... 4
2.1.2. State Aid, uncertainty and the need for Transfer Pricing Rulings ... 6
2.2. Principle of conferral and Tax Sovereignty within the Internal Market ... 6
3. Assessing a selective advantage on Transfer Pricing Rulings. General Court
Enforcement ... 8
3.1. On the way to the Arm’s Length Principle ... 8
3.2. The Arm’s Length Principle for State aid purposes: definition, legal nature and application 8 3.2.1. Inherent in art. 107 (1) TFEU. Applicable regardless Member States have it into their domestic law and in what form ... 9
3.2.2. MNEs’ profits are taxable as if they had arisen from a transaction carried out at market prices 9 3.2.3. The ALP as a tool or benchmark ... 9
3.3. Nevertheless, the Commission does not have the power to define the “normal” taxation autonomously ... 10
4. The Principle of Equal Treatment meant by art. 107 (1) TFEU ... 11
4.1. Market Prices to achieve Equal Treatment ... 11
4.1.1. Does art. 107 (1) TFEU require the Arm’s Length Principle as a profit allocation method? The Forum 187 Judgement ... 11
4.1.2. Does the Arm’s Length Principle lead to a level playing field? The ALP as a selective method 13 4.1.2.1. Lack of uncontrolled transactions: The Arm’s Length Principle is not an exact science 13 4.1.2.2. Lack of comparability: Inherent features of MNEs ... 15
4.1.2.3. Lack of consistency when applying the ALP: The Huhtamäki Case ... 16
4.2. Non-Market Based Method to reach Equal Treatment ... 18
4.2.1. Global Formulary Apportionment as an alternative ... 18
4.2.2. Balancing EU Principles: would a Member State be allowed to get rid of the Arm’s Length Principle? ... 19
4.3. Then, what is meant by “treated on equal terms”? ... 21
4.3.1. Determining the reference framework. ... 21
4.3.1.1. The Purpose of Transfer Pricing Rules ... 23
4.3.1.3. The CIT as the reference framework. The Justification of a potential derogation 24
4.3.1.4. The Reference framework in the absence of TP provisions implemented into the domestic law 26
5.
Conclusions ... 27
Bibliography ... 28
Table of Literature ... 28
Table of Case Law ... 30
Commission Decision ... 30
European Court of Justice ... 31
General Court 33 Table of Official Documents ... 33
Annexes ... 34
Annex A – State aid Criteria ... 34
Notion of “undertaking” ... 34
Granted by a Member State or through State resources ... 34
List of Abbreviations used
ALP Arm’s Length Principle APA Advance Pricing Agreement Art., Arts. Article, Articles
BAoTR Balanced Allocation of Taxing Rights BEPS Base Erosion and Profit Shifting C-, T- Court Case Number
CCCTB Common Consolidated Corporate Tax Base CFC Controlled Foreign Company
CIT Corporate Income Tax
CJEU Court of Justice of the European Union CUP Comparable Uncontrolled Price Method
EC European Commission
ECLI European Case Law Identifier ECJ European Court of Justice
EU European Union
FA Formulary Apportionment
GC General Court
MNE Multinational Enterprise
MS Member States
OECD Organisation for Economic Co-operation and Development
PE Permanent Establishment
TEU Treaty on European Union
TFEU Treaty of Functioning of the European Union
TP Transfer Pricing
TR Tax Ruling
UK United Kingdom
Executive Summary
This thesis provides an analysis and critic evaluation of the recent “EU arm’s length principle” endorsed by the Commission to the State aid control when assessing the selectiveness of a tax ruling. Indeed, it addresses the question of whether the selectivity analysis based on the European arm’s length principle disguises a tax harmonization on direct taxation within the EU.
The keystones of this paper are the recent Fiat and Starbucks decisions insofar as they are the first European judgements in the matter. Yet, extensive ECJ case law and scholars’ literature are considered as well.
In this context, it is first assessed what is meant by the prohibition of State aid and whether the arm’s length principle is actually required to enforce such a European principle. Then, to the extent that Member States are exclusively sovereign to decide how to define their tax jurisdictions, it is discussed whether accepting the arm’s length principle as inherent to State aid control would prevent Member States to opt for a non-market based profit allocation method (such as a global formulary apportionment system). As an inconsistency is found, it is finally suggested a selectivity analysis that might be followed by the European Courts in future decisions that reconciles the two conflicting fundamental principles of the EU (i.e. the principle of conferral and the prohibition of State aid).
Main Findings
The national tax law does not offer guidance, but instead establishes what should be considered as “normal” taxation and provides, at the same time, for the benchmark to be used when assessing whether a provision is selective under State aid rules. Indeed, if an external benchmark is used (such as the so-called, “EU arm’s length principle”), it is being threatened the tax diversity existent in the EU as a consequence of the exercise of the Member States’ sovereignty.
In this sense, a selective advantage must be assessed only when comparing with the general applicable domestic law in force, as doing otherwise, in essence, would override the legal framework as approved by the sovereign legislators.
TP provisions are necessarily at least a part of the reference system as, at this stage of the European integration, Member States are still exclusively competent to define in which way profits should be allocated among tax jurisdictions.
Identifying the domestic TP rules as - at least part of - the reference system which provides for the “normal” taxation would allow the Commission to apply the domestic benchmark as it has been implemented into the national legislation and require a reliable market-based approximation without overstepping its competencies under EU Law.
1. Introduction
1.1. Background and Motivation of the Study
Following the financial crisis in 2008, the G20 countries prioritized the fight against tax evasion and avoidance schemes and launched the BEPS Action Plan.1 Since then, the international tax policy
agenda around the globe has been modified and the European Union has been playing a key role in enforcing the BEPS initiative.2
The Commission as “guardian of the treaties”3 and, especially after the LuxLeaks scandal4 in 2014
where hundreds of tax rulings were made public, launched investigations against MNEs operating in the EU. The EU Competition Commissioner Vestager, stated, in this regard, “all companies, big or small, multinational or not, should pay their fair share of tax”.5
In those circumstances, a few years later, on the 24th of September of 2019, the General Court had the chance of, for the first time, assessing a tax ruling under EU State aid rules. The Fiat6 and Starbucks7
judgments confirmed the Commission’s approach to test the existence of a selective advantage under State aid rules based on the so-called “EU arm’s length principle”.8 In this sense, the Court seems to
consider the ALP as a general principle in taxation inherent to art 107 (1) TFEU, which requires to tax the profits arising from an economic activity “as it had arisen from transactions carried out at market prices”.9
At the same time, the fair taxation agenda, both in the International and the European arena, has made some headway by proposing unitary taxation approaches10 to tackle corporate tax evasion. In this line,
firstly, the OECD is proposing the “Unified Approach”,11 under the Pillar One, with a “new Profit Allocation
1 The “Base Erosion and Profit Shifting” (BEPS) Action Plan is aimed to fight tax planning schemes used by MNEs
getting advantage from loops and mismatches among tax systems to prevent paying tax. For further information, see ‘About - OECD BEPS’ <https://www.oecd.org/tax/beps/about/#history> accessed 10 May 2020.
2 Sjoerd Douma and Alexia Kardachaki, ‘The Impact of European Union Law on the Possibilities of European
Union Member States to Adapt International Tax Rules to the Business Models of Multinational Enterprises’ (2016) Volume 44 INTERTAX page 746.
3 The Commission is in charge of surveillance whether Members States implement domestic rules in line with EU
law and whether the latter is applied correctly and on time. For further information, ‘What the European Commission Does in Law’ (European Commission) <https://ec.europa.eu/info/about-european-commission/what-european-commission-does/law_en> accessed 10 May 2020.
4 The Grand Duchy was considered by MNEs a “magical fairyland” to drastically reduce their tax liability. For
further information, see ‘Leaked Documents Expose Global Companies’ Secret Tax Deals in Luxembourg’ (ICIJ) <https://www.icij.org/investigations/luxembourg-leaks/leaked-documents-expose-global-companies-secret-tax-deals-luxembourg/> accessed 10 May 2020.
5 ‘Tax Advantages Granted by Luxembourg and the Netherlands to Fiat and Starbucks Are Illegal under EU State
Aid Rules’ (European Commission - European Commission)
<https://ec.europa.eu/commission/presscorner/detail/en/IP_15_5880> accessed 10 May 2020.
6 ‘Joined Cases T‑755/15 and T‑759/15’. Fiat and Luxembourg v. Commission, ECLI:EU:T:2019:670. It upheld the
Commission’s 2015 decision finding that the tax ruling granted by Luxembourg to Fiat was not in line with EU State aid rules
7 ‘Joined Cases T-760/15 and T-636/16’. Starbucks and the Netherlands v. Commission, ECLI:EU:T:2019:669.
The Court annulled the Commission’s 2015 decision finding that the Netherlands granted a selective advantage to Starbucks.
8 See press release for further information, ‘Statement by Commissioner Margrethe Vestager Following Today’s
Court Judgments on Two Tax State Aid Cases (Fiat in Luxembourg and Starbucks in the Netherlands)’ (European Commission) <https://ec.europa.eu/commission/presscorner/detail/en/STATEMENT_19_5831> accessed 10 May 2020.
9 For further detail, see section 3.2.2. of this thesis. Starbucks v. Commission (n 7) paragraph 149; Fiat v.
Commission, (n 6) paragraph 141.
10 It is usually based on a formulary apportionment method that follows a non-arm’s length approach. For further
detail, Kerrie Sadiq, ‘Unitary Taxation – The Case for Global Formulary Apportionment’ [2001] IBFD; OECD
Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations (OECD Publishing 2017)
paragraph 1.16 and on.
Rule going beyond the Arm’s Length Principle”. And secondly, the European Commission launched an Action Plan for Fair and Efficient Corporate Taxation,12 which keystone of is the CCCTB,13 a common
system within the EU to calculate the tax base of enterprises. Under this proposed methodology, first the consolidated taxable income is calculated and then it is shared among the Member States in which the company is active, according to an agreed formula.
1.2. Research question and scope of the Study
In these circumstances, this thesis aims at discussing whether the selectivity analysis based on the European arm’s length principle disguises a tax harmonization on direct taxation within the EU. Furthermore, as the EU ALP is found to be inconsistent with EU law, an alternative to the current approach is suggested.
The thesis opens with some legal background regarding the conflicting European principles: the prohibition of State aid in Competition Law, on the one hand, and the principle of conferral and the tax sovereignty of Member States, on the other hand (Chapter 2). This is followed by an analysis of the Fiat and Starbucks recent judgements focused on the concept, legal origin and application the European ALP (Chapter 3). This assessment is then challenged by a discussion of the Principle of Equal Treatment meant by the prohibition of State aid and whether the ALP actually reaches such equal treatment (Chapter 4, Section 1). After that and being aware of the new trends to tackle tax abuse based on unitary taxation mechanisms, it is analysed whether a non-market based profit allocation method would treat in the same manner comparable undertakings (Chapter 4, Section 2). As it is concluded that the EU ALP is not in line with EU law, especially does not comply with the sovereignty of Member States, it is finally suggested a selectivity analysis that might be followed by the European Courts in future judgements (Chapter 4, Section 3).
Pillar One (9 October - 12 November 2019)’ 21.
12 For further detail, ‘Commission Presents Action Plan for Fair and Efficient Corporate Taxation in the EU’
(European Commission) <https://ec.europa.eu/commission/presscorner/detail/en/IP_15_5188> accessed 10 May 2020.
13 ‘Questions and Answers on the CCCTB Re-Launch’ (European Commission)
2. The Normative Framework of Fiscal State Aid
2.1. Prohibition of State Aid in Competition Law
The European Union – formerly the European Economic Community – was created after the Second World War to pursue, among other aims, economic cooperation between the Member States.14 In order
to achieve it, an internal (single) market would be a key feature.15
The drafters of the EU’s constitutive charter agreed on the view that a single market would never work out if Member States were able to unfairly rival between them by, among other ways, granting aid to certain investors in such a manner that their commercial activities would be redirected to the Member State providing such aid – in scenarios in which they would otherwise not naturally invest in that Member State-.16
In this context, Title VII of the TFEU provides for the primary legal framework concerning the concept of State Aid within the EU. Specifically, Chapter 1 of Section 2 establishes the rules on Competition applicable to aids granted by Member States.
The article 107 (1) TFEU constitutes the substance of the prohibition of State Aid. It states that “save as otherwise provided in the Treaties, any aid granted by a Member State or through State resources in any form whatsoever which distorts or threatens to distort competition by favouring certain undertakings or the production of certain goods shall, in so far as it affects trade between Member States, be incompatible with the internal market”.
According to settled case-law,17 the concept of aid has a broad scope and it covers not only subsidies
– granted by a positive benefit -, but also aids stemmed by a favourable tax treatment that mitigates the tax burden that normally would be imposed on a taxpayer. This means that, as the Court has repeatedly held, article 107 (1) “does not distinguish between measures of State intervention by reference to their causes or their aims but defines them in relation to their effects”.18
Pursuant to art. 108 TFEU, the Commission has the exclusive competence to decide on compatibility of State aid with the internal market and to initiate formal investigations when it considers otherwise. Nevertheless, where any of the parties concerned disagree with the Commission’s decisions, it may refer the matter to the Court of Justice. The latter has the ultimate power to determine whether a measure constitutes State Aid or not.19
Apart from the TFEU, the State Aid legal framework is also constituted by secondary and soft law. Due to their importance and relation with the topic discussed in this thesis, the following sections will be focused on the Commission notice on the notion of State Aid as referred to in art. 107 (1) TFEU. 2.1.1. The Commission notice on the notion of State Aid as referred to in art.107 (1) TFEU
The Commission, “with a view to contributing to an easier, more transparent and more consistent application of [the notion of State Aid] across the Union”20 updated, in 2016, its Notice on the notion of
14 ‘The History of the European Union’ (European Union, 16 June 2016)
<https://europa.eu/european-union/about-eu/history_en> accessed 10 May 2020.
15 Oyabradoh Enodeh, ‘The Notion of (Fiscal) State Aid: Decisions of the EU Commission and Case Law of the
EU Courts’ 17, page 1.
16 Werner Heyvaert, ‘European Union / International - Recent Developments in the Field of EU Fiscal State Aid:
Three Cases Concerning Benelux Tax Schemes and the Position of the United States – Foreword’ (2020) 60, No. 5 European Taxation 4.
17 ‘Case C-387/92, Banco Exterior v. Ayuntamiento de Valencia, ECLI:EU:C:1994:100, Paragraph 13’. 18 ‘Case C-159/01, The Netherlands v Commission, ECLI:EU:C:2004:246, Paragraph 51’.
19 For further detail as regards the notification of new aid and recovery of illegal aid procedures, see ‘Council
Regulation (EU) 2015/1589 of 13 July 2015 Laying down Detailed Rules for the Application of Article 108 of the TFEU’.
State Aid.21
This Notice concerns the notion of State aid as referred to in article 107 (1) of the Treaty and is willing to provide for further clarifications on how this article has been interpreted by the Court of Justice of the European Union.22 With regard to questions that have not been dealt with in the European Court yet,
the Commission suggests its understanding of the notion of State aid in that context. Nevertheless, in the latter case, such interpretation will be a subjective guidance, as the European Courts are the primary reference to interpret the Treaties.23
Broadly, according to the Notice, a tax measure may constitute State Aid within the meaning of the article if the following cumulative requirements are met. Firstly, the existence of an undertaking will be required. Secondly, an economic advantage (aid) which the undertaking would not be entitled to under normal market conditions has to be identified. Thirdly, this advantage should have its origin directly or indirectly in State resources and it should be attributable to the State. Fourthly, it should be selective in the sense that only “certain undertakings or the production of certain goods” are being favoured. And fifthly, the aid should (at least) threat to distort the competition by affecting the trade between Member States.
In the upcoming sections, the “economic advantage” and the “selectivity analysis” requirements are going to be further analysed.24 As it will be justified below, these two criteria will be the keystone of the
subsequent analysis and an overview of the current interpretation is essential.
2.1.1.1. An economic advantage
An advantage within the meaning of article 107 (1) TFEU, is defined as any economic benefit which the recipient undertaking could not “have obtained under normal market conditions”, that is to say in the absence of State intervention.25 Whenever the financial situation of the undertaking is improved, an
advantage has been obtained.
This means that, when assessing whether an advantage has been granted through a tax measure, a comparison should be performed between the tax burden triggered by the undertaking, including the controversial tax measure and excluding it.26 In this sense, only advantages will be covered by the scope
of State aid rules, not being arguable that the existence of a discrimination - exceptional burden - might be regarded as the grant of aid in favour of other undertakings subject to the “normal” system.27
Hence, the very existence of an advantage might be considered only when comparing with “normal taxation” or a “system of reference”.28 Furthermore, when assessing whether a tax ruling is granting an
economic advantage, the objective of the tax measure through which it is conferred is crucial.29 In other
words, the determination of a benchmark in the light of the reference system will be necessary and a question arises as to what “normal” taxation is and how it should be achieved.30
2.1.1.2. Selectivity Analysis
A selective measure within article 107 (1) TFEU must favour “certain undertakings or the production of
of the European Union’ (Official Journal of the European Union 2016) (2016/C 262/01) paragraph 1.
21 This Notice was preceded by the Commission Notice 98/C384/03 of 11 November 1998 on the application of
State aid rules to measures relating to business taxation and by a Draft Notice in 2014.
22 The CJEU includes both, the Courts of Justice and the General Court. 23 As stated by art. 19 TEU.
24 See Annex A for further detail regarding the other criteria to be considered.
25 Case C-39/94, SFEI and Others, ECLI:EU:C:1996:285, Paragraph 60.’; Commission Notice (n 6), paragraph 6. 26 ‘Case C-173/73, Italy v. Commission, ECLI:EU:C:1974:1, Paragraph 13’.
27 Sjoerd Douma, ‘Chapter 22: State Aid and Direct Taxation’, European Tax Law. Volume 1. General Topics and
Direct Taxation. (Wolters Kluwer 2018) page 471.
28 ‘Case T-211/04’, Gibraltar v. Commission, ECLI:EU:T:2008:595, para 146. 29 Douma (n 27) page 471.
certain goods”. Then, not all more beneficial measures will fall into the scope of aid, but only those which grant an advantage in a selective manner.31 In this context, a measure might be de jure or de facto
selective, as it should be assessed the favourable effects on certain undertakings regardless of whether the applicability of the tax measure is formulated in general terms.32
The “advantage” and “selectivity” criteria are so closely connected that settled case-law considers that a measure which grants an advantage is prima facie selective.33 At this point, recent judgments are
assessing the selectivity condition through a discrimination test:34 whether a measure is selective will
depend on whether the advantage is conferred on certain undertakings or the production of certain goods with regard to comparable undertakings.35
While a clear distinction between the “selectivity” and “advantage” criteria is made by the Notice, the Court of Justice and the Commission have not maintained a rigorous separation between them.36 For
instance, as it will be further developed in subsequent sections, the FIAT and Starbucks decisions are a clear example of this confusion when assessing whether a “selective advantage” has been granted.37
As regards the selectivity analysis, the Notice provides for a three-step approach38 which is also widely
held by the European Courts.39 First, the system of reference or “normal” taxation must be identified. In
this sense, this reference framework “constitutes the benchmark against which the selectivity of a measure is assessed”.40 Second, it should be analysed whether the controversial measure constitutes
a derogation from that system insofar as it differentiates between economic operators who, in the light of the objectives41 pursued by the system, are in a comparable factual and legal situation.42 If the
measure favours certain undertakings or the production of certain goods, it will be considered prima
facie selective. In the third step, assuming a prima facie selectivity, it needs to be established whether
the derogation is justified by the nature or the general scheme of the reference system.43 For instance,
it might be a possible justification the need to fight against BEPS schemes, the principle of tax neutrality, the progressive nature of income tax and its redistributive purpose or the need to avoid double taxation, among others.44
Identifying the reference system will be a crucial step, as to whether a derogation from it exists, will depend on the objective pursued by that system. For instance, a broad scope of the reference system expands the scope of the selectivity assessment, thereby making more likely a derogation from it being considered illegal state aid. However, where a narrow reference system is identified, the scope of the assessment is reduced and, therefore, it allows Member States to introduce a difference of treatment to a certain extent.45
As stated by the ECJ, “the determination of the reference framework has particular importance in the
31 Commission Notice (n 20) paragraph 117.
32 This was the case of a tax reform in Gibraltar, which de facto favoured offshore companies. See Gibraltar v.
Commission, (n 28); Commission Notice (n 20) paragraph 121.
33 ‘Case C-487/06, British Aggregates v. Commission, ECLI:EU:C:2008:757, Paragraph 83’.
34 ‘Joined Cases C-20/15 P and C-21/15, Commission v. World Duty Free, ECLI:EU:C:2016:981, Paragraph 54’. 35 For further detail regarding the so-called “two-step approach” or discrimination test: Paul-John Loewenthal,
‘Chapter 3. Fiscal Selectivity: A Notion in Need of Clarity’, New Perspectives on Fiscal State Aid. Legitimacy
and Effectiveness of Fiscal State Aid Control. (Carla De Pietro, Wolters Kluwer 2019) page 38.
36 ibid page 34.
37 Among others, Joined Cases T-760/15 and T-636/16, Starbucks and the Netherlands v. Commission,
paragraph137; Joined Cases T-755/15 and T-759/15, Fiat and Luxembourg v. Commission, paragraph 129.
38 Commission Notice (n 6) paragraph 128.
39 ‘Joined Cases C-78/08 to C-80/08, Paint Graphos and Others, ECLI:EU:C:2011:550’ paragraph 49. 40 Commission Notice (n 6), paragraph 132.
41 Regarding the determination of the objective of the tax system, Jérôme Monsenego, Selectivity in State Aid Law
and the Methods for the Allocation of the Corporate Tax Base (Wolters Kluwer 2018) page 75.
42 Commission v. World Duty Free (n 26), paragraph 54.
43 ‘Joined Cases C-78/08 to C-80/08, Paint Graphos and Others, ECLI:EU:C:2011:550’ (n 39) paragraph 49. 44 In any case, it must be an intrinsic basic or guiding principle of the reference system, not being possible to justify
a derogation through external policy objectives aiming at stimulating certain (economic) behaviour. Commission Notice (n 20) paragraph 139; Douma (n 27) page 474.
case of tax measures, since the very existence of an advantage may be established only when compared with 'normal' taxation”.46 Indeed, by precluding advantages, the selectivity test establishes a
principle of equal treatment. The difficulty comes from determining which kind of equal treatment.47
2.1.2. State Aid, uncertainty and the need for Transfer Pricing Rulings
Tax systems are usually tricky and the application to each particular case might be confusing. Precisely, to tackle this uncertainty, tax rulings are needed insofar as they safeguard genuine investments48 while
granting legal certainty for both taxpayers and tax administration.
Rulings have been harshly criticized from a tax policy perspective for their lack of transparency insofar as often no one else than the parties concerned is aware of the content of the agreement reached.49
However, no further issue should arise where the tax administration only anticipates in the ruling the outcome that would have been agreed on anyway at a later stage.50
The Commission stands for a legitimate function of the tax rulings within the EU as long as it respects the State aid rules. Hence, where a ruling agrees on a transfer pricing that does not reflect a reliable application of the ordinary tax system, that ruling may confer a selective advantage to the recipient.51
Due to the unique features of State Aid rules when applying to Tax Rulings, the EC made public the “DG Competition Working Paper on State Aid and Tax Rulings” to offer some guidelines to Member States when applying the arm’s length principle. This paper is comprised of a summary of its preliminary orientations and it does not bind the Commission.52 Moreover, since 2002 the EU Joint Transfer Pricing
Forum, who assists and advises the Commission on TP tax matters, has been elaborating reports “to create a common TP framework for co-ordinating and making the domestic rules work effectively and efficiently together”.53 Among all the initiatives, it should be noted the Communication in the field of
dispute avoidance and resolution procedures and on Guidelines for Advance Pricing Agreements within the EU54 and its complementary documentation.55
2.2. Principle of conferral and Tax Sovereignty within the Internal Market
By way of the founding Treaties56, Member States have conferred competences to the EU to achieve
common aims.57 Indeed, the EU relies, among others, on the principle of conferral (of powers) as
provided by art. 5 TEU, according to which “the EU shall act only within the limits of the competences conferred upon it by the Member States (…). Competences not conferred upon the Union (…) remain with the Member States”. In this sense, the use of competencies by the EU is governed by the principles of subsidiarity and proportionality.58 Furthermore, pursuant to the principle of autonomy, Member States
are not obliged to adapt their domestic legislation to the different tax systems to avoid the double
46 ‘Case C-88/03, Portugal v. Commission, ECLI:EU:C:2006:511, Paragraph 56’. 47 Monsenego (n 41) page 35.
48 Raymond Luja, ‘EU State Aid Law and National Tax Rulings’ page 6.
49 Michael Lang, ‘Tax Rulings and State Aid Law’ [2015] British Tax Review page 394. 50 For the sake of simplicity this discussion will be left out of the discussion.
51 Commission Notice (n 6), paragraphs 169 and 170.
52 ‘DG Competition Working Paper on State Aid and Tax Rulings’ (2016) paragraph 10.
53 ‘Joint Transfer Pricing Forum’ (Taxation and Customs Union - European Commission, 13 September 2016)
<https://ec.europa.eu/taxation_customs/business/company-tax/transfer-pricing-eu-context/joint-transfer-pricing-forum_en> accessed 7 May 2020.
54 Commission of the European Communities, ‘Communication from the Commission to the Council, the European
Parliament and the European Economic and Social Committee on the Work of the EU Joint Transfer Pricing Forum in the Field of Dispute Avoidance and Resolution Procedures and on Guidelines for Advance Pricing Agreements within the EU.’ (2007) COM(2007) 71 final.
55 ‘Commission Staff Working Document. Report Prepared by the EU Joint Transfer Pricing Forum Accompanying
Document to the Communication from the Commission (...) in the Field of Dispute Avoidance and Resolution Procedures and on Guidelines for Advance Pricing Agreements within the EU.’ (2007) SEC (2007) 246.
56 ‘Treaty on European Union (Consolidated Version 2016), OJ C 202, 7.6.2016’; ‘Treaty on the Functioning of the
European Union (Consolidated Version 2016), OJ C 202, 7.6.2016’.
57 Tom Binder, ‘The EU Concept of State Aid’ (2020) 60 European Taxation 9, page 1. 58 Article 5 (1) TEU (n 56).
(non)taxation arising due to the exercise in parallel of their fiscal sovereignty.
As no competence on direct taxation has been conferred, the organs of the EU cannot exercise its own tax jurisdiction in this regard.59 Each Member State is free to decide whether to levy a particular event
and, if so, in which circumstances and what form, regardless of how other Member States are exercising their taxing rights. In the absence of harmonization, unless a measure is considered discriminatory, there is nothing that the EU can do to tackle these forms of harmful tax competition.60
In these circumstances, legislators often implement tax provisions whereby certain taxpayers benefit from reduced tax liability. Through these tax incentives the Governments often pursue other objectives rather than raising the public revenue – which would be the main function of taxes -.61 As a matter of
policy as well, lawmakers will have to decide how to allocate their fiscal jurisdiction using the most suitable profit allocation method. Again, the EU nothing has to say, insofar as Member States also keep exclusive rights as regards the allocation of their powers of taxation with a view to eliminating double taxation.62
However, the “fiscal sovereignty ends where discrimination begins”.63 According to settled case-law,
Member States must exercise their competences consistently with the Treaties,64 which means
observing the four freedoms and the prohibition of State aid, both the core of the internal market.65
A question then arises: being aware of the considerable impact of EU Law limitations – especially the ones stemmed from State aid rules – on Member States sovereignty, where should the line be drawn?
59 Marjaana Helminen, ‘Chapter 1: EU Tax Law as Part of the Legal System’, EU Tax Law. Direct Taxation (Books
IBFD 2019) page 2.
60 Loewenthal (n 35) page 33.
61 For instance, tax incentives are maybe adopted to achieve social, environmental or economic policies.Edoardo
Traversa, ‘Tax Incentives and Territoriality within the European Union: Balancing the Internal Market with the Tax Sovereignty of Member States’ [2014] World Tax Journal page 316.
62 ‘Case C-336/96, Gilly v. Directeur Des Services Fiscaux Du Bas-Rhin, ECLI:EU:C:1998:221, Paragraph 30’. 63 Loewenthal (n 35) page 33.
64 ‘Case C-10/10 , European Commission v. Republic of Austria , ECLI:EU:C:2011:339, Paragraph 23’. 65 Binder (n 57).
3. Assessing a selective advantage on Transfer Pricing Rulings. General
Court Enforcement
3.1. On the way to the Arm’s Length Principle
On 24th September 2019, the General Court – Seventh Chamber, Extended Composition – made public both the Starbucks66 and Fiat67 Judgements. They are the first decisions where it is discussed whether
State Aid as referred to in art. 107 (1) TFEU was being granted through a Tax Ruling. In Starbucks (Netherlands), as the transfer pricings agreed on were considered in line with the ALP,68 no illegal aid
was found. It was decided otherwise in Fiat (Luxembourg),69 as an aid incompatible with the internal
market was declared.
When assessing the existence of a selective advantage,70 the General Court considers the ALP as part
of its principal analysis.
Following the three-step approach applicable to tax measures, the GC first identify the “normal” taxation - as only if the ruling mitigates the burdens normally included in the budget of the undertaking, an advantage might be considered-.71 With no further reasoning, the GC establishes that the CIT should
be the reference framework. Besides, it makes clear that the objective of both the Dutch and Luxembourg corporate tax is to levy the profits of all resident companies, whether stand-alone or integrated, and in the light of that objective, both types of companies are in a similar factual and legal situation.72 Second, a derogation from the reference framework has to be assessed and it is in this
analysis when the ALP is applied.
The following sections will thoroughly examine the interpretation of the arm’s length principle as applied by the General Court, being beyond the scope of this thesis how is applied case-by-case. Other interesting tax issues such as the role of the OECD TP Guidelines, the (non)possibility of neutralization, who supports the burden of proof or the supremacy of TP methods will also be left out of this study.73
3.2. The Arm’s Length Principle for State aid purposes: definition, legal nature and application
According to the ALP, transactions between associated companies should be remunerated as if they had been agreed by independent ones.74 This principle is aimed to assure that “intra-group transactions
were treated for tax purposes by reference to the amount of profit that would have arisen if the same transaction had been executed by independent companies”.75
Indeed, the Court differentiates the fiscal position of integrated and stand-alone companies insofar as intra-group transactions are not subject to market forces and, therefore, transfer pricing does not necessarily reflect market conditions.76
66 Starbucks v. Commission (n 7). 67 Fiat v. Commission, (n 6).
68 For further detail, Mieke Van Dulment, ‘The Netherlands Following the Recent Starbucks NL Decision’ (2020)
60 European Taxation 6.
69 For further detail, Sanja Vasić, ‘Luxembourg Following the Recent Fiat/Chrysler (T-755/15 and T-759/15)
Decision’ (2020) 60 European Taxation 8.
70 Starbucks v. Commission (n 7) paragraph 137; Fiat v. Commission, (n 6) paragraph 129. 71 ‘Case C-522/13, Ministerio de Defensa and Navantia, ECLI:EU:C:2014:2262, Paragraph 22’. 72 Fiat v. Commission, (n 6) paragraph 129; Starbucks v. Commission (n 7) paragraph 137.
73 For further information, CFE Tax Advisers Europe, ‘Opinion Statement ECJ-TF 1/2020 on the General Court
Decisions of 24 September 2019, in Cases C-760/15 & T- 636/16, The Netherlands v. Commission (Starbucks), and Cases T-755/15 and T-759/15, Luxembourg v. Commission (Fiat Finance and Trade), on State Aid Granted by Transfer Pricing Rulings’.
74 Starbucks v. Commission (n 7) paragraph 138; Fiat v. Commission, (n 6) paragraph 130. 75 Starbucks v. Commission (n 7) paragraph 138; Fiat v. Commission, (n 6) paragraph 130. 76 Starbucks v. Commission (n 7) paragraph 148; Fiat v. Commission, (n 6) paragraph 140.
3.2.1. Inherent in art. 107 (1) TFEU. Applicable regardless Member States have it into their domestic law and in what form
The GC refrained from judging the legal nature of the arm’s length principle. Yet, to prevent a too broad unlawful aid concept, it asserted that the EC’s approach “must not be taken out of context and cannot be interpreted as meaning that (…) there was a general principle of equal treatment in relation to tax”.77
The Commission held that the ALP necessarily forms part of its State aid assessment when testing a tax measure granted to an MNE, “irrespective of whether the Member State had incorporated that principle into its national legal system”78 and, if so, in which form. Indeed, the ALP is applied by the EC
as a “general principle of equal treatment in taxation which fell within the application of Article 107 TFEU”.79
Such a statement is grounded on the Forum 187 judgement80 which stands, when analysing the Belgium
TP rules applicable to certain companies, that “transfer prices do not resemble those which would be charged in conditions of free competition”. Moreover, the ECJ observed that when assessing a tax regime, it has to be compared “with the ordinary tax system, based on the difference between profits and outgoings of an undertaking carrying on its activities in conditions of free competition”.81
When the Netherlands and Luxembourg’s tax authorities issued the tax rulings, the arm’s length principle was implemented into their domestic laws. However, one may wonder whether this principle would have been claimed if any of the Member States had other profit allocation method in force. Therefore, a question arises as to whether the alleged arm’s length principle, pursuant to 187 Forum judgement, derives from art. 107 (1) TFEU as a way to enforce the equal treatment required.82
3.2.2. MNEs’ profits are taxable as if they had arisen from a transaction carried out at market prices When performing the selectivity test, the Court held that “where national tax law does not make a distinction between integrated undertakings and stand-alone undertakings for the purpose of their liability to CIT, that law is intended to tax the profit arising from the economic activity of such an integrated undertaking as though it had arisen from transactions carried out at market prices”.83 There
it is found the “State- aid-specific arm’s length principle”.84
Therefore, it might be compared the fiscal burden resulting from the application of the tax ruling with the tax liability that would have resulted “from the application of the normal rules of taxation under national law of an undertaking, placed in a comparable factual situation, carrying on its activities under market conditions”.85 Besides, based on paragraph 95 of 187 Forum Judgement, the GC held that stand-alone
and integrated undertakings, insofar as they are comparable, are to be treated on equal terms.86
Related with these statements, the Chapter 4 will analyse, first, whether stand-alone and integrated companies are in a comparable factual and legal circumstances for State aid purposes and second, whether requiring “a market-based outcome” would achieve an equal treatment between them.
3.2.3. The ALP as a tool or benchmark
Finally, the Court held that the ALP is to be used in the assessment of whether the methodology accepted by the tax rulings had departed “from a methodology that resulted in a reliable approximation
77 Fiat v. Commission, (n 6) para 161.
78 Starbucks v. Commission (n 7) paragraph 139; Fiat v. Commission, (n 6) paragraph 131. 79 Ibid.
80 Forum 187, ‘Joined Cases C-182/03 and C-217/03, Belgium and Forum 187 ASBL v. Commission,
ECLI:EU:C:2006:416’ paragraph 96.
81 ibid paragraph 95.
82 Questions further analysed in chapters 4 and 5 of this thesis.
83 Starbucks v. Commission (n 7) paragraph 149; Fiat v. Commission, (n 6) paragraph 141.
84 Ruth Mason, ‘Implications of the Rulings in Starbucks and Fiat for the Apple State Aid Case’ [2019] Tax Notes
International.
85 Starbucks v. Commission (n 7) paragraph 149; Fiat v. Commission, (n 6) paragraph 141. 86 Starbucks v. Commission (n 7) paragraph 150; Fiat v. Commission, (n 6) paragraph 142.
of a market-based outcome”.87
Indeed, the Court states that the ALP is a tool or a benchmark that allows the EC to check whether the transfer pricing agreed corresponds to pricing under market conditions and, based on its outcome, to determine whether an advantage had been granted within the meaning of art. 107 (1) TFEU.88 When
performing the assessment, an advantage can only be identified if the difference between comparable outcomes “goes beyond the inaccuracies inherent in the methodology used to obtain that approximation”.89
While the statement makes clear that the ALP would be a tool or a benchmark to assess the existence of an advantage, during the whole analysis the Court refers to the determination of a “selective advantage”. Indeed, the ALP is used to find out whether the tax ruling deviates from the normal taxation. At this point, it might be questioned whether transactions at market-prices constitute an inherent benchmark for State aid control or, against the Commission’s perspective, the domestic law should be considered.
3.3. Nevertheless, the Commission does not have the power to define the “normal” taxation autonomously
It is recognized that direct taxation falls nowadays within the competence of Member States. As no delegation pursuant to art. 4 (1) TEU had occurred, the Member States still maintain the competence to designate bases of assessment and to spread the tax burden across the different factors of production and economic factors.90 Nevertheless, the GC stands that these competences should be exercised
consistently and in line with EU Law.91 Hence, a measure in matters of direct taxation is still within the
scope of State aid control insofar as it may cause a distortion incompatible with the internal market.92
In those circumstances, the CG delineates the Commission competences when establishing that in the current stage of harmonization of EU law, the latter is not entitled “to define in an autonomously manner the “normal” taxation of an integrated undertaking, by disregarding national tax rules”.93
At this stage, one may wonder whether the arm’s length principle applied by the Commission and the Court under State aid control constitutes a hidden harmonization on direct taxation as it requires an allocation of taxable profit with a market-based outcome. Indeed, one may ask whether Member States are still entitled to opt for a non-market based profit allocation method. For instance, whether it could be possible to introduce the European CCCTB94 or the new initiatives under the BEPS project on the digital
Economy.95
87 Starbucks v. Commission (n 7) paragraph 140; Fiat v. Commission, (n 6) paragraph 132. 88 Starbucks v. Commission (n 7) paragraph 151; Fiat v. Commission, (n 6) paragraph 143. 89 Starbucks v. Commission (n 7) paragraph 152; Fiat v. Commission, (n 6) paragraph 144. 90 Gibraltar v. Commission, (n 28) 97.
91 ‘Case-269/09, Commission v. Spain, ECLI:EU:C:2012:439, Paragraph 47’.
92 Starbucks v. Commission (n 7) paragraphs 142 and 143; Fiat v. Commission, (n 6) paragraph 104. 93 Starbucks v. Commission (n 7) paragraph 159; Fiat v. Commission, (n 6) paragraph 112.
94 See, ‘Questions and Answers on the CCCTB Re-Launch’ (n 13).
4. The Principle of Equal Treatment meant by art. 107 (1) TFEU
The prohibition of State aid requires the same treatment of undertakings that are – in the light of the reference system - in comparable factual and legal circumstances insofar as the internal market might be threatened otherwise. Therefore, a principle of non-discrimination is meant by art. 107 (1) TFEU, coming the difficulty from understanding what is meant by treated on “equal terms”.96
4.1. Market Prices to achieve Equal Treatment
After identifying the CIT as the suitable reference framework, the Commission used a legal argument97
to justify the arm’s length principle as an external benchmark that enforces the equal treatment required by art. 107 (1) TFEU.98 In this sense, the ALP would grant an equal treatment between stand-alone and
integrated companies by requiring the calculation of the taxable profits in a way that would end up in a reliable approximation of market prices.99
Due to the supremacy of EU law,100 if the arm’s length principle is truly inherent in article 107 (1) TFEU,
the equal treatment required by the prohibition of unlawful State aid must be enforced through it. In this sense, when estimating the taxable profits of associated companies, intra-group transactions ought to be remunerated according to a reliable approximation of market prices - regardless of what is provided by the domestic law -.
4.1.1. Does art. 107 (1) TFEU require the Arm’s Length Principle as a profit allocation method? The Forum 187 Judgement
As it has been detailed, the Commission argues in its Notice that the arm’s length principle forms necessarily part of the State aid assessment under article 107 (1) TFEU. Indeed, the ALP is to be used independently of whether this principle has been incorporated into the domestic tax system.101
Nevertheless, in the Fiat and Starbucks decisions, the General Court avoids a pronunciation in this matter.
A question, therefore, arises as to whether the ECJ, when solving the Fiat’s appeal102, ought to confirm
the arm’s length principle as intrinsically inherent in State aid control. The Commission refers to the Forum 187 judgement103 to justify the use of such a principle. Hence, the purpose of this section is to
determine whether the ECJ truly pretended to establish and endorse the ALP in that judgement. The Forum 187 judgement, far from being related with APAs, concerns the Belgian special tax regime for coordination centres.104 This regulation was applicable to associated entities of an MNE that offered
administrative or financial services intra-group which derogated from the ordinary tax system. Among other selective measures,105 that system allowed the exclusion of certain expenses when applying the
CUP method106 to calculate the taxable income of these coordination centres. This regime was
96 Monsenego (n 41) page 30.
97 From an economic perspectives, are unclear the effects of the EU Tax Jurisprudence. Indeed, it is arguable
whether preventing some discriminatory obstacles to fundamental freedoms necessarily leads to a more level playing field increasing tax neutrality. For further detail, see Rita de la Feria and Clemens Fuest, ‘The Economic Effects of EU Tax Jurisprudence’ [2006] European Law Review page 1.
98 Monsenego (n 41) page 30.
99 Commission Notice (n 20) paragraph 172.
100 ‘Case C-6/64, Costa v. E.N.E.L., ECLI:EU:C:1964:66’. 101 Commission Notice (n 20) paragraph 172.
102 The Commission, as the defeated party in Starbucks has decided not to appeal the GC decision. 103 Forum 187 (n 80).
104 To be qualified as a “coordination centres” under this special tax regime certain requirements had to be met. As
a benefit, the special tax regime provided for a special rule to calculate their taxable profit based on a standard rate according to the cost-plus method. For further detail, see footnotes no. 105 and 106 below.
105 The exemption from property tax on buildings used for professional activities was also considered selective. For
further detail, see T Joris and W De Cock, ‘Is Belgium and Forum 187 v. Commission a Suitable Legal Source for an EU “At Arm’s Length Principle”?’ [2017] European State Aid Law Quarterly page 608.
106 The CUP method compares the price charged for property or services transferred in a controlled transaction to
considered incompatible State aid.
The ECJ when performing the State aid control held the following controversial statements: first, that the tax regime for coordination centres had to be compared with “the ordinary tax system, based on the difference between profits and outgoings of an undertaking carrying on its activities in conditions of free competition”107 and second, when reaching a conclusion, that the “transfer prices [did] not resemble
those which would be charged in conditions of free competition”.108
Did these claims endorse an obligation to apply the arm’s length principle, regardless of the profit allocation method provided by the domestic law? Some aspects should be borne in mind.
Firstly, it should be mentioned that both statements were held when assessing whether an economic advantage was being granted, which is a different test under State aid control than the selectivity one. Belgium had incorporated the arm’s length principle into its domestic law; however, the measure was considered selective not because a deviation from it was found, but because the special tax regime was only applicable to certain undertakings.109
Secondly, the arm’s length principle as such was never mentioned in that decision neither as a benchmark nor as a tool nor as a general principle of equal treatment in taxation. In this sense, it has been claimed that the Commission and the GC are interpreting in a very broad manner the “conditions of free competition” requirement.110
Thirdly, in the Forum 187 Judgement the term “conditions of free competition”111 was never defined.112
In this context, associated enterprises could also be deemed as acting under “conditions of free competition”, as they are subject to market forces as well when doing businesses.113 Moreover,
stand-alone undertakings might also have the opportunity to manipulate the pricing of transactions (for instance, when a franchisor/producer is carrying on businesses with the franchisee/distributor under conditions that the latter would not have otherwise under conditions of free competition).114 Hence, “it is
inaccurate […] to argue that the transactions of stand-alone entities do not deviate from market conditions”.115
Fourthly, when referring to “the difference between profits and outgoings”, legal scholars have interpreted that it was purely a reference to the principle of net taxation prevailing in the corporate income tax.116 In this sense, the arm’s length principle would not exclusively meet this statement, as other profit
allocation methods – in contrast with the arm’s length principle which was implemented in the Belgian tax system – also rely on the concept of income or net taxation117 (i.e. profits remaining after the
deduction of costs) and they can be applied as part of the corporate system.
All things considered, it can be concluded that the arm’s length principle was not irrefutably enforced by the ECJ in the Forum 187 decision when requiring conditions of free competition. The wording of that
et seq. TP Guidelines, OECD (n 10).
107 Forum 187 (n 80) para 95. 108 ibid para 96.
109 Joris and De Cock (n 105) page 614. 110 ibid.
111 A question arises as to whether the ECJ had been referred to “conditions of free competition” if Belgium would
not have had incorporated the arm’s length principle into its domestic law at that moment, but a non-market based allocation method (i.e., formulary apportionment).
112 Raymond Luja, ‘Do State Aid Rules Still Allow European Union Member States to Claim Fiscal Sovereignty?’
2016/5–6 EC TAX REVIEW page 323.
113 Pursuant this possible interpretation, it would not be necessarily and exclusively required to simulate the pricing
between independent parties (i.e. an arm’s length pricing). Monsenego (n 41) page 33.
114 Svitlana Buriak and Ivan Lazarov, ‘Between State Aid and the Fundamental Freedoms: The Arm’s Length
Principle and EU Law’ (2019) 56 Common Market Law Review page 12.
115 ibid.
116 Monsenego (n 41) page 32.
117 For instance, a safe harbour or a system of formula apportionment could also be implemented as part of a
corporate tax system. A contrario, any other profit allocation method which does not form part of an income tax, such a destination-based cash flow tax, would be incompatible with the ECJ findings. ibid page 33.
decision does not necessarily lead to the application of the ALP when remunerating intra-group transactions and even if the ECJ meant so, it is questionable whether the ALP actually reaches an outcome where undertakings on comparable legal and factual circumstances are treated in a similar manner as required by State aid rules.
4.1.2. Does the Arm’s Length Principle lead to a level playing field? The ALP as a selective method Assuming that a principle of non-discrimination is meant by art. 107 (1) TFEU, which implies a general principle of equal treatment in taxation, a question arises as to whether the arm’s length principle actually reaches a level playing field as required by the State aid rules.
In other words, it might be asked whether the ALP would lead to an equal treatment between comparable undertakings and, whether stand-alone and integrated enterprises are to be comparable for State aid purposes – especially in the light of Transfer Pricing rules -.
4.1.2.1. Lack of uncontrolled transactions: The Arm’s Length Principle is not an exact science
The Commission, when considering which one was the most suitable reference framework to be applied in the Fiat Case, established that the Circular118 - which implemented the arm’s length principle into the
Luxembourg domestic law - was too vague for being considered as defining the “normal” taxation.119
Nonetheless, the keystone of the selectivity analyses finally performed by both the EC and the General Court was grounded in the arm’s length principle which, as any principle, is by definition an interpretative legal concept that leads to different understandings and outcomes.120
The ALP was intended as a legal fiction to approximate allocation of profits for the purposes of preventing abusive schemes. However, its application does not necessarily lead to precise and reliable pricings which represent economic and business reality.121 In fact, as a vague legal concept might lead
to different interpretations and, therefore, a variety of outcomes which may result in a more beneficial treatment for certain undertakings.122 Moreover, as mentioned above, it is inaccurate to argue that the
transactions performed by stand-alone enterprises, which are used as a benchmark, are per se not deviated from market conditions.123
The undeniable starting point is the fact that associated enterprises may take part in economic transactions that stand-alone companies would not undertake.124 Nevertheless, that would not
automatically lead to qualifying such a transaction as artificial and against the ALP.125 Instead, that lack
of uncontrolled transactions to compare with will result in need of accepting the “arm’s length range”.126
Therefore, the comparability will be performed between “similar transaction by similar entities in similar circumstances”.127
As Navarro stated, there are four crucial features of MNEs which are not adequately dealt with when the ALP is applied,128 which will henceforth require the use of approximations. First, MNEs’ organisation
tend to prevent transaction costs arisen within imperfect market scenarios due mainly to information
118 ‘Circular, L.I.R. N°164/2, 28 January 2011’.
119 Fiat’s Commission Decision, ‘Commission Decision (EU) 2016/2326 of 21 October 2015 on State Aid SA.38375
(2014/C Ex 2014/NN) Which Luxembourg Granted to Fiat (Notified under Document C(2015) 7152)’ para 321.
120 Monsenego (n 41) page 125. 121 Buriak and Lazarov (n 114) page 18. 122 Monsenego (n 41) page 119.
123 See section 4.1. of this thesis when referring to, for instance, the advantage of mispricing when dealing between
franchisor/producer and franchisee/distributor.
124 The 2017 OECD TP Guidelines in its para. 1.10 explicitly referred to economies of scale and the benefits of
integration (i.e. synergies) in relation to which the separate entity approach followed by the ALP would provide no allocation.
125 TP Guidelines, OECD (n 10) para 1.11.
126 Because Transfer Pricing is not an exact science, the application of the ALP will result in a range of
approximations that will be equally reliable. ibid para 3.55 et eq.
127 Sadiq (n 10) page 277.
128 Aitor Navarro, ‘The Arm’s Length Standard and Tax Justice: Reflections on the Present and the Future of
asymmetries, bargaining and decisions costs. Second, MNEs when cooperating and sharing objectives and risks, create positive synergies and economies of scale which are not reachable by stand-alone undertakings unless they expressly decide so and actively cooperate among them. Third, when intermediates are avoided along the value chain, MNEs are able to sale products or provide services at marginal costs – which result in a lower price that will attract demand -. Finally, MNEs are able to coordinate their transfer pricing policy and decide where and in which form is more suitable to allocate each stage of the value chain. In that sense, integrated companies have the chance to make decisions as regards the production while the business strategy is settled.
Under the arm’s length principle, especially when the Profit Split Method129 is applied, deviations from a
reliable market-based outcome are allowed. However, the GC provides for a limit when staying that any variation should not “go beyond the inaccuracies inherent in the methodology used to obtain that approximation”.130 In this sense, derogations from the open market forces may lead to a difference in
treatment between stand-alone and integrated companies as well as among the integrated ones. Indeed, some scholars sustain that insofar as approximation of market prices are often needed when applying the ALP, such a profit allocation method is (at least) de facto selective131 - especially in
situations where it lacks a direct benchmark in the open market to compare controlled and uncontrolled transactions -.
In this sense, “the reliance on the comparability analysis under the arm’s length principle may capture the diversity of behaviours observed on the open market, but it cannot be taken for granted that such average behaviours correspond to those of the parties to a given intercompany transaction”.132 The
application of the ALP implies some incorrectness which may lead to differences in treatment among stand-alone and integrated undertakings, acting the former as they please and the latter being obliged to simulate market forces when remunerating intra-group transactions.133
Accepting the “market-based prices” as a benchmark to approximate intra-group MNEs profits would lead, in fact, to neutralize the inherent economic advantage of integrated companies134 which does not
mean that then undertakings are treated in the same manner as it is required by article 107 (1) TFEU. Indeed, as stated above, applying the EU arm’s length principle is de facto selective as, by requiring the application of proxies to simulate the market prices - when calculating the taxable profit of associated enterprises -, constitutes a deviation from the general rule, i.e., taxation on the accounting profits.135
Moreover, following the Commission argumentation, a residual profit would remain untaxed since, in comparable situations, an independent undertaking would obtain lower profits.136 This line of reasoning
was adopted by the ECJ in the Belgium Excess Profit Case137 when dealing with the Belgian excess
profit exemption regime which allowed resident companies and PEs of foreign entities to deduct the so-called “excess profit”. That excess profit was calculated as a subtraction of the “profit actually recorded” by the Belgian group and the estimation of the average profit that a stand-alone company carrying on comparable activities in comparable circumstances would have earned.138The difference will remain
129 For further detail about this TP method see TP Guidelines, OECD (n 10) para 2.114 et eq. Insofar the division
of the residual profit is exclusive of integrated companies, the degree of subjectivity and approximation needed is greater – than, for instance, when the CUP is used -.
130 Following this approach, a disagreement is not per se a selective ruling through which unlawful aid is being
granted. In order to consider so, the deviation should be manifest, “should go beyond the inaccuracies”. Starbucks v. Commission (n 7) para 196; Fiat v. Commission, (n 6) para 204.
131 Nevertheless, it cannot be said that the ALP is intrinsically selective as, for instance, when the CUP is applied,
no subjective approximation is required. Monsenego (n 41) page 124. As regards de jure selective, see Ibid page 111.
132 ibid page 115. 133 ibid.
134 As it has been stated before, MNEs operate at marginal costs which are lower than market prices. This have a
positive impact in the demand of their products and services. See Buriak and Lazarov (n 114) page 20.
135 ibid page 10. 136 ibid page 20.
137 ‘Cases T-131/16 and T-263/16, Belgium v. Commission, ECLI:EU:T:2019:91’.
138 ‘Commission Decision (EU) 2016/1699 of 11 January 2016 on the Excess Profit Exemption State Aid Scheme
