Do you know we also publish a journal on State aid?
The European State Aid Law Quarterly is available online and in print, and our subscribers benefit from a reduced price for our events.
This question, among others that were also related to the ongoing investigations, monopolized the lively third session, led by Ms Julia Rapp from the European Commission. At the beginning of the session Ms Rapp examined the 1998 Fiscal State aid Notice as well as past cases on tax schemes favouring multinationals from the 2000’s, e.g. the cases on coordination and offshore centres. The focus then shifted to the five ongoing investigations, the timing of those investigations and the effective tax rates that companies pay in various Member States. The Commission’s legal approach to the ongoing cases was set out on the basis of the reasoning in the opening decisions. More specifically, it was explained that the Commission considers the reference framework to be the corporate tax system, where taxable profit is calculated on the basis of the difference between the company’s income and charges, and that this was confirmed by Paint Graphos. It was believed that tax administrations could confer an advantage on a company obtaining a tax ruling if they did not ensure that transfer pricing arrangements within groups resembled impartial terms of companies transacting on the market, and that this was confirmed by Forum 187. The Starbucks opening decision was used as a case study to better illustrate how the Commission’s approach would apply in practice. Finally, the session concluded with a discussion of the European Parliament’s TAXE Report and of the thorny issue of aid recovery, in relation to the legitimate expectations and legal certainty defences to recovery.
The fourth session, led by Professor Douma, began with a presentation of his views on administrative discretion and its significance in the ongoing investigations. He considered that the possibility of judicial review of tax authorities’ decisions on individual tax rulings was crucial in assessing the rulings’ selective character. Professor Douma then placed State aid rules within the general picture of EU law and the division of competences between the EU and Member States. In this context, three separate strands of criticism were highlighted. Firstly, some would say that there is currently too much respect for tax sovereignty, others that there is too much respect towards the State aid prohibition, and many that the ECJ’s State aid assessment model is unclear. His suggestion for finding a way out of the maze lied in understanding Robert Alexy’s ‘Theorie der Grundrechte’, according to which rules are to be distinguished from principles and principles are, in essence, optimization requirements. In Professor Douma’s opinion, the picture becomes clearer once we perceive both tax sovereignty and the State aid prohibition as principles that have to be adjusted to co-exist with each other.
The seminar’s first day ended with a presentation by Mr Christian Ahlborn, from Linklaters law firm. This session focused on three Spanish cases, namely the Spanish Internet Gambling, Goodwill and Tax Lease cases. After providing the factual background for all cases, Mr Ahlborn delved into a very detailed, step-by-step selectivity analysis for each. It is worth noting that the unique constitutional arrangements of Spain, especially in relation to the fiscal prerogatives of its autonomous regions, added an extra layer of complexity. As regards the gambling case, the crucial question was whether online and land-based gambling operators were in a comparable legal and factual situation, to which the Commission replied in the affirmative. As regards the Spanish Goodwill cases, the Commission’s mistake, according to the General Court, was that it inferred selectivity only from a derogation even though it could not identify either the favoured or the excluded undertakings. Finally, in relation to the tax lease case, it consisted of a combination of various measures that ended up favouring the shipbuilding sector and, in the Commission’s view, amounted to illegal State aid. One of the many complexities of this case was identifying the direct and indirect beneficiaries, and Mr Ahlborn elaborated on this in order to answer certain questions from the audience. Overall, he concluded that we are lately viewing a battle between the Commission and the General Court in State aid cases as regards the selectivity condition, which will only give us a clear winner once these cases reach the ECJ.
The seminar’s second day started off with a summary of the points raised and the debates that took place on the first day. Afterwards, Professor Douma opened his session on environmental taxes and State aid. He started off by citing several paragraphs of the Draft Notice on the Notion of State aid, whose formal adoption is long-awaited by practitioners and academics alike. For instance, he indicated that certain paragraphs of the Draft Notice seem to contradict each other and also conflict with the Fiscal Aid Notice of 1998, citing as an example the contrast between paragraph 135 of the former and 13 of the latter Notice on the relevance of environmental objectives to the selectivity analysis. The session then became more interactive, with the speaker using a hypothetical example to spur debate: is a measure providing a beneficial rate for income resulting from the production of wind energy selective? Lively discussion ensued, with the vast majority of participants agreeing that it would constitute State aid for many reasons, e.g. because it only applied to renewable energy, and, moreover, only one part of renewable energy. After an in-depth discussion of the NOx and British Aggregates cases, Professor Douma ended the session by reminding us how environmental taxes can be used as both a carrot and a stick in gearing companies towards or away certain types of behaviour.
The second session of the day, led by Professor Nicolaides, dealt with tax measures imposed by sub-national authorities. It was submitted that the crucial concept here is that of regional selectivity, and that the decisive step in all regional selectivity cases is to decide what the reference framework is. If it is the national tax system and rate, then the measure will be found to be regionally selective. The key case in this area, namely the Azores case, was discussed extensively, especially in relation to the three-step test that it established (political, procedural and economic autonomy). The refinement of this test by subsequent cases was also commented upon (e.g. Case C-428/06). There was also analysis of more recent examples of regional selectivity related to measures adopted by other sub-national bodies like regional airports. For this purpose, the Lübeck airport case (T-461/12) was reviewed in more detail. In accordance with German federal law, German airports have power and jurisdiction to set landing charges. For this reason, the General Court held that airlines landing on other airports were not in comparable situation and that the reference framework was that of Lübeck airport. Thus, if every plane landing on it was taxed equally, no selectivity arose – the underlying legal system of Germany proved decisive.
Professor Nicolaides then proceeded to the third session of the day, where he tackled the topic of parafiscal charges and the hypothecation of tax measures. At first he explained the difference between parafiscal charges and excise duties: tax revenue on the former is raised to support the taxed industry itself, while with excise duties the state uses the revenue as it wishes, their effect being mostly deterrent. A major complication in this area arises when a tax measure is hypothecated to a State aid measure. Hypothecation means that the method of financing of the aid measure forms an integral part of the State aid measure. In such cases, the Member State must notify both the aid and the tax measure that finances it, thus leading to a counterintuitive result: the application of Article 107 TFEU to taxes per se. A further complication in this area relates to the second State aid condition, i.e. the use of state resources. Some parafiscal charges are levied by industry or professional associations, thus making their classification as state resources dubious. Through a series of Commission Decisions and Court cases Professor Nicolaides shed light on this problem and helped the participants understand how the law applies by analyzing various scenarios. The remaining part of the session was devoted to the issue of compatibility and specific cases on hypothecation, e.g. the UK milk levy and Nazairdis cases.
The penultimate session of the seminar, led by Mr Péter Staviczky, focused on the application of State aid rules to tax measures in the Hungarian practice. Fiscal State aid was one of the main tools for attracting foreign direct investment in the 90’s (due to budgetary reasons) and was then also a major obstacle to Hungary’s accession to the EU. Thus, Hungarian fiscal aid cases abound and four of them were used to demonstrate how they were caught by Article 107(1) TFEU. The first case concerned Hungarian tax deductions for intra group interest, where only half of the net interest revenue (received from affiliates and paid to affiliates) formed part of the corporate tax base. The Commission maintained that the measure led to a selective advantage in cross border situations for net interest receiving companies and adopted a negative decision. The second case was rare in its factual basis. It concerned the Hungarian tax donation system, which applied to cinema, sport and performing art donations. Donors received automatic tax exemptions for donating money in support of such causes and the donated amount was deducted both from the tax base and their tax payable. The Commission opened an in-depth investigation to ascertain the measure’s State aid character and its findings were interesting. On the donors’ level no State aid existed: the measure was open to all donors without any limitations and was thus not selective. However, for the beneficiaries, e.g. theatres and sports clubs, the measure was selective because the tax advantages to the donors only accrued if the beneficiaries belonged to the cinema, sport and performing art sectors. Despite the fact that the aid was found to be compatible with the internal market, this case is a good example that a measure can be both general and selective when looked at through different lenses. Finally, Mr Staviczky discussed two environment-related cases, namely the product‐levy case and the case of the excise duty reduction for gas, which perfectly complemented the first session of the day by Professor Douma. This session ended with practical advice on State aid notifications.
Professor Nicolaides concluded the two-day seminar by summarizing the main points of consensus and contention of the two preceding days. He then opened the floor to participants for the final problem solving session where specific “real-life” questions were raised. To briefly mention two, one question was raised in relation to VAT and State aid. Another question was about the State aid compatibility of prospective car taxes in Denmark. An interesting discussion on the specificities of the proposed car tax ensued, with the participants examining whether the criteria of Article 107 TFEU were fulfilled.
Overall, the seminar met participants’ expectations and improved understanding of the peculiarities of fiscal State aid. A general wish was expressed that a similar event be organized once the ongoing Commission investigations into tax rulings are concluded and their outcome is published.
*Dimitrios Kyriazis is a Doctoral Researcher and Tutor in Law at the University of Oxford.