Favourable tax treatment to alleviate “structural disadvantages” suffered by certain companies is a selective measure that falls within the scope of Article 107(1).
Member States use taxes not just to raise revenue but also as instruments of public policy. They impose taxes on activities they want to discourage [e.g. smoking, driving] and they relieve from taxes activities they want to encourage [e.g. regional investment]. There is nothing intrinsically wrong with the use of taxation to achieve wider public policy objectives. In fact, most economists would argue that taxation is one of the most efficient policy instruments because it gives the right incentives to market operators. However, the moment a tax becomes an instrument that pursues a purpose other than raising revenue, it is very likely that State aid rules come into play. Then Member States often forget that the social benefits to be had from achieving targets which are extraneous to the tax system cannot justify exemption from the prohibition of State aid.
This article reviews Commission decision 2015/2817 concerning the construction and operation of a container terminal at the port of Piraeus in Greece. The decision deals with several tax measures for the benefit of the company Piraeus Container Terminal [PCT] which is a subsidiary of the Chinese company COSCO, one of the largest transport companies in the world.
The decision is important because the Commission examines in considerable detail each of the tax measures and explains at length why all of them are selective. The Commission rebuts numerous Greek claims that the measures are general in nature, goes on to find that they contain State aid and concludes that the aid is incompatible with the internal market and, therefore, has to be recovered.
Given the fact that tax measures are now under intense scrutiny by the Commission, the article focuses on the reasons that led the Commission to conclude that the PCT obtained a selective advantage.
The tax measures
The Commission examined several measures, including variations of the main tax exemptions which fell into the following six categories:
- Exemption from income tax on interest accrued during the period of the construction of the terminal.
- VAT credit refund.
- Loss carry-forward.
- Choice of depreciation method.
- Exemption from stamp duty.
- Protection of foreign direct investments.
The Greek government relied on a series of arguments to claim that the measures were either general or merely addressed certain structural, economic or financial disadvantages suffered by infrastructure developers and managers. In relation to the latter argument, Greece contented that it was justified to take measures in favour of infrastructure developers because they were not in the same situation as other companies and, therefore, the Commission could not claim that the measures were discriminatory. Since developers were in a different situation they had to be treated differently. The Commission rejected all of the arguments either on the grounds that they were “extrinsic” rather than “intrinsic” to the logic of the tax system or that such disadvantages, even if true, affected other sectors as well, not just infrastructure development.
Extrinsic v intrinsic objectives
There is much confusion and misunderstanding of what is extrinsic or intrinsic to a tax system, so it is better to explain this at the outset.
Tax systems have basically the same primary objective: to raise revenue according to the ability to pay. The tax base is normally defined in terms of income or profit. The burden of contributing to the financing of state activities and the provision of public goods is normally shared according to the ability to pay. That is why tax systems are also progressive. The rate of tax increases as income or profit increases. These are the basic intrinsic features of tax systems.
Policy objectives such as regional development, promotion of research or protection of the environment may be very worthy but they are extrinsic to the tax system. Governments that levy lower tax rates on activities linked to regional, research or environmental investment, may have very good reasons why they do so but they cannot credibly argue that they aim to raise revenue by lowering taxes [they may be able, of course, to raise overall revenue by lowering overall rates, but in principle this possibility does not justify partial tax reductions. Partial tax reductions, like price discrimination, may increase overall revenue only if the reductions are granted to activities which are sensitive to the tax rate (they have high tax elasticity) and only if there is no substitution between activities subject to differential taxation]. If lower tax rates apply only to certain regions or activities, they will invariably be found to be selective. An exception to this principle that was introduced last year in the case law is when a deviation from the normal tax is open to all companies which are liable to pay the normal rate of tax. In this case the objective of encouraging research or environmental investment is still extrinsic to the tax system but the deviation is of general nature, rather than of selective application to some sector or activity.
It is also worth pointing out at this stage that the Commission observed in its decision that there are only two situations where state intervention does not confer an advantage:
i) when the intervention is in compliance with Altmark criteria in the case of services of general economic interest and
ii) when the state acts as a private investor. [Paragraph 83]
Three-stage test of selectivity
Whether a tax measure contains State aid hinges in most cases on whether the measure is selective. The Commission began its analysis of the existence of selectivity by reiterating the three-stage test: i) identification of the normal or reference tax system, ii) determination of whether the measure in question is a derogation or exemption from the normal system [by differentiating between undertakings which are in comparable situations], iii) assessment of whether the derogation or exemption can be justified by the logic or intrinsic objectives of the tax system [paragraphs 88-89]. For example, it is intrinsic to the aim of a tax system which penalises polluting activities to exempt non-polluting activities.
The Commission applied this three-stage test to each of the six categories of measures. At several points in its decision, the Commission was at pains to explain that if extrinsic objectives were allowed to determine the reference system, then every policy objective or every special treatment of an industry or sector would become the reference system [e.g. paragraphs 93, 153 or 176]. The consequence would be that special tax measures would never be considered to constitute State aid. This is the main reason why the Commission rejected the various Greek arguments that infrastructure projects had special characteristics that set them apart. Their special characteristics could justify the granting of State aid but could not justify their classification as general measures. It is obvious that it is nonsensical to claim that a sector-specific measure is a general measure, given that it is in fact a deviation from the normal tax that applies to all the other sectors and activities of the Greek economy.
At several points and in relation to several of the tax measures, the Greek authorities advanced the argument that infrastructure developers or operators suffered from a structural disadvantage [e.g. paragraphs 106-108]. Such disadvantages can be the large sums required, the duration of construction, the long recoupment period or the high level of risk. The Commission noted that the fact that all infrastructure developers could benefit from the tax measures, not just the PCT, was not enough to prove that the measures in question were general. They were selective because they derogated from the normal tax system.
Although the Commission was correct in its analysis, it should have elaborated that there is a more fundamental reason why the view that remedying structural disadvantages as a general measure is wrong. Large up front investment, long recoupment period or high level of risk are characteristics that can be found in other sectors or economic activities. For example, recoupment of investment in a new medicine may take many years. Research in new technologies may be very risky. Apart from the fact that the objective of remedying these disadvantages is extrinsic to the tax system, if indeed it were the aim of the Greek authorities to address these disadvantages, they should have widened the scope of the special tax measures to cover many other sectors and activities. The fact that they had not done it was strong evidence that the measures were not part of a well-thought out policy but ex post justification.
The Greek authorities also pointed out to several past Commission decisions which appeared to lend credence to their position that the special tax measures for the support of large infrastructure projects did not involve State aid.
The Commission rejected the view that precedents were relevant to this case. First, the Commission recalled that, according to the case law, past Commission decisions were not relevant. Each measure and every case have to be assessed on its own merits. [Paragraph 232]
Second, the Greek authorities also referred to the Commission decision concerning public subsidies to the new Athens airport. In this connection, the Commission explained that the finding of no aid in that case was because some activities were not economic in nature, or because they were not liberalised at that point in time. [Paragraph 234]
Third, in other cases concerning roads, bridges and tunnels, the Commission had limited its assessment to determining the proportionality of the state support or whether the operators derived an advantage after their selection through a competitive procedure. [Paragraph 235]
Having established that the tax measures contained State aid, the Commission proceeded to assess their compatibility with the internal market.
It first considered the regional aid guidelines but found that the measures in question were not conform with the requirements of those guidelines. The measures provided operating aid. Even though the port of Piraeus was in an Article 107(3)(a) area which was eligible for operating aid, the aid was not preceded by an analysis of the regional handicaps and why operating aid would have been capable of remedying those handicaps. In addition, there was no explanation why the aid was limited to a single undertaking if its alleged aim was to remedy handicaps affecting the whole of the area. Also, the aid was neither digressive, nor limited in time, as required by the regional aid guidelines. [Paragraph 251]
Then the Commission considered the compatibility of the aid directly on the basis of the Treaty and, in particular, Article 107(3)(c). Given the size and importance of the port of Piraeus in the Mediterranean, the Commission found that the aid promoted an objective of common interest.
With respect to the necessity of the aid, it could be established that the net revenue from the project was not sufficient to remunerate the investment costs [paragraph 255]. However, the Commission went on to reject the position of the Greek authorities that the aid was necessary.
“(257) The Commission has consistently considered that port infrastructure projects require considerable capital investments that can only be recovered in the very long term and their economic viability may not always be ensured without public support. However, in this case, PPA, the awarding authority that conducted the tender procedure for the selection of the concessionaire of the Port of Piraeus, had already estimated that the project’s economic viability would be ensured, something that is proven by the fact that according to the tender documents the selected beneficiary was meant to undertake the whole investment on its own expenses. In addition, PCT undertook the extension of Pier II and the construction of Pier III, by assuming on its own all the investment costs that this project would entail. When it submitted its bid that was accepted by PPA, it had estimated that its investment in the Port of Piraeus would be profitable for it without the need of any public support, as otherwise it would not have submitted the bid or it would have submitted it with a reservation as regards the profitability of the project in the absence of a specific fiscal treatment. Moreover, the fact that Cosco aimed at turning the port of Piraeus into the first container terminal in the Mediterranean Sea demonstrates the potential of this port, as well as the profitability of the investment project that was never questioned. Therefore it cannot be considered that the measures under examination were necessary to ensure the economic viability of the investment project.”
“(260) Moreover, the beneficiary never invoked the existence of a funding gap that needed to be covered by the measures under examination. The fact that PCT only quantified the amount of the aid after the opening of the formal investigation procedure by the Commission, i.e. almost 5 years after the signature of the concession contract, demonstrates that the amount of aid was not taken into consideration by PCT in its initial business plan, and in particular when Cosco decided to undertake the investment. As regards the Commission decisions invoked by the beneficiary, where the Commission approved non-notified aid in cases where the aid had not been quantified in advance, the Commission notes that the cases mentioned are not applicable in the current case, as they do not concern funding of port infrastructure, where a specific funding gap has to be determined even for an ex post analysis of compatibility. Consequently, this aid cannot be considered necessary for the implementation of the project, as PCT would in any case undertake it.”
These arguments of the Commission are not robust. The fact that PCT did not quantify the aid does not mean that the special tax measures were not taken into account when it submitted its bid. And, more importantly, it is possible that the port construction and operation are profitable without any further state assistance precisely because of the lower taxes that PCT has to pay. The Commission comments only in footnote 190 that “the economic viability and profitability of the investment project has already been confirmed by the fact that PCT’s investments in the port of Piraeus have already very positive financial results”, but it does not show that these “very positive” financial results would have remained positive without the tax concessions.
Therefore, this part of the assessment is, in my view, incomplete. Perhaps this is the reason why the Commission concluded its assessment with the statement that “(261) in any case, as already explained above, the measures under examination consist in uncapped fiscal advantages that constitute operating aid which is normally prohibited. Such aid can only be accepted in exceptional specifically determined conditions. In the context of the compatibility analysis of the funding of port infrastructure project on the basis of Article 107(3)(c), this type of aid cannot be considered as compatible.”
It is true that uncapped tax benefits are operating aid without any ceiling. In the absence of an ex ante study that could justify the granting of operating aid, perhaps due to specific regional or project handicaps, the Commission was right to find that the aid was incompatible. But it should have buttressed its analysis by showing that even by excluding the benefits from the special tax measures, the project was viable and therefore the aid was unnecessary.
The inevitable consequence of the incompatibility of aid is its recovery. The Commission ordered Greece to recover it both from PCT and its parent company COSCO.
 The full text of the decision is published in OJ L269, 15 October 2015, and can be accessed at: