1
University of Amsterdam
Master’s Thesis in International and European Law: European Competition Law and Regulation
“Does Article 107(1) allow for external review of the national reference system in fiscal State aid? A consideration of harmful tax competition and its implications for the ‘selectivity’ criterion.”
Student: Simone O’Carroll Supervisor: A. Schrauwen
2
In 2014, the Commission launched investigations into the tax arrangements of prominent multinationals in Ireland, Luxembourg, the Netherlands and Belgium. These investigations were undertaken within the legal framework of Article 107 - 108 of the Treaty which prohibits the provision of State aid that is incompatible with the internal market. The Commission has reached decisions in a number of these cases finding that the advanced tax rulings have resulted in a ‘selective advantage’ to the various multinational enterprises and contravened the prohibition in Article 107(1). The investigations can be seen as part of a broader objective on the part of the Commission to reign in harmful tax competition within the internal market and are in line with movements being made internationally by the OECD as part of its BEPS1 project. The use of State aid rules to combat harmful tax competition within the internal market has however proved controversial as it is perceived as impinging on the reserved competence of Member States in the area of direct taxation. The preliminary findings of the Commission in their investigations placed considerable emphasis on the application of the ‘arm’s length principle’ to transfer pricing agreements. This has been restated in the recently published decisions of the Commission in respect of Starbucks, Fiat and the Belgium Excess Profits case. The decisions reveal a novel approach to incorporating the arm’s length principle into EU law which could have broader implications for the application of the ‘selectivity’ test under Article 107(1).
Introduction
In June 2014, the Commission commenced investigations into the use of advanced pricing agreements2 made between Member States and multinational enterprises (MNE), namely Ireland and Apple, Luxembourg and Fiat Finance and Trade and the Netherlands and Starbucks. These investigations were expanded to include Amazon and Luxembourg and the excess profit rulings in
1 Organisation for Economic Co-operation and Development (OECD) project on Base Erosion and Profit Shifting.
BEPS refers to tax planning strategies that exploit gaps and mismatches in global tax rules to artificially shift profits to low or no-tax locations where there is little or no economic activity, resulting in little or no overall corporate tax being paid.
2 Advanced pricing agreements are tax rulings used to confirm transfer pricing arrangements, which are the
prices charged for commercial transactions, such as the pricing of goods and services including royalty payments and loan interest repayments that are provided between the various entities in a group of companies.
3
Belgium as well as the general tax ruling practice in Gibraltar3. In December 2014, the Commission decided to conduct a general enquiry into the tax ruling practices of all Member States.
By June 2016, decisions4 have been reached in a number of the investigations with findings being made against the Member States involved for the provision of unjustified State aid. Aside from the specific facts in each case, the Commission has consistently held that the respective Member States deviated from standard practice in their application of the arm’s length principle as informed by the OECD guidelines5 in determining the relevant transfer price and thereby allowing the MNE to benefit from a lower tax burden. The OECD guidelines provide for the determination of a transfer price in accordance with the ‘arm’s length principle’,6 which is the internationally accepted standard
requiring
that commercial and financial transfers between associated enterprises should not differ
from the arrangements that would be made between independent companies.
This is to ensure that when profits are allocated between company groups the profits will not be transferred in an artificial way so as to lower the taxable profits in high tax jurisdictions and move them to countries with a lower tax rate or tax havens thus reducing the overall tax burden of the corporate group as a whole.The investigations into the tax rulings come at a time when there has been a greater recognition internationally of the harm that the tax avoidance of multinationals can cause to the revenues of States which became a serious issue in the aftermath of the 2008 financial crisis. In 2013, it was
3 See Commission Decision in Case SA.38373 (Apple) of June 11, 2014 [2014] OJ C369/22 of October 17, 2014;
Commission Decision in Case SA.38944 (Amazon) of October 7, 2014 [2015] OJ C44/13 of February 6, 2015; Commission Decision in Case SA.38375 (Fiat) of June 11, 2014 [2014] OJ C369/37 of October 17, 2014; Commission decision in case SA.38374 (Starbucks) of June 11, 2014 [2014] OJ C460/11 of December 19, 2014; Commission Decision in Case SA.37667 (Belgian Excess Profit Rulings) of February 3, 2015 [2015] OJ C188/24 of June 5/2015
4 See Commission Decision in Case SA.38375 (Fiat) of October 21, 2015 [2014] OJ C369/22 of October 17, 2014;
Commission Decision in Case SA.38374 (Starbucks) of October 21, 2015; Commission Decision in Case SA.37667 (Belgian Excess Profit Rulings) of 11 January, 2016.
5 OECD Transfer Pricing Guidelines 6 OECD Model Tax Convention, Article 9.
4
reported that annually, an estimated 4-10% of global corporate income tax revenues or approximately USD 100 – 240 billon is lost each year through BEPS7. The increased scrutiny of taxing practices of multinationals can be observed in the efforts made by the OECD in this area. In July 2013, the OECD released its Action Plan on BEPS which addressed a number of global tax issues including treaty abuse, hybrid mismatches and transfer pricing8. Within the EU context a number of initiatives have been taken by the Commission in this area which include proposals for a Common Consolidated Corporate Tax Base (CCCTB)9 and a directive on the exchange of information on tax rulings adopted by the Council in December 2015. On January 28th, the Commission adopted the Anti Tax Avoidance Package which includes an Anti Tax Avoidance directive10 which was adopted by the Council as recently as the 21st June 2016. The package aims generally at ensuring effective taxation within the Union where all companies pay taxes where they generate their profits as well as increased tax transparency to ensure fairer taxation.11
This need for increased transparency in corporate tax rulings can be traced back to the controversy which existed in the aftermath of the infamous ‘Lux Leaks’ scandal which provided a new impetus for the Commission in its drive to tackle harmful taxation within the EU. In parallel to the above mentioned legislative initiatives the Commission launched its investigations into the tax rulings of Member States in 2014 under its powers of review within the State aid legal framework
Yet the question remains whether State aid is the appropriate means of tackling harmful taxation practices. State aid requires that a measure must favour “certain undertakings or the production of
7 See http://www.oecd.org/ctp/beps-about.htm
8 The final OECD/G20 BEPS package was endorsed in February 2016 by G20 Finance ministers and includes a
number of recommendations to tackle global tax issues.
9 The CCCTB would create a single set of rules on how EU corporations calculate their EU taxes and provide the
ability to consolidate EU Taxes. A proposal was first introduced in March 2011. See http://ec.europa.eu/taxation_customs/resources/documents/taxation/company_tax/common_tax_base/ccct b_citizen_en.pdf
10 COM (2016) 23 final, Proposal for a Council Directive laying down rules against tax avoidance practices that
directly affect the functioning of the internal market.
5
certain goods”. In other words, it must be ‘selective’. This normally takes the form of a measure deviating from a general scheme which allows an economic advantage to accrue to a selective group of undertakings. Within the area of direct taxation, the ‘normal’ or general reference scheme is that, as implemented in the respective Member States. The approach taken by the Commission in its investigations could be interpreted as departing from this principle so as to allow for an external review of the national reference framework through the application of international standards, namely the arm’s length principle under the OECD guidelines. This would be in line with an expansive approach of selectivity which was potentially endorsed by the Court of Justice in the Gibraltar case12. Such an expansive understanding of the scope of selectivity would however significantly alter the State aid formula and would have serious consequences for the taxing autonomy of Member States. In the aftermath of the initial investigations into tax rulings, a number of decisions have been published which provide an insight into how the Commission will approach the selectivity requirement in fiscal State aid cases.
In
May 2016,the Commission published its first decision in respect of tax rulings
in the case of the Belgium Excess Profits regime where it appears that the Commission may be retreating from such a broad stance on the test of selectivity and may be re-aligning itself with the traditional test, albeit with a novel approach regarding the incorporation of the arm’s length principle into EU law. The Commission’s interpretation suggests that non-discrimination is a fundamental aspect of the review of national tax regimes which derives from the general principle of equal treatment under EU law and its application within the framework of State aid law. The extent to which such a principle will allow for the review of national tax regimes for abuse of law and harmful taxation remains unclear.
Methodology
6
The purpose of this thesis is to determine whether the application of the test of ‘selectivity’ under state aid law as applied by the Commission in its recent decisions on tax rulings stretches the scope of the concept beyond the limits imposed in Article 107(1).
It is intended to rely primarily on the case law of the European courts and the decisions of the Commission in respect of tax rulings as well as national law, where relevant, to determine the outer limit of the selectivity test. Reliance will also be placed on the Commission’s draft notices and communications in respect of State Aid. Finally, I intend to consider and incorporate some of the varying perspectives of academic writings in this field.
In conclusion it is hoped to ascertain whether the Commission’s use of state aid law to deal with member states who offer tax advantages to multinationals residing in their jurisdictions is the correct approach. The Commission’s findings in respect of the tax rulings place considerable emphasis on the OECD guidelines and the tax standards prescribed to the exclusion of a comprehensive analysis of whether there is a deviation from national law. In this regard, it is hoped to show that a more restrictive approach to selectivity will assure that the correct balance is struck between the sovereign prerogatives of Member States in the area of direct taxation and the observance of the prohibition of illegal State aid under Article 107(1).
I will thus first consider the notion of State Aid and then the relationship between fiscal aids and harmful tax competition and the role and limits of state aid law in this area. I will focus on the case of Commission v Gibraltar 13 which may well be open to an expansive interpretation of the application of
State aid law and perhaps the dispensing of the requirement of a national reference framework in certain fiscal aid cases. I will then turn to a consideration of the Commission’s recent investigations
7
into the tax rulings of Member States and the implications of the Commission’s reasoning on the scope of the ‘selectivity’ test.
Chapter 1
The Notion of State Aid 1.1 Article 107(1)
Article 107(1) prohibits state aid, providing “any aid granted by Member States 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”. There are thus four elements to the test which must be met cumulatively: (i) the measure must be imputable to the State and financed through State resources; (ii) it must confer an advantage on the recipient; (iii) that advantage must be selective; (iv) the measure must distort or threaten to distort competition and affect trade between Member States. If the measure falls within the meaning of Article 107(1) it is automatically incompatible with the internal market unless it can benefit from the exceptions in Article 107(2)(3).
The concept of economic advantage is quite broad. As it involves the transfer of State resources, it not only covers traditional subsidies but also refers to tax reductions which can take the form of rebates, exemptions and other fiscal measures which result in a loss of tax revenue to the State14. It is considered an advantage as it involves the reduction of liabilities normally borne by an undertaking through the ordinary course of business, thus giving the undertaking a competitive advantage in comparison to others. This in turn can have an impact on the internal market, distorting competition and affecting trade.
14 It embraces not only positive benefits, but also measures which, in various forms, mitigate the charges which
are normally included in the budget of an undertaking and which, without therefore being subsidies in the strict meaning of the word, are similar in character and have the same effect. (Case 30/59, De Gezamenlijke Steenkolenmijnen in Limburg v High Authority [1961] ECR 1, 19
8
Any aid which comes within the ambit of Article 107(1) must be notified to the Commission prior to implementation. A failure to notify can result in a subsequent finding of illegal State aid by the Commission which will require a Member State to retrieve the aid awarded up to a period of 10 years. This can have serious consequences for any undertakings that have benefitted from illegal State Aid and in the context of the recent investigations into the tax rulings of Member States may require the payment of multiple millions even billions of Euro in unpaid tax.15
1.2 The Selectivity Test
Selectivity is generally the decisive criterion in assessing the compatibility of a measure with Article 107(1). Selectivity is normally what determines the boundaries between what measures are general and therefore do not need to be notified and those that have the relevant specificity and therefore come within Article 107(1). As Article 107(1) determines that only aid that favours certain undertakings or the production of certain goods is to be deemed state aid it is necessary to determine as a preliminary step what is the normal or general rule from which the specific or selective measure provides a derogation. Differentiation of treatment on the basis of products, sectors, regions etc will be deemed prima facie selective. Once a derogation is found then it may be justified by the overall logic or nature of the scheme. This ‘three step’ methodology was first clearly formulated by the Court of Justice in Adria-Wien Pipeline case where it stated:
“The only question to be determined is whether, under a particular statutory scheme, a State measure is such as to favour certain undertakings or the production of certain goods within the meaning of Article 107(1) of the Treaty in comparison with other undertakings which are in a legal and factual situation that is comparable in the light of the objective pursued by the measure in
15 In the Fiat and Amazon cases the estimated aid is calculated at between 20-30 million euro while Bloomberg
reports that Apple may be required to repay 8 billion USD. (see report at
http://www.bloomberg.com/news/articles/2016-01-15/apple-may-be-on-hook-for-8-billion-in-taxes-after-europe-probe)
9
question […] According to the case-law of the Court, a measure which, although conferring an advantage on its recipient, is justified by the nature or general scheme of the system of which it is part does not fulfil that condition of selectivity.”16
Selectivity will not be found when a measure applies generally to all undertakings in a similar legal and factual situation. Therefore any economic advantage that is provided which benefits all undertakings in a Member State does not come within the definition of state aid.17 Similarly, in the absence of harmonization in the area of direct taxation, and therefore the lack of any Europe-wide ‘general’ reference system, the traditional view has been that it is the general level of taxation of a Member State which forms the reference system18. Consequently, fiscal measures such as a general rate of corporation tax are deemed general measures and fall outside the remit of Article 107(1).
1.3 General vs Selective Measures
One of the most difficult aspects of the selectivity test can be determining what measure is the ‘general rule’ and what is the exception. Advocate General Saggio observed that “it does not seem possible to identify with any degree of certainty a criterion of a general nature which provides a clear demarcation line between general and selective measures.”19 It is however accepted that measures which apply generally to all undertakings in a similar legal and factual situation will not be regarded as selective. Therefore any economic advantage that is provided which benefits all undertakings in a Member State falls outside the ambit of Article 107(1).20 Furthermore, tax exemptions or rebates can be introduced by a Member State if they are justified in accordance with their legislative or regulatory
16 Case C-143/99 Adria-Wien Pipeline and Wietersdorfer & Peggauer Zementwerke [2001] ECR I-8365, para 41. 17 Case C-66/02, Italy v Commission [2005] ECR I-10901, para 99.
18 Case C-88/03 Portugal v Commission [2006] ECR I-7115, para. 56
19 Cases C-400/97 C402/97, Administration del Estado v Juntas Generales de Guipuzcoa [2000] ECR I-1073, per
Advocate General Saggio at para 33.
10
regime whereby all persons who fulfill certain criteria and are in a comparable legal and factual situation are capable of benefitting. 21
When determining whether a fiscal measure is selective the first step is to ascertain what the general reference framework of the tax is. The Court of Justice held in the Paint Graphos case that:
“in order to classify a domestic tax measure as ‘selective’, it is necessary to begin by identifying and examining the common or ‘normal’ regime applicable in the Member State concerned. It is in relation to this common or ‘normal’ tax regime that it is necessary, secondly, to assess and determine whether any advantage granted by the tax measure at issue may be selective by demonstrating that the measure derogates from that common regime inasmuch as it differentiates between economic operators who, in light of the objective assigned to the tax system of the Member State concerned, are in a comparable factual and legal situation.”22
A derogation can be either in the form of material or regional selectivity. Regional selectivity relates to specific measures introduced in a geographical region23 whereas material selectivity can refer to the provision of aid to specific sectors or undertakings or certain sized undertakings (for example measures only applicable to MNEs that are not justifiable). A derogation can arise in one of three ways, either the measure is specifically exempt from the normal provision (de jure), or there is a degree of discretion on the part of the tax or administrative authorities which can result in selectively determining the eligibility of certain undertakings for aid, finally, a measure can be de facto selective,
21 Case T-399/11, Banco Santander SA v Commission EU:T2014:938, para 49; Case T-219/10, Autogrill Espana SA
v Commission EU:T:2014:939, para 45
22 C-78/08 to C-80/08 Minestero dell’Economica e delle Finanze e.a. V Paint Graphos e.a. [2011] ECR n.y.r; citing
Case C-88/03 Portugal v Commission [2006] ECR I-7115, para. 56
11
which can arise when what appears to be a general measure applying objective criteria, is in reality capable of benefitting only a certain specified group of undertakings.24
The Court of Justice has made clear that Article 107(1) does not distinguish between the causes or the objectives of State aid, but defines them in relation to their effects.25 Therefore the objective pursued by a fiscal measure is not sufficient to exclude the measure outright from a classification as aid.26 General objectives of an economic nature, or of environmental or social policy will thus not be shielded from review under State aid law27.
The fact that a measure relates to either an exemption or rebate from a general rule or that the measure or fiscal criteria are so narrowly defined will not impact on the finding of selectivity. For instance a measure which introduced a levy on aggregates for environmental purposes but did not extend to all producers of aggregates was defined too narrowly according to its object and was therefore deemed to be the provision of aid to the undertakings not subject to the new tax. The tax base introduced was deemed too narrow and could not be justified in accordance with the underlying logic of the scheme.28 Not only must measures not be selective, but equally a general measure must be applied coherently throughout the tax regime.
A derogation can be defined quite broadly whereby “neither the large number of eligible undertakings nor the diversity and size of the sectors to which those undertakings belong provide any grounds for
24 Case 203/82, Commission v Italy [1983] ECR 2525, para 4. A measure provided that the Italian State would
take over a portion of employers’ contributions into their employees’ sickness insurance fund. A greater reduction of contribution was allowed in respect of female workers which was deemed to indirectly benefit the textile industry which had a higher proportion of female workers than in other sectors and was thus deemed to be unlawful State aid.
25 Case C-487/06 P British Aggregates v Commission [2008] ECR I-10505, para 85. 26 Ibid, para 84.
27 See, supra note 16, Adria-Wien Pipeline, which concerned an environmental levy allowing for a rebate which
was restricted to those in the manufacturing sector. The rebate was deemed selective for excluding other sectors, namely the services sector, who were comparable, both sectors liable to cause an equal amount of damage to the environment.
12
concluding that a State initiative constitutes a general measure of economic policy”.29 As such, what
may appear to be general, can in fact be deemed selective. The question is whether the ‘effects’ of the measure result in the treatment of comparable undertakings in an unequal manner in accordance with the objective of the measure. This incorporation of an ‘effects’ based analysis into the Court’s jurisprudence has thus resulted in a broadening of the scope of what can be deemed selective, meaning that the Court and the Commission have greater control and powers of review over tax measures. This has largely been at the expense of Member States’ fiscal autonomy and has been criticized as some differentiations in fiscal policy which are necessitated and legitimate may be captured by State aid law30. Bartosch writes, in this context, that the “….the notion of material selectivity is (from the perspective of the Member States) a worryingly broad one31.
In line with the effects based analysis, the preliminary step itself of determining the national tax system has also been called into question when the parameters of the State’s fiscal regime cannot be clearly defined or when it has been defined in such a way as to reverse engineer a general system which is in fact discriminatory. This can be seen in the Gibraltar case and has consequences in respect of how harmful taxation can be dealt with through the State aid rules. One reading of the judgment suggests that the objective of the Gibraltar Government which was to introduce various manageable tax bases was in fact substituted by the Court of Justice with what it deemed an appropriate objective, namely the taxation of all corporate entities in Gibraltar, and as the ‘effect’ of the Gibraltar measure as designed by the Government, deliberately excluded offshore undertakings that were in a comparable situation, it was selective.
29 Ibid.
30 See concerns regarding the broadening of the scope of selectivity test, namely, GA Jacobs, Opinon of 7 May
1996, C-241/94 France v Commission, para 30; GA Geelhoed, Opinion of 18 Sep. 2003, C-308/01 Gil Insurance, para 74; GA Maduro, Opinion of 12 January 2004, C-237/04 Enirisorse, paras 45 et seq.
31 A. Bartosch, Is there a need for a rule of reason in European State aid law? Or how to arrive at a coherent
13 Chapter 2
2.1 The interplay between Fiscal Aids and Harmful Tax Competition
“State aid review should be viewed as the appropriate instrument to remedy most of the situations giving rise to inappropriate low taxation that results from the exploitation of legal or regulatory mismatches in member states’ taxation of cross border transactions, including those resulting from the application of income tax treaties.” 32
This is the view of Rossi-Maccanico, who is a member of the Legal Affairs department of the Commission. However as fiscal State aid is concerned with tax measures that are selective within a Member State it is not immediately clear what role it can play in preventing harmful tax competition which is primarily an inter-state concern.
The objectives of the prohibition of State aid are two-fold. Firstly, the support of industry or undertakings in one Member State through the provision of tax exemptions or rebates, will bolster undertakings within that jurisdiction, leading to them obtaining a competitive advantage within the jurisdiction but also potentially within the EU and as a result can distort competition and affect trade within the internal market resulting potentially in a proliferation of subsidies. Furthermore, such measures can also be internally harmful, as the likelihood is, that public policy will be dictated by powerful special-interest groups who will secure benefits to the detriment of other competitors within the localised market upsetting the competitive environment.
Harmful tax competition refers however to beggar-thy-neighbour policies, whereby States introduce tax benefits for the sole purpose of attracting foreign business with the result of reducing the revenues
14
of that other State. Unlike, normal State aid, harmful tax competition normally results in a form of reverse discrimination as it tends to treat foreign undertakings in a more favourable manner than domestic undertakings. State Aid, on a preliminary view, appears therefore to be an odd instrument to be used to combat harmful taxation.
Prior to 1997, the Commission was cautious in using its powers under State aid rules in the area of direct taxation. Member States under Article 115 have retained their competence in this field and any attempts to harmonise taxation to address matters such as harmful taxation would require movement on the political level and the unanimity of all Member States. It is clear however that State aid rules do have a role to play in the area of direct taxation as can be seen in the early case of Italy v Commission33 where the Court held that the residual autonomy in taxation did not affect the
application of State aid law. It is of course a truism of EU law that in the area of direct taxation, while Member states retain their autonomy they must exercise this in accordance with their EU obligations, as is clear in the cross-border taxation cases which impact on the fundamental freedoms34.
The question however remains what role State aid rules can play in the fight against harmful taxation. While tax competition in itself is not harmful,35 it became clear after the release of the Monti report36 that harmful taxation was an issue within the community. Harmful tax competition can be described as the use of tax schemes, rules or other services by States which have the sole purpose of attracting foreign mobile business activity37. The Council, in response to the report, adopted a resolution on a
33 Case C-173/73, Italy v Commission [1974] ECR 709
34 Case C-319/02 Manninen [2004] ECR I7477, para. 19; Case C-292/04 Meilicke and Others [2007] ECR I-1835,
para. 19; Case C-157/05 Holbock [2007] ECR I-4051, para 21.
35Tax competition can be beneficial as it forces States to increase efficiencies in its tax system which can
ultimately result in lower tax pressures on tax payers. Such systems are attractive for domestic and foreign business alike and cannot be regarded as harmful.
36 Monti Group, Taxation in the European Union (First Report of Monti Group), SEC(96) 487 [1996]
37 The OECD has listed a number of effects of harmful tax competition, including “distorting financial and,
indirectly, real investment flows; undermining the integrity and fairness of tax structures; discouraging compliance by all taxpayers; reshaping the desired level and mix of taxes and public spending; causing undesired shifts of part of the tax burden to less mobile tax bases, such as labour, property and consumption; and increasing
15
code of conduct for business taxation (Code of Conduct)38 which identified a number of harmful practices within the internal market39. The code was specifically designed to detect measures that distort the location of business activity in the Community, in that they apply only to non-residents and grant them more favourable tax treatment than that which is normally applicable in the Member State. Many Member States in light of the Code of Conduct urged the Commission to review its policy in the area of State Aid with a view to employing its powers under the State aid rules to combat harmful taxation. The Commission drew up guidelines on the application of the State aid rules to measures relating to direct business taxation in 1998 (1998 Notice)40.
The code of conduct thus informs the approach of the Commission in the field of State aid where its relevance in decision making is reflected in paragraph 30 of the guidelines which states “the qualification of a tax measure as harmful under the code of conduct does not affect its possible qualification as State aid. However the assessment of the compatibility of fiscal aid with the common market will have to be made, taking into account, inter alia, the effects of aid that are brought to light in the application of the code of conduct.”41
The Commission has used the 1998 Notice and the Code of Conduct to carve out its policy in the application of State aid rules to harmful taxation practices. In the absence of harmonisation on a legislative level, the documents have had a considerable influence on the boundaries of State aid law and sovereign fiscal competence. As Fantozzi observes “the discretionary character of the State Aid
the administrative costs and compliance burdens on tax authorities and taxpayers”. See OECD (1998), Harmful Tax Competition: an emerging global issue, OECD Publishing, p.16
38 [1998] OJ C2/1.
39 Five criteria were identified in paragraph B of the Code as a means of evaluating harmful tax regimes, these
included: Ring-fencing or making advantages available only to non-residents; advantages granted when no real economic activity or no substantial economic presence existed in the State; transfer pricing rules for profit determination and lack of transparency of the regime.
40 [1998] OJ C384/3.
41 A. Fantozi, “The Applicability of State Aid Rules to Tax Competition Measures: A Process of “De Facto”
16
decision making process gives the Commission a substantial freedom of manoeuvre in taking State aid decisions. Moreover, the lack of formal legislation has given to the “soft law approach”, in the State aid field a significance which goes beyond the meaning of the sentence in other Community law fields.”42
While there can be some overlap between harmful tax competition and illegal State aid, the use of the State aid framework to tackle harmful tax practices has met with some criticism notably Advocate General Jaaskinen in his opinion on the Gibraltar case where he found that:
“the legitimate objective of combating harmful tax competition cannot justify distortion of the European Union’s legal framework established in the area of competition law applicable to State aid, or even the adoption of ad hoc solutions conflicting with the rule of law as enshrined in Article 2 TEU”.43
One of the primary difficulties with an over-expansive use of the State aid rules to combat harmful taxation is that it may require the dispensing of the initial step in the ‘selectivity’ test which is to determine the national reference system or benchmark. Once that system is coherent it should not be possible under State aid law to find an abuse of law per se simply because the consequence of the regime is detrimental to tax competition.
42 For example, the origins of the use of the OECD guidelines by the Commission as a means of reference in
determining whether transfer pricing arrangements are harmful can be traced back to their reference in the Code of Conduct where it states that “account should be taken of (…) whether the rules for profit determination in respect of activities within a multinational group of companies departs from internationally accepted principles, notably the rules agreed upon within the OECD (..)”. See supra note 40, para B No. 4. Despite the non-binding nature of the guidelines, the Commission included them as a point of reference in its decision in the Belgium and Forum 187 Case when assessing whether transfer pricing arrangements amounted to State aid. The OECD Guidelines have since been used in a more comprehensive manner in the initial findings of the Commission in the tax rulings investigations. Indeed the arm’s length principle according to the Commission can be deemed as hard law given that in the recent Fiat and Starbucks case the Commission deemed the principle as having been endorsed by the ECJ. (The Starbucks case, para 264 and the Fiat case, para 228).
43 Advocate General Jaaskinen, Opinion of 7 April 2011, Case C-106/09P and C-107/09P, Commission v
17
This reality has also played out in the field of direct taxation where Member States are deemed to retain their autonomy, and in the area of the fundamental freedoms. The lack of harmonisation in the field of direct taxation has resulted in disparities to arise between the tax regimes of the various Member States which can result in mismatches leading to double taxation or double non-taxation. The former being an impediment to free movement while the latter can allow for tax avoidance and aggressive tax planning on behalf of MNEs. In the field of fundamental freedoms it has been held that when these disparities arise as a result of two States running their tax regimes in parallel there can be no remedy under EU law, as in the absence of harmonisation, the Courts cannot demand that one regime unilaterally alter their system to bride the ‘gap’ in the law, as neither system, in itself is at fault. 44
Equally, in the area of State aid law, without harmonisation and a uniform reference system, the applicable system must be that of the Member State, as Luja finds “one cannot expect an EU member State to compensate for hybrid situations and foreign check the box regulations by giving up its own legal standards unilaterally. Neither does current State aid doctrine demand anti abuse regulations for situations that do not lead to the abuse of domestic law”.45 As a consequence when the tax system is internally coherent and the measures introduced are ‘general’ within the meaning of State aid, they traditionally do not come within the ambit of Article 107(1)46.
It thus followed that without tax integration on the EU level the role of State aid in the area of harmful taxation is limited as the focus of Article 107(1) is on one Member State and its regime as opposed to the EU as a whole. Thus internal disparities such as when the domestic system fails to abide by its own
44See generally, B. Terra, P. Wattel, European Tax Law, Kluwer Law International (2012) p 59
45 R. Luja, “EU State Aid Rules and their Limits”, (2014) issue 4 Tax Notes Int’l, p 353
46 No aid decision C(2009)4511 final, Dutch Group Interest Box, Official Journal of the EU (OJ) L 288/26 of 4
18
logic can be addressed but external disparities are outside its remit. State aid rules can however still prove effective, for example in the Foreign Income case in Ireland, where multinational corporations were treated as ´exempt’ from taxation on foreign income despite the application of a credit system in the normal tax regime. The Commission found that the system operated on ambiguous criteria and the exemptions were not within the logic of the tax system.47
This position marks the traditional approach to State aid law in the area of harmful taxation. The recent agreement by the Council of a draft directive addressing tax avoidance will have some impact on the parameters of State aid review of national tax measures that result in tax avoidance. The directive adopted under Article 115 of the TFEU which requires unanimity of all member states aims at strengthening anti-tax avoidance rules by tackling both aggressive tax planning and the abuse of law. The draft directive seek to address situations in which MNEs can take advantage of disparities that result from the diversity of domestic tax regimes on the EU level. The directive is based on the 2015 OECD BEPS recommendations and provides for a coordinated approach by Member States in addressing specifically four areas, namely hybrid mismatches, controlled foreign companies, as well as rules on interest limitation and exit taxation. It further provides for a general anti-abuse rule which is intended to cover any gaps that arise in the application of the specific anti-abuse rules applying to arrangements in respect of corporate tax liability that are not genuine. The objective of the directive is to provide coordination amongst Member States to ensure a minimum standard of protection for domestic corporate tax systems and to improve the resilience and functioning of the internal market as a whole against cross-border tax avoidance practices. As these principles will become part of domestic law they will in turn form part of the national reference system from which deviations in respect of State aid law can be assessed. The objective of protecting the integrity of the internal market will mean that in applying tax law, tax administrations or regulators will have to look beyond
19
potential internal inconsistencies that may arise or the abuse of national law but will also have to take account of the internal market and external cohesion in the areas covered by the directive.
Some commentators have suggested that the directive may however not have too great an impact on national tax systems as most Member States have already sought to implement the OECD BEPS recommendations and have anti-abuse measures in place. Instead, it has been suggested that the new directive is likely to simply cause another layer of complexity in determining the applicable law48. Dourado suggests however that the Anti Tax Avoidance Package reflects an acknowledgement by the Commission that there is no international standard in respect of BEPS that can be adhered to, instead the general nature of the OECD’s recommendations have led to the generation of further disparities when States seek to implement them unilaterally49. While the EU is seeking to provide coherence in respect of its implementation of BEPS recommendations, the individual provisions of the directive itself are quite general and may equally allow for further disparities when the Member States seek to implement them. To the extent that national laws may not currently contain such provisions, Member States have until December 2018 to transpose the directive into domestic law or align their current framework with that of the EU.50
While the recent directive does not deal with transfer pricing rules or tax rulings. The directives concerning increased transparency will certainly impact on how MNEs seek to allocate their profits between members of the corporate group. In respect of the recent investigations into tax rulings it is necessary to consider how alongside these new transparency requirements the State aid rules can be utilised to address harmful tax practices that arise as a result of an abuse of transfer pricing and
48 See W. Bongaerts, I. Ijzerman, “EC Anti-Tax-Avoidance Package: Responses from European Tax Practices
[2016] 35(3) ABA Tax Times, found at
http://www.americanbar.org/groups/taxation/publications/abataxtimes_home/16jun/16jun-pp-bongaerts-ijzerman-ec-anti-tax-avoidance-package.html
20
advanced pricing agreements and how these investigations in turn impact on the meaning of ‘selectivity’ within Article 107(1).
Tax rulings in themselves are not considered harmful and indeed, by increasing the legal certainty of MNEs’ in their taxing requirements they can in fact be beneficial to a State by ensuring an attractive business environment. The issue with many rulings is that they lack transparency, again this itself is not per se an issue, however such a lack of oversight can allow certain special benefits to be advanced to MNEs by the national taxing authorities that go under the radar. Such benefits are thus selective as they are not awarded to ordinary tax payers and in that respect they will come within the ambit of Article 107(1).
State aid law as a consequence of the Gibraltar decision may however in itself be shifting towards a broader application of the selectivity test in that it leaves open the possibility of a system that can be checked by external anti-abuse standards and potentially by international best practices particularly with regard to the recent investigations, in the field of transfer pricing rules and methodologies.
CHAPTER 3
3.1 The Gibraltar Case
The Gibraltar case51 is considered a landmark judgment in the development of the selectivity test and has proved quite controversial in the academic52 literature particularly as it appears that the entire Gibraltar corporate tax system was under review despite the general nature of the measures that were introduced. The case perhaps best exemplifies the difficulties in determining the correct reference framework or benchmark when trying to ascertain whether a contested fiscal measure is selective. The reasoning of the Court of Justice, can at times, be quite opaque leading to a degree of
51 Case C-106/09P and C-107/09P, Commission v Government of Gibraltar [2011] ECR I-nyr
52 R.Luja, “(Re)shaping Fiscal State Aid: Selected Recent Cases and Their Impact”, (2012) Intertax 120; J. Lang,
21
uncertainty as to the actual full impact of the decision. On one reading, it potentially allows for the testing of a Member State’s tax regime against external standards. Such a reading would represent a significant departure from the previous case law of the court and would open the way for the Commission to test Member States’ tax regimes for abuse and harmful tax competition.
In 2002, the Government of Gibraltar introduced a tax reform whereby it sought to repeal its entire corporate tax legal regime. Three tax bases were introduced for all companies in the Gibraltar territory which included a payroll tax, a business property occupation tax (BPOT) and a registration fee. Gibraltar notified the Commission of the proposed changes who in turn found that the measures introduced were both regionally and materially selective. They were deemed to be regionally selective as they allowed for a lower tax rate to be implemented in the territory of Gibraltar than that which was applicable in the UK.53The measures were considered to be materially selective as they excluded companies that were not profitable from their application, furthermore, the capping of liability to 15% in respect of the payroll tax and BPOT favoured companies that generated high profits while the payroll tax and the BPOT favoured offshore companies in that they had no liability under the corporate tax system given their lack of physical presence on the territory.
The General Court decided against the Commission both in respect of regional and material selectivity. It found that the measures were not regionally selective as the Gibraltar government was sufficiently autonomous in the implementation of its tax measures54. In respect of material selectivity, the Court reaffirmed the competence of Member States in the area of direct taxation and in determining the structure of their tax systems. Significantly, in applying the test for selectivity it held that the Commission had not fulfilled the first step in the selectivity analysis as it had failed to identify the
53 The UK tax regime, according to the Commission was the applicable reference framework.
54 It found that the UK did not intervene in the determination of the new reforms, that the financial
consequences of the reform would not be off-set by a subsidy or aid by the UK and there was no causal link between any loss of revenue and financial transfers from the UK. These findings by the General Court on regional selectivity were not appealed by the Commission.
22
‘normal’ regime from which a derogation could be ascertained. It was therefore not possible to determine whether the measures at issue could be justified by the nature or general scheme of the system. As a result of this omission, the Commission had not observed the three stage analysis as required and had as a consequence exceeded its powers under the Treaty by assuming the position of the Member State in designing the ‘normal’ regime.55
The Commission appealed the decision on the basis of the General Courts interpretation of material selectivity. The Commission claimed that the requirement to first identify a normal regime under the tax system was erroneous as such an approach would disregard the possibility that a Member State may introduce a tax system which is inherently discriminatory by its very structure and that through a judicious selection of the criteria to be applied in the alleged ‘normal’system of taxation, Gibraltar had produced to a large extent the effects of a scheme which manifestly incorporates State aid for certain categories of undertaking.
Advocate General Jaaskinen, delivered his opinion in line with the position taken by the General Court opining that if the Court of Justice were to follow the Commission’s approach it would require a form of “methodological revolution”56 He found that the Commission was asking the Court to establish a
new concept of an “inherently discriminatory” tax system in its attempt to tackle harmful taxation within the internal market. He found that the contested measure, would classify as selective an advantage that applies to more than 99% of Gibraltar undertakings. This new approach advocated by the Commission, he found, was at variance with the three step derogation test as set out in the Commission Notice of 1998. He stated “given the diversity of tax measures, it is becoming more and more complex to trace a dividing line between general measures and selective measures. Consequently, the determination of the reference framework, difficult though it may be, is
55 Joined cases T-211/04 & T-215/04 Government of Gibraltar and United Kingdom v Commission [2008] ECR
II-3745, para 145 and para 184.
23
fundamental in ascertaining whether the regime in question is ‘abnormal’ and therefore ‘selective’”. He found that to take the position of the Commission would be “to no longer assess an advantage on the basis of a comparison between the measure and the generally applicable regime, but by virtue of a comparison between the tax regime as it exists and another hypothetical and non-existent – system.”
57
On appeal, the Court of Justice largely agreed with the findings of the General Court58. Regarding the alleged material selectivity, it found that the capping of profits and the requirement of generating a profit were general measures. Advantages that arise as a result of the application of a general measure without distinction are not selective, finding that the benefit arose as a “consequence of a random event that the undertaking in question is unprofitable or very profitable.59 The Court however found that the omission of offshore companies from taxation was indeed selective.
It criticized the General Court for failing to observe the ‘effects’ or the substance of the reform, having regard only for the regulatory technique used. This meant that the Court had excluded from the very outset any possibility that the fact that offshore companies that had incurred no tax liability may be classified as obtaining a ‘selective advantage’. It found:
“in order for a tax system to be classifiable as ‘selective’ it must be designed in accordance with a certain regulatory technique; the consequence of this would be that national tax rules fall from the outset outside the scope of control of State aid merely because they were adopted under a different regulatory technique although they produce the same effects in and/or in fact. Those considerations apply particularly with regard to a tax system which, as in the present case, instead of laying down general rules applying to all undertakings from which a derogation is made for certain
57 Ibid, para 98
58 Regional selectivity was not appealed by the Commission and therefore the Court did not pronounce on it. 59 Ïbid, para 83.
24
undertakings, achieves the same result by adjusting and combining the tax rules in such a way that their very application results in a different tax burden for different undertakings.”60
The Court was of the view that offshore companies and those resident in Gibraltar were comparable with regard to the objective of the proposed tax reform, namely to introduce a general system of taxation for all companies established in Gibraltar.61 Given the fact that the narrow tax bases, were limited only to payroll and occupation tax meant that offshore companies were excluded from the outset. The court found that the “criteria forming the basis of assessment which are adopted by a tax system must also, in order to be capable of being recognized as conferring selective advantages, be such as to characterize the recipient undertakings, by virtue of the properties which are specific to them, as a privileged category, thus permitting such a regime to be described as favouring ‘certain’ undertakings or the production of ‘certain’ goods within the meaning of Article 107(1)”.62
The fact that offshore companies were not taxed was seen not as a random consequence of the Gibraltar system but the inevitable consequence of the fact that the bases of assessment were specifically designed so that offshore companies would not be liable to tax.
3.2 Analysis
The Court of Justice identifies in the judgment what it perceives to be the tax regime or reference framework as being a general system of taxation for all companies in Gibraltar. In many ways the terminology of the Court follows that of the previous case law, whereby a regime is determined as a first step, and on that basis, the reform introduced can be assessed for any discriminatory elements which have the effect of favouring certain undertakings in a similar legal and factual situation in light
60 Ibid, supra note 13, 91 - 93 61 Ibid, supra note 13, para 101 62 Ibid, para. 104
25
of the object pursued. However, given that the reference regime was, according to the General Court not identified by the Commission it is unclear how the Court of Justice was able to conclude that the relevant reference framework was the one identified by the Court.
In light of the system established by Gibraltar under the reform there does not appear to be a reference system beyond the three tax bases introduced. Therefore no possibility of finding a derogation due to the exclusion of off-shore companies. The tax bases as introduced were in accordance with the stated objectives of Gibraltar which was to simplify the tax system by introducing more manageable tax bases such as payroll and BPOT which reflected the small territory of Gibraltar and its limited administrative resources. The objective of Gibraltar was clearly not the introduction of a general system of taxation for all companies, as found by the Court.
There can therefore be two alternative interpretations of the judgment. The Court arrived at the conclusion that the reference system as designed was in fact the corporate tax system, and this system was deliberately reduced in scope to exclude off-shore companies. By emphasizing the ‘effects’ based analysis the court could find a derogation even when the regulatory technique used had obviated this by “adjusting and combining the tax rules.” The Gibraltar tax regime has already been subject to review and had previously been found to be in contravention of the State aid principles. The Court may have deemed that the new system was simply a perpetuation of the old and was thus clearly a means of abusing a regulatory technique. Lang suggests that perhaps the Courts reasoning was based on a ‘finding of fact’63, namely, that the reference system that Gibraltar had initially intended to introduce was a comprehensive tax regime and then this was deliberately narrowed to exclude certain companies from the tax base. In fact, it is questionable, had Gibraltar simply introduced either a payroll tax or a BPOT on their own whether the Court would have jumped to a finding of a corporate tax regime.
26
Some authors suggest that the judgment allows for the dispensing of the first step of analysis. Prek and Lefevre find that ‘the standard approach is not the only one, and that the fact that a given situation does not fit in it, should not lead to the conclusion that the tax measure at stake is a general measure, but rather that the benchmark against which the existence of a differential treatment must be examined is solely composed of the circle of undertakings that are in a comparable legal and factual situation in view of the objective of the measure at stake.”64
The question of course is who determines the objective of the measure. This is normally the domain of the Member State and should not be reviewable. The Gibraltar case has given rise to whether even the objective of the measure is reviewable by some sort of external standard.
Such an interpretation suggests that the Court did indeed substitute the reference system as determined by Gibraltar, with what it deemed to be the appropriate standard, a hypothetical ‘corporate tax regime’. This standard would have as its objective the taxation of all undertakings in Gibraltar including off-shore companies. The fact that the off-shore companies were not subject to any tax due to the ‘regulatory technique’ used meant that they benefited from a selective advantage or, within the terminology used by the Court, they were ‘privileged’. On this reading the Court is essentially introducing a hypothetical external reference framework. Ismer and Piotrowski have found that Gibraltar does potentially represent such a departure and finds that the determination of the objective of the tax system by the Court is necessary in such cases, as otherwise State aid law could be easily circumvented by a mere redefinition of the objective of the tax system. They believe that there are two aspects to the test for selectivity, internal consistency to ensure that measures are in accordance with the logic of the national tax regime and as a result of Gibraltar, an external
64M. Prek, S. Lefevre, “The Requirement of Selectivity in the Recent Case-Law of the Court of Justice”, [2012] 2
27
consistency test. The latter would be applied in cases of de facto selectivity when there is no reference system that can be relied upon. They believe that such a view is justified, as they find that tax systems should not only be internally consistent (as is traditionally required under Article 107), but that they should also be externally consistent. They find that such a reading would allow for external standards including the use of the ‘arm’s length principle’ in matters such as tax rulings or more broadly allow the Commission to test measures for harmful tax competition.65
Such a broad interpretation has however been criticized, notably Lang finds it would be “a striking example of judicial legislation if the Court in State aid tax cases were to compare national measures with hypothetical tax measures visualised only by the Court itself”. 66 The broader interpretation however would allow for the review of tax measures if they indicated that they were allowing for abuse of international taxation disparities or harmful tax practices against a hypothetical best practice such as the OECD Guidelines.67
Chapter 4 Tax Rulings
4.1 State Aid and Tax Rulings
65 R. Ismer, S. Piotrowski, “The Selectivity of Tax Measures: A Tale of Two Consistencies” [2015] 43 Intertax,
559, p.567-569
66 J. Lang, ‘The Gibraltar State Aid and Taxation Judgment – A “Methodological Revolution”?’ (2012) 4 EStAL
805, at p. 812 (See generally for a broader critique of the findings of the Court, he finds that the better reading of the judgment is that the Court made a finding of fact, namely what the reference system was, that Gibraltar had initially intended to introduce a comprehensive tax regime and then deliberately narrowed its base.)
67The Paint Graphos67 case is interesting as while it did not find the measure in question to be State aid, it
suggested that State aid law may require accompanying anti-abuse measures. The case concerned the possible abuse of a provision which treated cooperatives as transparent, meaning that only the participants would be taxed without taxing at the level of the entity itself. Alternative interpretations of the legal form by different taxing jurisdictions could allow for tax avoidance by corporations. The Court allowed for different treatment, finding that corporations and cooperatives were not comparable in light of their economic activities. However it found that Member states should ensure that appropriate laws are in place to prevent abuse of the
provision, thus suggesting that State aid law mandates some form of anti-abuse measure. C-78/08 to C-80/08 Minestero dell’Economica e delle Finanze e.a. V Paint Graphos e.a. [2011] ECR n.y.r
28
The recent investigations into the tax rulings of Member States are key in determining the Commission’s interpretation of the scope of the ‘selectivity’ test and how it can be applied to harmful tax competition within the Union.
Under traditional State aid principles individual rulings can come within the remit of Article 107(1) if they deviate from the practices and rules of the national tax regime. This can be either through the provision of rulings only for specific undertakings when the tax regime generally does not provide for them68 or it can arise when the tax authorities have discretion in how they apply the tax rules. In the Kimberly Clark Sopalin69 case it was held that the treating of taxpayers on a discretionary basis may
mean that the individual application of a general measure takes on the features of a selective measure, particularly, where the exercise of the discretionary power goes beyond the simple management of tax revenue by reference to objective criteria.
“Measures which prima facie apply to all undertakings, but are (or may be) limited by the discretionary power of the administration, are selective. This is the case where meeting the given criteria does not automatically result in an entitlement to the measure.70
The exercise of discretion however will not necessarily lead to a finding of selectivity, in particular where the margin of discretion is transparent and non-discriminatory or where the degree of latitude is limited by objective criteria not unrelated to the tax system.71 In the area of tax rulings, when an
68 For example the tax rulings may be selective as they allow for ‘certainty’ in how intergroup transactions will
be interpreted by the tax authority. This is not normally available to SMEs. In the Report on the Implementation of the Commission Notice on the application of the State aid rules to measures relating to direct business taxation, February 2, 2004, C(2004)434 p. 9, the Commission found that “measures open to all can, none the less, be regarded as selective where the eligibility criteria in practice restricts the potential number of recipients. This is the case with measures that apply only to multinational companies or large companies”.
69 Case C-241/94 France v Commission [1996] ECR I-4551 para 23 and 24.
70 Draft Commission Notice on the Notion of State Aid pursuant to Article 107(1), p 45, to be found at
http://ec.europa.eu/competition/consultations/2014_state_aid_notion/draft_guidance_en.pdf, para 124.
71 Case C-6/12, P Oy [2013] ECR I-0000, paras 26–27 where the objective of the avoidance in trading in losses is
29
administration or tax authority uses its discretion to issue a ruling to confirm the taxable basis or income based on a calculation method that deviates from the normal practice of the system it will result in State aid.
Tax rulings first came before the Court of Justice as an issue in the Belgium and Forum 187 case72 which concerned Belgian legislation that allowed for a flat rate tax of 8% on the basis of operating costs [minus staff costs] to be imposed on Coordination centres. In order to qualify the Coordination centres had to be part of a multinational group in at least four countries having inter alia a turnover of BEF 10 billion. The regime had originally been found to be in compliance with State aid but in 2003 the Commission again investigated and found the measure was selective as it applied only to large groups and excluded certain costs. The decision is important as it lays the basis for the findings of selectivity in the recent tax ruling investigations and is referred to in the recent Draft Notice on the Notion of State aid 201473. The Court held in the case that rulings which allow certain taxpayers to use alternative methods for calculating taxable profits compared to other similar taxpayers may involve State aid. The Commission states in its recent Draft notice that, “rulings allowing taxpayers to use alternative methods for calculating taxable profits, eg. the use of fixed margins for a cost-plus or resale-minus method for determining an appropriate transfer pricing, may involve State aid”. It further provides that “tax rulings should only aim to provide legal certainty to the fiscal treatment of certain transactions and should not have the effect of granting the undertakings concerned lower taxation than other undertakings in a similar legal and factual situation (but which were not granted such rulings)”.74
The Notice states that the following advance administrative rulings involve selectivity when:
72 C-182/03 & C-217/03 Belgium and Forum 187 v Commission [2006] ECR I-5479.
73 Draft Commission Notice on the Notion of State Aid pursuant to Article 107(1), p 45, to be found at
http://ec.europa.eu/competition/consultations/2014_state_aid_notion/draft_guidance_en.pdf.
30
The tax authorites have discretion in granting administrative rulings;
The rulings are not available to undertakings in a similar legal and factual situation;
The administration appears to apply a more “favourable” discretionary tax treatment compared with other taxpayers in a similar factual and legal situation;
The ruling has been issued in contradiction to the applicable tax provisions and has resulted in a lower tax.75
The latter point is significant, as in order for State aid to arise it must depart from national tax provisions or practices. As Luja states “whether a tax benefit is present should be determined by comparison to the normal tax system in a Member State. It is the national tax system that is the only relevant benchmark and not the tax system of another Member State or any desirable/best practices as defined by international organisations. Determining the proper benchmark for analysis is the first step in tackling tax benefits via the State aid system.”76
The recent investigations by the Commission however into tax rulings indicated an alternative approach. Considerable emphasis was placed on the arm’s length principle and the OECD guidelines as a reference point in the Commission’s reasoning in determining how the relevant tax authorities should implement tax rulings. The use of these criteria as a benchmark with which to test national tax administrations was highly questionable as the guidelines themselves are not binding and in some instances were not part of national law, the suggestion therefore was that the Commission was willing to use international best practice or an external standard, in their review of Member States’ tax regimes.
75 Ibid, para 177.
76 R. Luja, EU State Aid Law and National Tax Rulings 2015; to be found at
31
4.2 Current Investigations
Formal investigations commenced by the European Commission in 2014 in respect of prominent multinationals have resulted in final decisions with respect to the advancement of unlawful State aid in respect of Fiat, Starbucks and the Belgium Excess Profits Scheme. Final decisions are still pending in respect of Amazon and Apple however preliminary findings also indicated the presence of aid. Broadly speaking the State aid investigations referred to intra-group transfer pricing agreements and the individual tax rulings which approved them by the various taxing authorities. The Belgium Excess Profits Scheme being the exception, as it concerned the review of the entire scheme itself rather than individual rulings. The review concerned what the Commission perceived to be a deviation in the issuing of the tax rulings in the calculation of taxable profits in respect of Apple and the taxable basis relating to services such as royalties, finance or manufacturing activities of the remaining MNEs.
The Commission appeared to follow the same analyses in each case whereby it determined whether the ruling under investigation allowed the MNE to depart from normal market conditions in setting the commercial conditions for intra-group transfers. The transfer price to be used is compared with that which would be paid by an independent prudent operator. A failure to comply with this was viewed as infringing the ‘arm’s length principle’ in that it allowed for the lowering of the tax due and thus a finding of the advancement of State aid on the part of the taxing administrator.
In the Apple case the Commission, relying on the Belgium and Forum 187 case found:
“In order to determine whether a method of assessment of the taxable income of an undertaking gives rise to an advantage, it is necessary to compare that method to the ordinary tax system, based on the difference between the profits and losses of an undertaking carrying on its activities under normal market conditions. Thus, where a ruling concerns transfer pricing agreements
