European Court finds that Spanish tax rules were not unlawful state aid because they did not give a “selective advantage”
SUMMARY In two recent cases on fiscal state aid, the General Court of the European Union found that certain aspects of the Spanish tax system did not constitute “state aid” within the meaning of the Treaty on the Functioning of the European Union. The General Court found that the European Commission had failed to show that the fiscal measures at issue gave rise to a “selective advantage” for a particular category of undertakings. Absent this element of selectivity, the particular Spanish tax rules did not constitute unlawful state aid, and the Commission had erred in characterising them as state aid. Alleged state aid in relation to tax law and practice has hit the headlines several times in 2014, most notably in relation to advance tax rulings and transfer pricing. The Commission has opened investigations into advance tax rulings/advance pricing agreements obtained by Apple in Ireland; Starbucks in the Netherlands; and Fiat Finance and Trade and Amazon in Luxembourg. There is a natural tension between advantageous features of a tax system that have eligibility criteria and the EU state aid rules, which prohibit EU Member States from giving advantages financed by State resources that could distort competition. The recent cases on the Spanish regime demonstrate that whether fiscal measures are “selective” is likely to be the key question in state aid investigations into tax measures. Moreover, the two cases show that it may be difficult for the Commission to establish the “selectivity” of fiscal measures that apply generally, rather than to specific goods or undertakings.
WHAT IS STATE AID? Article 107(1) of the Treaty on the Functioning of the European Union (“TFEU”) provides that “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 [emphasis added] shall, in so far as it affects trade between Member States, be incompatible with the internal market”. Four elements must be present for a measure to constitute state aid: 1. An advantage for the beneficiary of the measure; 2. “Selectivity” in granting that advantage; 3. Funding by the Member State or through State resources; and 4. An effect on trade between Member States and a distortion or threatened distortion of competition. State aid can be justified in certain circumstances, e.g. to remedy damage caused by natural disasters1 , and may be justified by certain policy objectives, e.g. the economic development of deprived areas2 . Therefore, measures satisfying the four criteria above can be lawful in certain circumstances, by falling within one of the various exemptions, or by application to the European Commission (the “Commission”) for clearance in advance of implementing the measure. There is also a de minimis carve-out for any measures that would result in aid with a value of less than €200,000 over three fiscal years. From a fiscal perspective, the state aid rules are capable of catching any de facto or de jure favourable tax treatment granted by a Member State, such as tax rulings or exemptions from certain laws, where such treatment is given selectively to a particular taxpayer or category of undertakings or in respect of certain goods. If the Commission finds that a measure constitutes unlawful state aid, it must, in principle, require the relevant Member State to recover the aid, with interest, from the beneficiary (unless such recovery would be contrary to a general principle of EU law). The Commission can make such a determination up to 10 years after the aid was granted. BACKGROUND TO THE CASES The Commission took issue with a provision of Spanish tax law (the “Contested Measure”) that allowed a company subject to tax in Spain to amortise the share value representing the underlying goodwill on acquiring a shareholding of at least 5%, held for a minimum of 12 months, in a nonSpanish company. The non-Spanish investee company also had to be subject to a similar tax to the corporate income tax in Spain. Spanish companies could not amortise the value representing the underlying goodwill in respect of shareholdings in other Spanish companies, although this value could be amortised in relation to certain Spanish group transactions. In decisions in October 20093 and January 20114 , the Commission found that the Contested Measure constituted unlawful state aid under Article 107(1) TFEU on the basis that it gave an advantage to Spanish companies investing in non-Spanish entities over Spanish companies investing in Spanish entities. Autogrill España SA challenged the 2009 decision, and Banco Santander SA and Santusa Holding SL challenged the 2011 decision. THE GENERAL COURT’S JUDGMENTS In judgments issued on 7 November 2014, in near-identical terms in respect of both appeals, the General Court found that the Commission had failed to establish that the Contested Measure constituted state aid5 . Accordingly, the General Court annulled the relevant parts of the two decisions. The judgments focussed in particular on whether the Contested Measure was “selective” for state aid purposes. If selectivity of a fiscal measure is established, it is relatively easy to demonstrate the presence of the other criteria set out in Article 107(1) TFEU. The burden of proof is on the Commission to identify a specific class of persons favoured by a particular measure. The General Court held that, in order to establish “selectivity”, it is not enough to look at a measure that is potentially available to all undertakings within a taxing jurisdiction (e.g. a tax exemption) and find it to be selective simply because certain persons will fall within the terms of the measure while other persons fail to qualify for the benefit of the measure6 : the fact that reliefs and exemptions are subject to eligibility conditions is an inherent feature of a taxing system, and will not per se amount to selectivity. In order to evaluate selectivity in respect of a certain group of undertakings, a comparison needs to be made which identifies those to whom a beneficial measure applies, and other persons in a similar legal and factual situation who would be treated differently7 . For example, a measure only benefiting companies in a particular geographical area could be selective8 . The Commission argued that the Contested Measure was selective because it applied different treatment to Spanish companies investing offshore than to Spanish companies investing onshore. The Commission also argued that the Contested Measure was designed to give preferential treatment to Spanish companies investing in foreign entities over entities established in other Member States investing outside their home jurisdiction. The General Court did not agree that these factors were sufficient to demonstrate the selectivity of the Contested Measure. The mere fact that the Contested Measure had eligibility criteria did not mean it was “selective” for state aid purposes9 . The Commission had to show that an advantage was given to a particular group of companies or a particular group of products within a Member State as required by Article 107(1) TFEU. Moreover, while the distortion of competition and trade between Member States is relevant to Article 107(1) TFEU, it is not relevant to the criterion of selectivity, which has to be assessed by comparing undertakings within the particular Member State10. The distortion of trade condition is logically separate from the selectivity condition. The Contested Measure did not apply to a particular category of undertakings or a particular business sector, since it was prima facie available to any Spanish company that held the requisite percentage shareholding in a non-Spanish entity for the requisite period of time. Therefore, although the Contested Measure was limited to a category of economic operations (namely, holding shares in a foreign entity), it was not “selective” within the meaning of Article 107(1) TFEU. The General Court noted that a measure that applies independently of the nature of a business carried on, or invested in, is not selective11. The General Court held that acquiring shares in a foreign entity was a “purely financial” operation, independent of the nature of the business of the acquiror or the investee company12 . The General Court highlighted the need to carefully evaluate selectivity in terms of an advantage granted to a particular group of goods or undertakings. The Commission’s analysis did not meet the required legal standard in the view of the General Court. If the General Court had found that the Contested Measure was state aid, this could have set a precedent that any tax relief in the EU whose application is subject to specified conditions is capable of being unlawful state aid13 . COMMENT Establishing a “selective advantage” is key in demonstrating that a particular fiscal measure (or other type of measure) constitutes state aid. Identifying the correct comparators in deciding whether a particular group or person is advantaged is a crucial part of the analysis. Autogrill and Santander show that establishing a selective advantage requires the Commission to demonstrate that a fiscal measure applies to a particular category of undertakings, that is, to a specific identifiable group of undertakings, and gives them an advantage that is not available to others in a comparable legal and factual position. Where fiscal measures apply generally, or to a wide variety of
economic transactions rather than to a narrower category of undertakings, they should not constitute state aid since they do not grant a “selective” advantage within the meaning of Article 107(1) TFEU. For example, there should be no basis for a state aid challenge if a Member State generally imposes a low rate of tax (e.g. Ireland’s 12.5% corporate tax rate on trading income). Similarly, a participation exemption for capital gains and dividends in respect of significant percentage shareholdings should not constitute unlawful state aid unless accessing that exemption is subject to material financial thresholds or is conditional upon the taxpayer carrying on certain types of activity. It is likely that the Commission will appeal the Autogrill and Santander judgments to the European Court of Justice. A ruling by the higher court would not be expected in fewer than 18 months from any such appeal. With the Commission also progressing its investigations into the alleged state aid arising from the advance pricing agreements of Apple, Starbucks, Fiat Finance and Trade, and Amazon, the fiscal state aid debate will continue to occupy headlines and Commission resources. It will be interesting to see how the principle highlighted in Autogrill and Santander plays out in the context of the Commission’s investigations into tax rulings/advance pricing agreements entered into by tax authorities. The President of the Commission, Jean-Claude Juncker, recently announced the development of a directive for the automatic exchange of information on national tax rulings in order to increase transparency across the EU. Such a directive would be aligned with the aims of the OECD’s BEPS (base erosion and profit-shifting) project. Exchange of information on tax rulings is being discussed as part of action item 5 of the BEPS project, which proposes ways to combat “harmful tax practices”. The Commission’s position (see, for example, its recently published decision in respect of Starbucks14) is that a “selective advantage” is given where a Member State agrees a transfer pricing methodology with a taxpayer which is difficult (in the Commission’s view) to reconcile with the OECD-endorsed arm’s-length transfer pricing principle (see the 2010 OECD Transfer Pricing Guidelines15 and, more recently, the September 2014 draft OECD publication regarding transfer pricing and intangibles16). In some cases, the Commission may justifiably consider there to be state aid, e.g. if a Member State mandates a particular transfer pricing method in relation to operations carried on by a specified group of taxpayers (such as the modified “cost plus” method formerly applied to Belgian Co-ordination Centres). However, the point is often less clear. The Commission may arguably be in danger of overlooking the fact that agreeing transfer prices is not a precise science. As the General Court pointed out in Autogrill and Santander, it would be perverse if the application of any tax measure that is subject to certain conditions could lead to it being classed as state aid. Any fiscal system has particular rules (e.g. those allowing deductions from taxable income for interest payments), and those rules are likely to contain pre-conditions. Moreover, it would be an erosion of the fundamental right of Member States to fiscal sovereignty if Article 107(1) TFEU allowed the
14 Available at http://ec.europa.eu/competition/elojade/isef/case_details.cfm?proc_code=3_SA_38374 15 Available at http://www.oecd.org/ctp/transfer-pricing/transfer-pricing-guidelines.htm 16 Available at http://www.oecd.org/tax/guidance-on-transfer-pricing-aspects-of-intangibles- 9789264219212-en.htm-6- EU State Aid and Tax Law 2 December 2014 Commission a free hand to intervene in the application of national tax laws. It is encouraging for taxpayers and Member States that the General Court has recognised this risk and has provided welcome clarification. However, the concept of a “selective advantage” remains malleable despite the two judgments. The Commission’s ongoing fiscal state aid investigations show its willingness to develop this concept, not just in relation to black letter tax rules but also in relation to how Member States’ tax authorities exercise their administrative powers when giving advance rulings or settling tax disputes. This greater activism needs to be borne in mind by taxpayers when dealing with tax authorities in EU Member States.