I ageree that this is not a clear judgement in some aspects. I am not sure I agree with Professor Nicolaides’ first conclusion that the principle that infrassadstructure that is not constructed for commercial purposes does not confer an advantage to its users, rather I thought the CJ ruling had concluded that there was no transfer of State resources by the use of regulation without accompanying charges (the case being made that the charges were fines for unregulated use and there would be no forgone income if everyone obeyed the regulations). I thought the references to the intention of the regulation was not made to justify a claim of no economic advantage, but rather in connection with examining whether the regulation was compliant with non-State aid aspects of the Treaty (such as discrimination). The preliminary ruling seemed to make quite convincing arguments that economic advantage was actually provided to selected commercial undertakings which affected competition and Trade, so the only barrier to the conclusion of State aid was the issue of State resources. What I found surprising was that the CJ drew a distinction beteween the access to the infrastructure and the construction of the infrastructure, and therefore only considered the potential of state respurces being forgone as a result of the regulation of access. It does not seem justified to seperate the access to the infrasrtructure from its construction, as surely access is an integral part to the whole exercise of constructiong infrastructure. After all, infrastrucrture cannot be used without some access arrangements. In the case of the London Bus lanes, there seems likely to have been some investment costs in the project, even if the road itself already existed. Apart from the possible need to modify the road itself at junctions etc, there would innevitably be other costs such as lane markings, signage and public information campaign costs. It could trherefore be argued that the economic advantage was provided through the use of state resources used for the adaptation of the infrastructure and promotion of the access arrangements.

  2. von Phedon Nicolaides

    We agree that the judgment is confusing at several points. With respect to conferment of advantage, please see paragraphs 48 & 63. With respect to intentions, it is a fundamental principle that intentions do not matter. The way the Court refers to them contributes to the rather convoluted reasoning in this judgment. But I fully agree with your point concerning construction and access to infrastructure.

  3. von José Antonio Rodríguez-Miguez

    Another interesting comment focused on an essential topic on state aid practice. The MEIP or, as Commission uses to say now, the “market economy operator (MEO) test” [See the Draft Commission Notice on the notion of State aid pursuant to Article 107(1) TFEU at http://ec.europa.eu/competition/consultations/2014_state_aid_notion/draft_guidance_en.pdf p. 21et seqs. ] it always a “risk” task for the Commission, as recent jurisprudence proofs. But, as Prof. NICOLAIDES points out again in this new interesting post, it’s basically a proof question, in which the Commission has the harder work. Thanks for this new post.

  4. von Dominic Rowles

    An excellent article, and a good case study on state aid to SGEI. The interactions between the state aid regime and the public procurement regime are particularly interesting as they are two sides of the same coin. EU officials from DG Comp and DG Grow however sometimes know surprisingly little about how each other’s regimes operate.

  5. von Ioannis Platis

    Very interesting article, as always. And a special reason to be glad is that I also wrote some thoughts on that judgment and I reached identical conclusions.

  6. ”…But if the fee had to cover all true costs, then any subsidised infrastructure would never be used [since by definition aid is necessary when a project is not commercially viable without aid] or users would be presumed to benefit from State aid – an unwelcome expansion of the scope of State aid and of State aid investigations…”. Could you please explain why users would be presumed to benefit from State aid? As I understand it, users would be deemed as State aid beneficiaries on condition that what they pay for the use of the infrastructure falls short of the incremental costs plus reasonable profit which represents a minor portion of the full costs recovery. Why full cost recovery would be considered as State aid? Even if you mean State aid in favor of the operator still confuses me.

  7. von Phedon Nicolaides

    Thank you for the question. As I explain in the couple of sentences preceding the text you quote, the practice of the Commission is indeed to consider that a fee that covers incremental costs is free of state aid. But I go on to point out that in fact this is legal convenience. A user fee that does not cover all the costs of a subsidised infrastructure must entail some benefit for users. However, I happen to think that if the Commission would apply such a rule, it would have to assess many cases of involving small amounts of aid. It would spend much effort in controlling insignificant amounts of aid. Not a good policy.

  8. Now it makes sense for me. What I failed to understand is that in your reference to ‘users’ in the quoted part, you mean ‘users (that pay a fee amounting only to costs plus profit)’. Thank you for taking the time to answer my question.

  9. von Allison Argent

    London has a strict policy on transportation services, including that of cabs. There are two kinds of licensed cab drivers in London: The Cabs and The Minicabs. Cabs are basically taxis they are also sometimes referred to as London black cabs although they now come in many different colours, most of them are still black. The minicabs on the other hand belong to the Licensed Private Hire which includes limousines, airport transfer services and chauffeurs. Heathrow Transfer Reservation offers Heathrow minicabs and cab transfers.

  10. von Moritz Am Ende

    Reminds me about a (not so recent) judgment by the EFTA Court regarding the notion of undertaking/”economic activity” carried out by public bodies (Kindergartens in the case at hand). The Court followed a rather different approach by applying case law from the freedom to provide services (para 78 ff).

  11. Full cost revovery under Water Framework Directive may involve capital costs as well (for example, costs of building the infrastructure). Thus my understanding of the decision is that if capital costs were partially covered by lawful state aid this would be disregarded when deciding which costs are relevant for the market test? Therefore if one would alow for regional aid due to affordability issues and then seek for recovery for difference between full cost recovery and incremental cost + profit the regional aid as an instrument would be futile?

  12. Sir, the claim “no other means of producing electricity generated radioactive waste [paragraphs 78-79]” is not supported by available scientific evidence, as convincingly argued for instance by UNSCEAR review of technologically modified exposures to natural radiation. This authoritative UN body has for instance endorsed the conclusion that “radionuclide concentrations in ashes and slags from coal-fired power stations are significantly higher than the corresponding concentrations in the earth’s crust”. In other words, coal used for electricity production provides an immediate counter-example to the said claim. It would be interesting to speculate whether the conclusion of the Court still holds in absence of that leg. JM

  13. von Phedon Nicolaides

    Thank you for your comment. Paragraph 79 of the judgment states ‘as only that method [i.e. nuclear power] generates radioactive waste arising from the use of such fuel.’ You may be right that there are radioactive by-products in coal-fired power stations. I, of course, cannot assess the correctness of the scientific information on which the judgment is based.

  14. von Alexander Rose

    Phedon, an interesting and insightful article. I’m pleased that you recognise the contribution that the UK has made during its membership. I also feel that the UK has been a strong contributor, particularly when it comes to State Aid law, in sharing best practice and putting forward novel initiatives which have later been adopted in other Member States. It is important to note that the referendum question was very narrow and leaves the UK parliament wide discretion as to what model to pursue in the future. Indeed, I don’t think a European wide competition authority would necessarily be opposed by the UK – after all companies operate across borders and it is important that subsidy regulation / enforcement has the same reach. It should also be noted that in 2013 when the UK government consulted domestic stakeholders about State Aid law the feedback (as set out in the Balance of Competencies report) was wholly positive. Change is inevitable in light of the referendum, but I would hope that the UK and EU could cooperate to keep the rules that both parties want in place after Brexit.

  15. von Dr. Hans Arno Petzold

    For the author, “It is not so clear what the extra pension costs of Deutsche Post may be presumably because it pays the same pensions for all its staff”. This sentence does reveal a considerable misunderstanding of the German System of OAP for employees (Angestellte) vs. public servants (Beamte). The former receive their pensions (Rente) from a social security system equally fed by employers and employees, plus funds from government Budgets. The latter do not pay into that System, nor do their employers, who have to take the full burden of the pension from their budget. Obliging a company to employ public servants, who maintain their Status, means having to carry a substantially higher financial burden, compared to “normal” employees, where the Company only has to budget the 50%-share of the social security contribution. Companies not obliged to employ Beamte may have a higher burden while the employees are still active, but a substantially lower one after their retirement: zero. Apart from pensions, Beamte receive further benefits directly from thier employer that are otherwise covered by social insurance systems. Summing up, the court was right in assuming a higher financial burden for Deutsche Post than for its competitors, without Deutsche Post being able to avoid this disadvantage.

  16. von Phedon Nicolaides

    Many thanks for the clarification. Much appreciated. Indeed I admit I was speculating on this issue (that is why I used the words “it is not clear” and “presumably”) because the Court did not carry out any exhaustive investigation. Of course, the Court does not have to carry out any such investigation itself, but it could have at least tried to consider how the extra pension costs affected Deutsche Post. As you indicate, companies that do not employ Beamte have an initial higher cost. This means that companies that do employ Beamte have an initial lower cost and therefore advantage. Perhaps Deutsche Post also derived an indirect advantage because it was able to retain more experienced staff rather than losing them to the competitors who entered the market after liberalisation. Perhaps Deutsche Post was able to reduce gross wages to offset the higher pension costs. The Court merely focused on one cost component – the pensions – and inferred that it reduced the overall competitiveness of Deutsche Post without considering how Deutsche Post may have responded to neutralise that disadvantage, if it was a disadvantage. Competitiveness is a composite concept and the Court did not seem to understand that. The intention of the German authorities was to compensate Deutsche Post, but what were the actual effects of the pensions obligation? Recent decisions of the Commission on compensation of postal operators for public service obligations have also considered benefits from these obligations, such as network effects, visibility, etc. Legally, the Court did not ask what were the actual effects of the pension obligation. It applied instead only half of the Altmark criteria

  17. von Dr. Hans Arno Petzold

    Thank you for picking this decision. It has several important aspects. Firstly, it recognizes that the activities of STPs and similar organizations are “economic activities” falling within the scope of art. 107 TFEU. Secondly, it makes clear that users of the STPs or similar structures receive State Aid, even if de minimis, and have to comply with the de minimis procedure. Both questions were so far under discussion. Furthermore, the reason why COM did not refer to “Innovation cluster” should be seen in the diverging direction of clusters and infrastructure. The focus of clusters lies on the co-operation of established enterprises (cf. art. 27 GBER), while incubators and the like focus on bringing start-ups to the market – a far more single target approach. To ask the question is justified, but from the practitioner’s view, the answer seems quite clear.

  18. von Zoran Sretic

    The article asks „how will the UK establish its own system of State aid rules?“ The UK state aid control system’s scope would be defined by a treaty regulating „Brexit“ (the „Brexit Treaty“). Akin to the EU contractual practice under family of association agreements (FTAs under WTO rules) the EU will probably insist on applicability of Article 107 TFEU and its „instruments of interpretation“ even beyond the exit. The instruments of interpretation will be linked to the EU institutions practice (i.e European Commission and the CJEU) hence their content will evolve accordingly. Under such scenario, space for development of autonomous UK legal concept of „common interest“ would be somewhat limited. In any case, such concept will gravitate around a notion of distortion of trade between the EU and the UK. In other words, the autonomous concept of aid reserved for distortion of the UK market alone would be redundant. Indeed, the leverage that the UK negotiators would have compared to administrative capacities of association states may provide more autonomy to the UK from the Brussels to develop more progressive industrial and state aid control policy. On the other hand, the EU association agreements so far have lacked a robust and effective system of control of compliance of state aid practices in parties to the agreement. Therefore, the real question would be if an institutional arrangement of the Brexit Treaty would provide effective means to control the aid compliance with its provisions. Likewise, another question would be if UK companies will have effective and non-discriminatory access to the EU’s system of state aid control to tackle any non-compliant aid with the future trade agreement.

  19. von Rhodri Davis

    Canada, South Korea and Switzerland all have free trade agreements with the European Union without having to sign up to EU state aid control. What irks me about this piece is the assumption that state aid control in the U.K. has to rest with the CMA in Central London. It is time to recognise that this kind of administration can be more cheaply delivered in other cities such as Cardiff or Leeds. What exactly are the implications on U.K. businesses, would they still be able to challenge breaches of state aid in EU countries?

  20. von Dr Eugene Stuart

    This case is one too far by the Commission and directly interferes in national sovereignty to set up progression in tax measures. The argumentation is very vague in places and it may well be a politically driven decision

  21. von Dr. Hans Arno Petzold

    Thank you for this post. Let me add another practical example: national or regional “Investment Banks”, often publicly owned or dominated, usually have their share of “private business” where they decide as a bank and under – more or less – market conditions. But they often also act as “intermediate bodies” in the sense of EU Reg. 2013/1303 (GPR for the ESI Funds). Here, the Banks execute the funding directives of the respective administration, often notified as aid schemes under the GBER. Even insofar as they have – within the Framework of the aid scheme – a certain discretion whether to fund or not to fund, this discretion is granted and limited by State decisions so that in General this funding will have to be imputed to the State and the measure as State aid.

  22. von Phedon Nicolaides

    Very good example. Thanks. Indeed, any (public or private) fund manager operating under a contract with a public authority would fall in that category too.

  23. Dear Phedon, Many thanks for another very informative contribution and the observation that it seems clear prima facie but not so clear when you think it through. I believe that you’re correct when you state that the identity of the funder of the activity should not matter. In this regard you write: The Court went on to add that “(49) it is not necessary for that private financing to be provided principally by the pupils or their parents, as the economic nature of an activity does not depend on the service concerned being paid for by those for whom it is performed”. This is also true. Costs must always be covered one way or another. This applies to non-economic activities too. But the significance of this statement is not obvious. I’d say that it is obvious and problematic at the same time. The first (non-problematic) example to come to mind would be the situation where a company pays for – say – one of the courses you teach to one of their employees. This is obviously economic in nature. The problematic example is where the church funds such services (paraphrasing the Court here) not seeking to engage in gainful activity, but in order to fullfil its social, cultural and educational obligations towards its congregation. In this example, the objective an entity pursues (lets call it enlightenment) makes the activity non-economic. This cannot be right in light of the Court’s standard considerations that its the effects, not the intention that matters. So I guess this is just the Court’s response to entering a minefield: gnomic caselaw.

  24. von Phedon Nicolaides

    Many thanks for the comment. We agree that regardless of motives, someone has to bear the full cost of the good or service that is provided. When I buy food for my children, I provide it to them free of charge, but I still have to bear the full cost. The fact that I am not acting economically or commercially does not make costs disappear. It merely shifts them from the users to the provider. A commercial operator would accept to bear costs that exceed the price it charges only temporarily and only because it is a profit-maximising strategy in the long-term. It is in this sense that, in the context of state aid law, the intentional shifting of costs away from the users indicates a non-economic activity. For example, the cost of shelters for the homeless is borne by voluntary contributions. The cost of public education is borne by taxpayers. But there is another reason that this shifting of costs is intentional. The users are not charged a price that covers cost because they are considered to be entitled to or in need of the particular good or service. Price is just a mechanism for allocating scarce resources according to ability to pay. But some goods and services are provided regardless of ability to pay so as not exclude those who cannot afford them. Ability to pay, in this instance, would be contrary to rights or needs. Therefore, if you think that intentions would not fit in state aid law, then this idea can be expressed in the form of the following two-step test: i) Is the good or service provided on a constant basis at a loss without contributing to the revenue of another good or service? ii) If yes, is “consumption” of the good or service in question rationed on the basis of non-price criteria and users have no obligation to buy any other good or service? The answer must be “yes” in both steps.

  25. Thank you for yet another eye-opening article. The following to-the point remark got me thinking: “Hence, whether something is economic does not depend on who funds it, but on whether those who provide it charge a price that covers its cost and, as a result, they get to be remunerated.” One could, in economic terms, argue that the Congregación (the output side) has charged a price that constitutes remuneration when the total amount of state subsidies based on an agreement and student fees (the input side) covers the cost of the education. The buyer would be both the students and the State. Under such terms, the Congregación would not be making a permanent loss. The Court, however, distinguishes between private funds (para 48) and public funds (para 50) on the input side. It is this distinction that seemingly enables the Court to reach the conclusion in para 56 that “the educational activities of the Congregación that are subsidised by the Spanish State could not, according to the Court’s case-law set out in paragraphs 41 to 50 of the present judgment, be classified as ‘economic’”. The antithesis being that if the funds had come from a private entity, the outcome would have been very different. The key to the Court’s distinction between private and public funds could perhaps be not the origin of the funds themselves, but the fact that the Congregación is “integrated into a system of public education” by way of the agreement described in para 55. Apparently, the Court has identified the part of the Congregación providing the education and the State as one, for the sake of the assessment of an economic activity, where the virtues of the State (providing the services at a permanent loss for the sake of solidarity) prevails. In fact, the Court states in para 50 that the State “is not seeking to engage in gainful activity”, but it does not assess or comment on whether that is the case also for the Congregación. Nevertheless, the Court concludes in para 56 that the activity is not of an economic nature. This implies that the Court takes it as granted that the Congregación is not seeking to engage in gainful activity. One could argue that the Court takes a holistic view, where only the characteristics of the service seen from the beneficiaries’ point of view matter: There is only a nominal payment for services that the beneficiaries are entitled to, and the state funds the services based on solidarity and universal access. The practical arrangements as to how the services are produced and delivered are not relevant. The providers of the actual service, whether they are public or private entities, carry out non-economic activities even though they may receive remuneration in economic terms. This line of thinking should be the same where a private charity funds the services, at a permanent loss. In such a case, the service provider would be “integrated in a system” of charity. If one were to extrapolate the Court’s findings, do you find this to be a fair generalization? Private entities do not carry out an economic activity when a) they provide services to beneficiaries who (by law) are entitled to them, b) the beneficiaries only pay a nominal fee and c) the State (or a private charity making pemanent losses on the services) covers the private entity’s remaining costs according to an agreement or contract set up in accordance with the EU rules on public procurement?

  26. von Carmen Butler

    Dear Professor Nicolaides, Thank you for your article, insightful and informative as always. One statement in your article raised a question for me. You wrote, ”An activity or product is non-economic when the beneficiaries or users have a right to it [e.g. health or education services], deserve it [e.g. research awards or prizes] or need it [e.g. shelters for homeless or rescue services].” I reacted to your statement that an activity can be non-economic when the beneficiaries or users deserve it. You gave two examples: research awards or prizes. There are certainly limits to when such activities can be considered non-economic, but the statement here is broad-reaching. I wonder if you could comment. Thank you.

  27. von Phedon Nicolaides

    Dear Vegard, thank you for your comment. It is a complex issue. Your last point on compliance with public procurement rules reminds me of the Altmark criteria. I think an entity selected on the basis of a competitive procedure, having its costs covered by the state, can be very well acting as an (efficient) undertaking

  28. von Phedon Nicolaides

    Dear Carmen Butler, Thank you for your comment. What I had in mind was a foundation awarding prizes to researchers who discover something significant [e.g. Nobel prize]. Naturally, state subsidies to research can also be granted to economic activities, even if the recipient undertakings may “deserve” them because they do something good for the economy.

  29. Dear Professor Nicolaides, Thank you for this interesting article. It would be very useful to get more insight information on the national/regional system the Dutch government uses to measure the total amount of de minimis aid per beneficiary. Apart from Luxemburg, I don’t know of any other member state who is developing a national de minimis database (as strongly advised by the Commission). How was the Dutch government able to convince/prove to the Commission that the total amount of public support granted to beneficiaries under this measure did not exceed the threshold contained in the de minimis regulation. Do you have more information on this?

  30. von Phedon Nicolaides

    Thank you for your query. There are several Member States, especially among the newer ones, who have developed national data bases for recording de minimis aid. However, many Member States rely on self-certification by aid recipients [whereby, granting authorities ask recipients to confirm that they are single undertakings and to specify all de minimis amounts granted to them in the current and previous two fiscal years]. As to how the Dutch authorities calculated the amount of aid, please see the text of the decision [a link is provided]. The Dutch authorities calculated the precise amount of the aid they granted to each undertaking. This is the only way to ensure that the de minimis threshold is not exceeded.

  31. von Dr. Hans Arno Petzold

    Well done, Erika, I fully agree and appreciate your clear-cut and outspoken positin!

  32. von Frances Selkirk

    If the UK does create an authority to regulate state aid then it would be an own goal to automatically base it in London. It would show the same out of touch thinking that meant the Brexit vote was such a surprise

  33. Prof. Nicolaides could, the Court, pull another argument in your opinion that State Aid rules cannot be misused to achieve effects contrary to their purpose. Namely, it appears to me that if the imputability was found (or if indeed regulatory authorities influenced the company not to purchase) the competition would have been distorted.

  34. von Phedon Nicolaides

    Thank you for your question. It is not possible to say what the Court could have said otherwise. No one knows that. But if the state acts as a regulator, then it must act in an impartial, non-discriminatory and consistent manner. I cannot imagine how a regulator can achieve public policy objectives [which, by definition, aim to raise overall welfare] by seeking to influence the behaviour of only one company, unless, of course, that company infringes the relevant regulations.

  35. von Dr. Hans Arno Petzold

    The problem with selectivity is, as you correctly elaborate, to find the proper comparison, or yardstick, as beautifully demonstrated in C-524/14 P European Commission v Hansestadt Lübeck (cf. EStAL 2/2017, p. 285). But already in C-403/10 P – Mediaset v Commission – the problem of different broadcasting systems arose, cable v terrestrial. It was thus quite manifest for the Commission to somewhat reduce their efforts in reasoning – and consequently the court followed them in most of the cases. In fact, in C-70/16 P, para. 61, the Court does not say that the Spanish measure was not selective, only that the Commission should have reasoned on this point. In the end, the result will be predictable, but the Commission is rightfully reminded to take nothing for granted or self-evident. This argument may sound formal, but judicial protection starts with upholding the formalities …

  36. von Victoria Rodríguez

    We have won! The judge considers that all the requirements to be considered a State aid are met. The entity carries out economic and non-economic activities in the property subject to the tax and meets all the requirements of the Article 107 of the Treaty on the Functioning of the European Union. It is a selective advantage, attributable to the state and capable of affecting trade between Member States and distorting competition. It has also considered that it is not possible to apply the exception of the Minimis Rule, because the religious entity has not proven that the threshold of 200.000,00€ has not been exceeded. The Congregation did not provide separate accounts of the subsidized and non-subsidized educational activity). In addition, the judge doubts that the application of this exception of the Minimis Rule to the case because of the difficulty that entails because it is a non-transparent tax in terms of the Regulation.

  37. von Phedon Nicolaides

    I very much agree that the Commission should take nothing for granted. Perhaps it thought that in this case the selectivity was self-evident. It was wrong. It had to be proven. And, yes, formalities must be respected.

  38. von José Antonio Rodríguez Miguez

    It is really a very controversial case, that shows again the complexity of the notion of State Aid and the alternative versus cumulative question about the difference inter state aid and/of public resources. I will read the Decisión again. Thanks again, Prof. Nicolaides, for showing us this very interesting cases.

  39. von Dr. Hans Arno Petzold

    Just some few remarks: Though electronic or instrument landing Systems (ILS) do increase safety of air traffic, they are indeed an economic factor as they allow airports to host larger airplanes, thus increasing the capacity of the Airport and its attractiveness for carriers and tour Operators. But ILS is not generally mandatory, as opposed to airport firefighting services (cf. EU Regulation 1108/2009). The question you put forward is at Hand, but the answer is probably: never. On the question of distortion of competition it is interesting to note that when it comes to “local” measure, the Commission wants to turn from the abstract to a more concrete assessment while here, in a rather complicated environment, they don’t, and don’t have to, according to the Court. This has a slight smell and taste of cherry picking …

  40. von Dr. Hans Arno Petzold

    This is an interesting judgment, indeed. The Slovak system does very much resemble the German social health insurance system, only that there are a lot more “competitors” offering their services to those obliged to join the system (about 90% of the population). Even with contribution regulated by law (a percentage of the income from employed work, up to a moving threshold), and little room for any additional premium, the insurers – all organised as public-law corporations – still manage to vary in amount and quality of service, be it by “economies of scale”, be it by lean management or other means of cost efficiency. Private insurance companies are not admitted to this market, but according to the GC, this is a market. It will be interesting to see how the ECJ thinks about this, and what consequences this may have also for the German system, which is under massive criticism anyway.

  41. von Phedon Nicolaides

    I very much agree with you. We see that both the Commission and EU courts tend to extend the frontiers of the market [as in the recent judgment on Charleroi airport]. But the implications of this judgment are rather unpredictable. Probably the most “explosive” part of the judgment is in paragraph 69: ‘the fact that the offer of goods or services is made without seeking to make a profit does not prevent the entity which carries out those operations on the market from being regarded as an undertaking, provided that the offer exists in competition with that of other operators that are seeking to make a profit. It follows that it is not the mere fact of being in a position of competition on a given market which determines the economic nature of an activity, but rather the presence on that market of operators seeking to make a profit.’ We used to think that similar economic and non-economic activities could co-exist. Now the General Court is telling us that if there is a profit seeking entity, the activities of a non-undertaking become economic in nature if they are ‘in competition’ [which means that they are similar enough]. The next judgment will have to explore further what it means to be ‘in competition’. But I am sure we all can think of many economic and non-economic activities that are sufficiently similar. Are they ‘in competition’?

  42. von Caroline Wehlander

    This importan judgment seems to confirm my analysis that not only for-profit but also non-profit and municipal schools offering their services free och charge but in competition in the Swedish publicly financed system (a system where pupils may choose the school they attend), must be seen as conducting an economic activity. The same goes for elderly care in systems of free choice in Sweden, where for-profit entities, non-profit entities and public institutions in competition offer their services which are financed to a large extent by the state. See Services of General Economic Interest as a Constitutional Concept of EU Law, 2016, Springer, p. 297-339. I based my analysis of the Swedish cases (free choice systems for school and elderly care) on a very detailed analysis of the notion of economic activity in the case law of the CJEU, which is shortened in the book (chapter 2) but considerably longer and more detailed in my PhD thesis.

  43. von Dr. José Antonio Rodríguez Miguez

    Undouble a very interesting case. Are very interesting and important your comments about the different between the Private Investor Test and the Creditor Investor Test even now both are presented under the general title of Private Operator Test. I am not sure this new general category (Private Operator Test) was the more useful; basically because under this umbrella it possible to think wrongly that all the test are the same and all have to be applied in the same way. The only common element is the market reference to a predicable behavior of a theatrical private operator but every test is different from the other because the perspective are different too. Probably in this very case the point of view of the General Court was more subtle and sensitive with this evidence. Obviously, as I always do, all we have to thanks Prof. Nicolaides for his excellent selection of cases and his brilliant comments. After reading many of them the only conclusion that appears clearly is the enormous complexity of State Aid practice and how many questions remain unsolved or almost not definitely solved at all.

  44. von Dr. Hans Arno Petzold

    I’m not sure whether I fully grasp your conclusions. The constructions to be found in ports do vary across the EU, so funder and owner are not necessarily identical. Nor are owner and operator, though especially public entities usually employ an operator, selected – hopefully – by a proper tendering process (cf. art. 56 b para. 7 GBER). In this case, the operator has to pay a “concession fee” to the owner which then, as determined by the tender, reflects the market price and excludes any State aid to the operator. If the owner operates himself (no “third party” involved), there is no question of State aid (cf. GBER). Concerning State aid to the users of the port (the shipping lines), I refer to the judgment C-524/14 P (Flughafen Lübeck) and the ECJ’s clarification on selectivity: the yardstick is the airport itself, not the neighbouring one (cf. para. 123 of this judgment on the opposite way of argument). It is thus not relevant whether or not the selected operator wants to maximise its profit, but that it does not unjustifiably discriminate against any user. In this constellation, a comparison to port fees in other ports is not required. Only in the case where the public owner of a port is also the operator, it will – in order to avoid State aid to the users (the shipping lines) – have to charge port fees according to market prices, at least cover its own full costs plus a usual market profit, which is easily to be determined by studying the annual (or quarterly) reports of other port operators. For me, the main clarification in this judgment concerns the obligations of the Commission and the necessary scope of its investigation, plus underlining that a factual monopoly caused by lapse of time may very well constitute State aid, be it intentional or not.

  45. von Rita Aleixo Gregório

    I tend to agree with the suggested line of reasoning in respect to the calculation of the recoverable amount. It seems a more plausible and frankly more realistic and fair method of calculation, considering that the purpose of recovery is to re-establish the situation that existed on the market prior to the granting of the aid. It seems to me that such line of reasoning would, nonetheless, introduce speculative elements around these matters as, most of the times, it would be difficult to demonstrate straightforwardly the exact amount of electricity that a company would have bought in the absence of a contract for the supply of electricity at a reduced tariff. I am not sure either whether this line of reasoning has been tested before the EU Courts, but I would be surprised if the Commission or/and the EU Courts would be willing to accept it, particularly considering the implications that it may have for the immediate and effective execution of a recovery order.

  46. von Dr. Hans Arno Petzold

    This is an interesting point, indeed. But I have doubts whether this argument will carry through. Any counterfactual scenario must be realistic and credible in itself. In this case, AoG would have had to argue that higher costs for electricity would have deteriorated their competitive position on the market, thus led to reduced sales, reduced production, and reduced consumption of energy. This would necessarily imply that they were not able to compensate for higher prices for electricity at some other point in their calculation or cost structure, and that this claimed reduction was inevitable and coherent in itself. I find this a rather daring approach, considering the complexity of markets and cost calculations, which I would not necessarily advise to my client, if only for the sake of the argument. N. B. In this case we are not talking about the presence of aid, but only discussing its amount, aren’t we?

  47. von Melvyn Waumsley

    There is of course a good argument for not calculating the aid from the counterfactual perspective. Given that there was an actual amount of electricity consumed with the help of the concessionary price, the resulting aluminium produced was driven by this reduced price, therefore the effects on trade relate directly to value of the unit price concession multiplied by the real number of units of electricity consumed in the period.

  48. von Phedon Nicolaides

    All good points. As I wrote, I do not think there is a precedent for the use of counterfactual in the case law. For sure it will not be easy. Since the argument put forth by AoG was hopeless, I suppose defendants will have to come up with more innovative approaches. I was testing a possible option.

  49. von Dr. José Antonio Rodríguez Miguez

    This is another very interesting case. As it is clearly noted in this post, the basic question is whether the levy of 5% over the gross revenues deriving from the provision of private services is o not relted with the real use of the facilities or inclues the materials, machinery… Is it a way to give these doctors a complementary salary or a genuine state aid to them. These doctors can work in private center or center opened by they own. It is similar, in such way, and mutatis mutandi, at what happen with universities and the “private” activities of the professors of the university institutes when work for firms…where is the limit?. I will read carefully the Judgment of the General Court of 5 February 2018 in case T 216/15, Dôvera zdravotná poist’ovňa et al v European Commission.

  50. The EU case law notion of economic activity as offering goods or services on a market is well established. I wonder whether it could be possible clarify further the meaning of offering goods or services on a market. In my opinion, the economic point of view is the right approach to answer the question, even for a correct legal glance on the topic. I think that there are two considerations to take into account. First, offering goods or services (set aside for the moment any consideration about the reference to a market) implies investments of scarce resources and the selling of the output of a production process, as clarify in the judgement: “(47) services normally provided for remuneration are services that may be classified as ‘economic activities’. The essential characteristic of remuneration lies in the fact that it constitutes consideration for the service in question”. So, we need to ascertain that those scarce resources have been transformed into a product for which someone – i.e. customers – is able and willing to pay a remuneration. The economic problem rises whenever limited resources, which have alternative uses, face up unlimited human wants. That means that the remuneration is the outcome of a customer’s choice. The finance flows from customers to producers. The other consideration regards what means market in the EU caselaw notion. As “(46) not-for-profit basis does not prevent the entity which carries out those operations on the market from being considered an undertaking, since that offer exists in competition with that of other operators which do seek to make a profit.”, I think that there have to be not only customers willing to buy, but also other competing firms, which have been offering other goods/services. In other words, a market exists if customers can enjoy the freedom to choice what to buy and consume, because of a large numbers of producers. The freedom to choice also implies the choice to give up goods/services the consumer do not want. For example, a supermarket employee could not place an unwanted product into a shopper’s basket and expect the shopper to pay for it at the checkout. This is called the principle of rejectability. On the market the sovereign is the consumer “whose buying or abstention from buying ultimately determines what should be produced and in what quantity and quality” (s. Mises, The Anti-Capitalistic Mentality). “The customers are “always right,” [they are] the patrons who have the power to make poor suppliers rich and rich suppliers poor. (…) The profit system makes those men prosper who have succeeded in filling the wants of the people in the best possible and cheapest way. Wealth can be acquired only by serving the consumers. The capitalists lose their funds as soon as they fail to invest them in those lines in which they satisfy best the demands of the public. In a daily repeated plebiscite in which every penny gives a right to vote the consumers determine who should own and run the plants, shops and farms. The control of the material means of production is a social function, subject to the confirmation or revocation by the sovereign consumers” (cfr. Mises cit.). Provided those conditions, we should assume that an economic activity in the EU law is an activity financed by the buyers of the goods/services produced. As a consequence, it does not exist any economic activity in nature. This is a logical contradiction. Any activity is economic as far as it can satisfy social needs. The problem is elsewhere and consist in looking into the financing of the activity concerned. The question is way how the activity is financed and by who, as stated by the EU Court. If “(56) the educational activities of the Congregación […] are subsidised by the Spanish State, [they] could not […] be classified as ‘economic’.” “(57) By contrast, […] educational activities that are not financed by the Spanish State, […], meet all the criteria […] for classification as ‘economic activities’”. I will appreciate very much any reaction to my comment.

  51. von Phedon Nicolaides

    Thank you for your comment. I agree that the market is based on voluntary transactions. Anything that is based on rights [the right to schooling] or compulsion [the obligation to contribute to the state pension system] is not a market transaction. EU courts would not disagree with you. I also agree that in this case the court has muddied the waters with its reference to the source of funding. Perhaps it intended to highlight that state-funded schooling is a right and therefore it is not a market-based transaction.

  52. The case T‑216/15, Dôvera zdravotná poist’ovňa et al v European Commission, is a very very interesting one because in my opinion it raises the underlying question of what be the legal foundations of a free (i.e. competitive) market. In order to decide whether an activity is economic or not, the profit motive does not matter – that is obvious if we distinguish the profit as purpose of operation of business, which is an ex ante choice of the entrepreneur, from the profit as actual ex post result of that operation, which depends on the choice of the consumer. As stated by CJEU, “It follows that it is not the mere fact of being in a position of competition on a given market which determines the economic nature of an activity, but rather the presence on that market of operators seeking to make a profit”. Yet, the presence on …. market of operators seeking to make a profit is a mere fact. The number and in particular the possibility to enter the market depends on the law. In the case T‑216/15, Dôvera zdravotná poist’ovňa et al v European Commission, the essential point is in my view, that the social security system be not a market at all because of the country’s legislation. It is due to that legislation, i.d. State intervention, if there are only three undertakings, or in other words, there is no competition and consequently no market, as health insurance is compulsory for each individual. This situation guarantees to realize quite easy profits, although improperly. In such situation, no company has any stimulus to compete and try to better up their services or to reduce the cost of their operation (i.e. to be innovative and efficient). The marginal company can simply rise up their prices, recording no losses to their turnover. We are facing a situation of rent granting. Paying a higher price because of the lack of real competition is quite equivalent, from an economic view, to paying taxes, which the State gives subsequently out to subsidise the activity concerned. Money does not flow from taxpayers into the State, but consumers pour it out directly in the firms’ hands as higher prices. In both cases, the undertaking receive an undue advantage, unjustifiable because paid by the consumer, always involuntary and often unaware as well. The consumer is simply exploited: they cannot choose a) neither to buy and above all not to buy the health coverage of their preference; b) nor the benefit from the insurance as “…Slovak health insurance companies are obliged to offer the same statutory benefits …”. There is no market within the Slovak compulsory health insurance system, either in respect of the compulsory statutory benefits or formally on the amount of contributions. Competition requires a market clear of State interferences, any form they can assume. Market requires freedom of contract – or better the freedom to carry out any economic activity without any other limitations than civil rights and freedom of consumers – and protection of private property. On these grounds, competition can emerge as a feature of the market. So that we could say that the purpose of competition law is twofold: 1) to increase the ever changing set of choice (above all the choice not to choose) rather than to protect the consumers’ rights, and 2) to protect market from State intervention (State aid law). Consumer is the sovereign.

  53. von LL.D. Caroline Wehlander

    I note with much astonishment that in its decision of December 2017 on public hospitals in Lazio, the Commission omitted to mention its important decision on the public financing of Brussels public IRIS hospitals (SA.19864) adopted 2016 after the annulment by the general court (T-138/10 CBI) of the Commission’s initial decision in the case. In the IRIS-decision the Commission found that both public and private hospitals in competition in the Brussels region conducted an economic activity, although both to a large extent were funded publicly, in particular the public hospitals which had particular SGEI-obligations. Clearly in the IRIS-hospitals decision, the fact that the public hospitals were to a large extent solidarity funded and provided their services free of charge on the basis of universal coverage did not keep the Commission (and not the General Court either) to find that their activity was economic. In the IRIS-decision this assessment was made at point 108 with reference to the CJEU reasoning in paragraph 58 of Smits and Peerbooms. In paragraphs 56-57 of the Lazio-decision, aptly brought to attention by prof. Nicolaides, the Commission does not reason on why publicly financed hospitals in competition with private hospitals, of which many are driven for-profit, were found to conduct an economic activity in the Brussels-region but not in Lazio. The Commission refers first vaguely to the situation in “other Member States” where “hospitals and other healthcare providers offer their services for remuneration, be it directly from patients or from their insurance” and second to paragraphs in the decisions of the General Court and of the Court of Justice in FENIN which are not relevant for the assessment of the economic character of healthcare (paragraph 39 in T-319/99 FENIN v Commission is about healthcare administration and not healthcare services). A correct application of state aid and procurement rules to social services markets that rapidly grow in the Member States is key for peoples’ trust in EU-institutions. I do hope that light on the important issue of whether for-profit hospitals in publicly-funded mixed systems may be seen as conducting a non-economic activity will soon be brought by the Commission itself and not first by scholars.

  54. von Phedon Nicolaides

    This is absolutely correct. But the point is that the decision could have explained why there was no burden sharing. I will review the decision on the restructuring of MPS in a forthcoming article.

  55. von Phedon Nicolaides

    Last week, the Commission appealed against the judgment of the General Court. Let’s see what the Court of Justice will have to say on this very important issue and how it will approach the concepts of market and competition.

  56. Hi Daniel,

    Thank you for the interesting blog post. Collusion through or with the aid of software is indeed a practice which poses some new questions for antitrust enforcement, although, as I argued in a paper on this topic (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2987705), I believe EU law can deal with most of these questions. I’m not familiar with the facts of the Partneo case, but isn’t a big issue here whether the car manufacturers are really competitors for the products concerned? Customers may see Renault and PSA as competitors when they are looking to buy a car, but when they already have a Renault and they need to buy Renault spare parts, do they really care so much about the pricing of spare parts of Citroen cars?

    Kind regards,


    • von Daniel Mandrescu

      Hi Jan,

      You make a fair point. I agree with you that once a car is bought the car makers will not compete with each other with regard to spare parts since these are not interchangeable. However, the price of spare parts can make a difference when it comes to choosing a car in the first place. The importance of the pricing of spare parts for the choice of car will depend on the type of customer you consider.

      Customers which purchase big orders of cars every few years (e.g. leasing companies/ big corporations with lease plans for employees/ public sector/ taxi’s drivers) may care quite a bit about the price of such parts when making their choice. For consumers the price of such parts maybe appear less interesting since most do not know the price of spare parts until they have to replace them. However even in their case it may play a role as the price of these parts will be taken into account when it comes to insurance where the differences between cars may be substantial, particularly when some parts (and cars) are often stolen ( https://www.autoblog.com/2018/02/19/car-theft-parts-prices/ ).

      Whether EU law can deal with future algorithmic / automated decision making, it remains to be seen. Many things can indeed be already tackled under the current system since we are not yet at the point where AI has removed the need for human intervention. However the less human interaction is involved in the business practices of undertaking the more difficult it will be to prove liability from both a substantive and a practical point of view.

  57. Hi Daniel,

    Thank you for the interesting blog post. Collusion through or with the aid of software is indeed a practice which poses some new questions for antitrust enforcement, although, as I argued in a paper on this topic (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2987705), I believe EU law can deal with most of these questions. I’m not familiar with the facts of the Partneo case, but isn’t a big issue here whether the car manufacturers are really competitors for the products concerned? Customers may see Renault and PSA as competitors when they are looking to buy a car, but when they already have a Renault and they need to buy Renault spare parts, do they really care so much about the pricing of spare parts of Citroen cars?

    Kind regards,


    • von Daniel Mandrescu

      Hi Jan,

      You make a fair point. I agree with you that once a car is bought the car makers will not compete with each other with regard to spare parts since these are not interchangeable. However, the price of spare parts can make a difference when it comes to choosing a car in the first place. The importance of the pricing of spare parts for the choice of car will depend on the type of customer you consider.

      Customers which purchase big orders of cars every few years (e.g. leasing companies/ big corporations with lease plans for employees/ public sector/ taxi’s drivers) may care quite a bit about the price of such parts when making their choice. For consumers the price of such parts maybe appear less interesting since most do not know the price of spare parts until they have to replace them. However even in their case it may play a role as the price of these parts will be taken into account when it comes to insurance where the differences between cars may be substantial, particularly when some parts (and cars) are often stolen ( https://www.autoblog.com/2018/02/19/car-theft-parts-prices/ ).

      Whether EU law can deal with future algorithmic / automated decision making, it remains to be seen. Many things can indeed be already tackled under the current system since we are not yet at the point where AI has removed the need for human intervention. However the less human interaction is involved in the business practices of undertaking the more difficult it will be to prove liability from both a substantive and a practical point of view.

  58. von dr. José Antonio Rodríguez Miguez

    Thanks very much, Professor Nicolaides, for quoting and comment this very interesting case on taxation and cross-border effect. Indeed, many state ais scheme as designed as a reaction in from of other member states actions. S harmonization in taxation matter is a real problem beyond state aid control, but not, as Court clearly explains a justification for the infringement of competition. This case remember all of us the crucial problem in Eu common project: more European vision or more Member States vision. State Aid rules are not the solkution for this problem that, in such way, put in evidence.

  59. von Dr. Hans Arno Petzold

    Interesting subject indeed, in case the UK does finally leave the EU. Even more difficult to predict if tthey were going to exit without agreement, this might deteriorate any chances for a common State aid rule book, but also reduce Whitehall’s chances to come to terms with the devolved admins/governments (by the way, what about Gibraltar?). Another point of interest: The White Paper (p. 11) contains “a commitment that UK courts would pay due regard to EU case law”. Presumably, this would include State aid case law. But how can the government commit independent courts? Isn’t there a division of powers, established since the 13th century? And how could a “Joint Committee” (of government officers?) arbitrate in case of diverging rulings from CFI/ECJ and UK courts? It seems to me rather unlikely that the UK was willing to give up long-standing constitutional principles, models the world over, for the sake of something as ludicrous as “Brexit” …….

  60. von Phedon Nicolaides

    Good questions. Indeed there is a lot of uncertainty. But see this document that was published recently

  61. von Pierpaolo Rossi

    Hello – I generally agree with the analysis – However your criticism about classifying exceptions of general application as selective measures ignores the notion of justification – Indeed, there are exceptions of general application which are not selective because justified by the nature and logic of the tax system of reference. If for example, Spain would have demonstrated that there was a tax logic in the tax lease structure by which one can amortize fully building costs but forfeit tax on gains because taxed at a later stage then the exception of lower tax now would have been justified and therefore not selective

  62. von Phedon Nicolaides

    Thank you for your comment with which I agree. I considered as self-evident and well-understood that something that is justified by the logic of the tax system is not selective. I wanted instead to focus on exemptions which are not justified by the logic of the system, because this this the crux of the issue. Under which conditions can they be considered as general or selective? My view is that the Court conflated ex ante application of the exemption with the ex post outcomes of the exemption. If the application is non-discriminatory [absence of discrimination at the heart of the concept of selectivity], then the exemption is general even if the outcomes may vary.

  63. von Apostolis Grammis

    In light of that recent case law, do you think that a “horizontal exemption”, like the reduction in social security contributions to undertakings employing manual workers (as in Maribel scheme), is a selective measure?

  64. von Phedon Nicolaides

    Good question. Probably yes, but it is far from certain. After the recent judgments and especially the judgment on the Spanish digital tv, what is certain is that it must be explained why all undertakings can employ manual workers or why employers of manual workers are not in a situation comparable to that of employers of non-manual workers.

  65. von Michael Papadakis

    Faidon, many thanks for this intriguing “food for thought” piece of analysis as well as the other excellent pieces of state aid case law commentaries. In my view there must be a significant difference between granting a taxi-licence, a doctor licence, or a banking licence and a licence to explore state owned natural resources. The latter is in my view about the sale of a public asset and ought to follow the abolished Communication on the sale of public land and buildings, the relevant section of the Notice on the notion of State aid and all the guidance around privatisations (the Aviation Guidelines of the 1990’s and the XIV EC Annual Report on Competition. I have not seen any specific reference to this framework. Very best Michalis Papadakis +447900167276

  66. von Phedon Nicolaides

    Thank you for your comment. Yes, there must be some difference. I am surprised too that the Commission did not investigate further whether the Polish authorities were acting beyond the boundaries of their regulatory tasks.

  67. von Phedon Nicolaides

    Your doubts are well founded. Regardless of the Commission’s doubts, the significance of this case is that it addresses the issue of whether a non-undertaking my charge a price [yes] and how much of the costs that price should cover [less than 50%]. This is useful to such entities as public hospitals, public schools, museums, libraries, etc.


    I am not convinced that for a private entity the fact of more or less than 50% public funding in the enterprise is definitive for the purpose of being defined as an “Undertaking” or not. An undertaking for State aid purposes is an entity carrying out economic activity, which in turn is defined as placing a product or service on a given market for a remuneration. In this case the remuneration comes in two parts, one from the State and one from the students. Given that the music school is independent of the state, when it negotiated the fee from the State for the education service it provides, it would only be able to accept a combination of public funds and student fees that would allow it to survive economically, and therefore this amounts to a “consideration” for the service. The fact that a private competitor raised a complaint, and the fact that the Commission could not conclude that trade between Member States was not affected is ample evidence of participating in a given market. The Commission seems to base its reliance for a finding of a non-undertaking in the field of education on the ECJ case-law. However, as the Commission acknowledges at (29) in its decision on this case, the ECJ was referring to education being provided by the State itself where the contributions of beneficiaries is in the minority. It could be inferred that the ECJ is considering a very different commercial situation to the one where the provider is independent of the State, and can not call on State resources to ensure that it breaks even. This case law might not be strictly comparable because of this difference.

  69. von Pierpaolo Rossi

    In this case, both the Commission and the General Court have misapplied the notion of aid scheme within the meaning of Article 1d of Reg 2015/1589, which does not refer in any way to discretion. This is not new, because the national tax rulings practices are still fundamentally misunderstood. Any such national practice corresponds to a tax scheme by which a taxpayer (generally a multinational group which is confronted with the problem of international double taxation), without need of further implementing measures, asks the tax administration to have access to special rules, which are also general, to define the tax base in a way that eliminates or reduces international double taxation. This is what even the General Court has misunderstood. Discretion is the general administrative discretion that any tax administration needs to have when applying such general rules. Administrative discretion is fully irrelevant provided that the administration acts within its general prerogatives, and cannot turn an aid measure into “individual aid”.

  70. von Phedon Nicolaides

    “This is an interesting observation. If the Commission appeals or when the Commission re-opens the cases, it will hopefully clarify the difference between general discretion and discretion which confers a selective advantage. More broadly, when general policy principles have to be applied objectively, governments assign their implementation to independent institutions which ensure impartiality and equal treatment [e.g. sectoral regulators, competition authorities, etc]. This is because it is very difficult to define ex ante rules that ensure impartial and objective implementation. In fact EU courts have said in multiple judgments that when an institution exercises decision-making discretion, it is important that it follows strict procedures. Procedural discipline is the safeguard to administrative discretion. Therefore, is it possible to define such rules or procedural safeguards with respect to the application of advance tax rulings? Is it possible to identify indicators of objective exercise of general administrative discretion? Perhaps the Commission will enlighten us either in its appeal or new decisions.”

  71. von Miguel Muñoz

    It is interesting to read AG Kokott Opinion in case C-74/16, Congregación de Escuelas Pías Provincia Betania, par 53-60, that also deal with the idea of “non-relevant” economic activity and proposes not to consider relevant an economic activity below 10% of the total activity. The ECJ did not explicitly address this issue in the final judgement.

  72. von Phedon Nicolaides

    It is perhaps time for the Court of Justice to “invent” the concept of ‘de minimis economic activity’. Sometimes strict doctrine does not make for sensible policy. Economic activities that are very small in relation to the main functions of a non-economic entity should not change it from a non-undertaking to an undertaking. A concept of de minimis economic activity will reflect today’s evolving economy, will bring legal certainty and will reduce administrative burden significantly. It follows naturally from the Commission’s best idea of the past decade: big on big, small on small.

  73. von Melvyn Waumsley

    Thank you for an excellent and stimulating analysis. I think your last point is the critical point, and one which I think will not be resolved by one or more judgements at the ECJ whilst the EU Treaty remains as it is. How do you fully develop and defend a single market without a common fiscal policy – i.e. federal Europe? The same might also be said about defending the Euro on the international markets, but let’s not go there!

  74. von Pierpaolo Rossi

    The General Court seems to elude some relevant considerations in what should be its analysis of the justification of the apparent selectivity of 1) the sectoral tax on the retail sector, in Poland and 2) the system of progressive rates for that tax, based on the objective of that tax. Any justification for the said two selective elements should be inherently consistent to the logic of the tax system in question. 1) Firstly, what was the logic of that sectoral tax on retailers? The first observation to make is that that tax is not a turnover tax applied on the price of the sales and borne by the customers (indirect tax, or consumption tax) but a personal tax on retailers (direct tax on their ability to pay). The question is then why only the retail sector should be hit by tax, while other economic sectors are excluded. If the logic of the tax system is to raise revenues and the ability to pay, other economic sectors (manufacturing) may have that same ability and should not be excluded from a tax on their turnovers (a breach of horizontal tax neutrality). 2) Second, the differentiation between rates maybe explained in terms of their lose correlation with (in)ability of the retailers with lower turnovers, but they are surely inconsistent with the sectoral nature of the tax. Retailers are effectively hit by the tax because of the aim of the PL tax system to charge a sector which is arguably viewed as less labour intensive (like manufacturing) than retail. In this respect, large surfaces with large turnover which comparatively employ more people are unjustifiably hit by a higher personal charge than smaller retailers. The question is then why only the large retailers should be hit by tax, while smaller retailers are exempted (a breach of vertical tax neutrality). The explanation for this otherwise illogic distinction can only be connected to the fact that the large retail distribution are mainly foreign, and accordingly the tax is design to de-facto restrict the establishment of large foreign retailers. In sum, with its judgement the General Court failed to adequately assess the nature of the tax in question, because: (i) the tax on the turnover of the retailer sector is a personal tax that is only apply on certain economic activities and therefore selective, while construed as a general tax (contribute to tax revenues and redistribute charges on the basis of ability to pay), and (ii) the tax applies mostly on undertakings that are more capital and labour intensive, without justification if not that of de-facto restricting the establishment of foreign groups which are likely to be large distribution chains and favour local shop-owners.

  75. von DR. José Antonio Rodríguez Miguez

    In this new Nicolaides’ post analyses the essential , but very unknown, role pf national Courts both, in Public and Private Enforcement, front ne conclusions of a Study on the enforcement rules and decisions of State aid by national courts, prepared by a consortium of academic and consultancy organisations. In a few minutes, we all can discover on database, the importance and potentially on this particularly necessary role. Congratulations and, once more again, thanks very much for your new contribution to the knowledge of the application of EU State Aid Rules Professor Nicolaides.

  76. Many thanks for signalling this and the quick recap. One thing that I missed in the report is the (increasing?) tendency of public authorities to rely on private enforcement or at least be involved in private enforcement to be freed from the state aid. In the Netherlands, this has happened several times now (with Residex being the case that made it to the ECJ). It seems like the authorities are also using state aid supervision to escape from aid where they have second thoughts.

  77. von Phedon Nicolaides

    Thank you for your comments! As more competitors or aid recipients resort to national courts, I expect that granting authorities will “discover” that state aid rules do not just constrain them, but can also save them from their mistakes.

  78. von Melvyn Waumsley

    Thank you again for another clear and erudite analysis. This has been both a learning and a confirming experience. As a regional development agency in a sparsely populated region we are sometimes faced with similar type of scenario concerning our actions as a landlord for industrial premises when tenants get into rent arrears. Our properties will have been built as a public purpose where there is a shortage but the market is not able to provide economically, but rent arrears are dealt with as a market economy creditor. Part of the solution when tenants are experiencing financial difficulties may be a restructuring of rental charges, such as a period or reduced rent and /or phasing payments to allow some business restructuring that improves the prospects of improved future rental prospects giving a market economy creditor out-turn better than insolvency proceedings. This does involve careful analysis of all of the factors of the specific situation, including the role of other creditors and insolvency law, rather like the case in this article where the lender also has financial considerations as a shareholder.

  79. Dear Phedon, many thanks for your observations. You write that: Therefore, the Court of Justice instead of using vague concepts such “connected, by their nature, their aim and the rules to which they are subject” and agonising about “separability”, it could have simply ruled that when an activity is necessary for and proportional to the exercise of officials powers or non-economic activities, the whole package is non-economic. This, however, raises the question with me whether and to what extent using concepts like ‘necessary’ and ‘proportionate’ are any clearer, simpler or easier to administrate compared to the Court’s – admittedly – vague terminology? Furthermore, if I understand the judgment correctly, the reasoning turns on the factual assessments set out in para. 43 (to which para. 45 refers). This in turn refers to the factual assessments by the GC set out in paras 96 and 51 of the GC judgment. These paragraphs don’t really add to the reasoning, but the paragraphs leading to para 51 of the GC judgment refer to the design choice by the Netherlands government to design TenderNed as an integrated application with two modules to comply with the public procurement directives. I’m not sure how the design at one stage of a system can determine the seperabilty and any later stage. Let me use one example. It is obvious that the organisation and approval of racing activities in Greece have long been in the hands of ELPA. That was probably a well-thought out design choice by the Greek authorities. Still, that did not keep the Court from ruling that the activities of organisation and approval can be separated without any problems, as it did in MOTOE. Perhaps, the Court’s deference in this case follows from the fact that we’re dealing with the concept of an economic activity and therefore the scope of the competition rules whereas MOTOE dealt with the activities that were firmly in the scope of the competition rules? If so, I do not see why there should be greater deference for MS choices on whether there is a market compared to the choices a MS makes within a market. Both are questions of EU law where the Court has jurisdiction. I’d love to hear your thoughts.

  80. von Phedon Nicolaides

    Thank you for your comment and please apologise for the late reply. I agree that the private creditor test may be applicable in this case, as a private creditor may indeed restructure rents to avoid more costly litigation and eviction procedures.

  81. von Phedon Nicolaides

    Thank you very much for your comment and please apologise the late reply. I share your concerns. I am not sure I have a better answer. But let me try to express my view succinctly and concisely. The crux of the issue is how to give effect to a non-market activity [i.e. something that falls within the prerogatives and exclusive powers of the state] without the state engaging in a market activity. The state needs to use tools or instruments to implement its decisions that stem from its state prerogatives. The problem is that these tools or instruments in almost all cases one can think of contain activities that are carried out on the market. Therefore, the state has a choice to use a market agent or manage the tools or instruments itself [e.g. to deliver cash from the central bank to commercial banks. Without actually delivery of the money, the issuing of money is useless.] If the state actually delivers itself, this is necessary to give effect to its prerogative of issuing money. If it is limited to the delivery of money and not delivery of pizzas, then it is also proportional. If you take any other view, then you enter into an interminable discussion on the limits of the state. For example, should the state only regulate rather than actually manage air traffic control? Should the state regulate rather than actually run prisons?

  82. von Caroline Wehlander

    I would really look forward reading professor Nicolaides’ view on the opinion of the General Advocate in the Dovera case.

  83. von Phedon Nicolaides

    I have read the opinion but I do not normally comment on opinions. As with all AG opinions, it provides an educational review of the case law. But I am not sure it offers a test that can be applied in other situations to determine whether what we may call “residual” competition can turn a non-economic activity into economic. It should not be surprising that some residual competition existed in that health insurance system. This is because I have never observed a non-economic activity which is totally devoid of some form competition [above minimum required standards] among those who carry it out. [e.g. teacher or nurse of the year awards, research or book prizes, etc] So there is always some competition, some emulation, some performance comparison. But how much is too much?

  84. von Caroline Wehlander

    Thank you so much for this comment. I found the opinion confusing in the sense that it seems to suggest that even the for-profit activity of privately-owned companies would be non-economic…

  85. von Pierpaolo Rossi

    I disagree with the comment that the notion that progressive turnover taxes being linked to ability to pay is incorrect. The ability to pay is the same that is found in other indirect taxes = its the customers’ ability to pay. I find that state-aid commentators tend to quickly equalise indirect taxes to direct taxes and this is in my view a source of misunderstandings. On a different note, the judgments are in my view stressing again the importance of the distinction between the notions of advantage and selectivity when reviewing tax measures, which seems to be lost in the recent practice. Advantage is a first step in determining whether a tax measure falls within the scope of State Aid review. When dealing with assigned taxes, the entire tax can be State Aid if it is indeed assigned to certain beneficiaries. When dealing with exemptions or reductions, the advantage is such if it relieves from taxes otherwise borne by the beneficiary. That analysis is not comparative as many believe but focuses on what the same beneficiary would normally be charged as tax (irrespective of comparable undertakings). If this is the case, the selectivity analysis may follow to verify if comparable undertakings get the same advantage and if not whether this is justified by the nature and scope of the tax system. In my view, the judgements are a useful reminder of how the notion of advantage and therefore of tax measures falling or not within the scope of application of state aid control should be applied. I wonder if the same rigorous approach has been used when examining the individual rulings as separate measures instead of general schemes. What is advantage there and why is it necessary to find that not is selective if individual rulings are just individual measures?

  86. von Carlos Oliveira

    I wonder whether a “credit moratorium ” can be used in the context of this Temporary framework. Or if such a measure could constitute “no aid” if it had a general scope, being applicable to all companies of all sectors of the economy. However, as a general measure like this would certainly harm creditors I cannot see how it can be seen as “no aid”(additional provisions might be necessary, such as the provision of guarantees to lenders). There are also calls from ultraperipheral regions for the application of more generous measures that consider their specific situation (already recognized in article 349 TFEU). I believe that the adoption of special measures to these regions cannot be done under the Temporary Framework (which does not apply to specific regions of a Member State) but rather using existing aid instruments.

  87. von Julian Jarrett

    My reading of the temporary framework is that it is only directed at member state governments and doesn’t provide any guidance, for instance, to municipalities who may be considering aid measures to individual local businesses (SME and non-SME) to reduce the impact from COVID-19. Advising local authorities in the UK, which has no notifiable schemes under the temporary framework, I am struggling to say with confidence that they can rely upon an 800,000 euro limit on COVID-19 related aid if there is no broader scheme which has been notified to the Commission. Is this your reading also? Or do you think it reads that the temporary framework rules if complied with may be just reported on SANI2 and that is the best you can do as a municipality (where no scheme is relevant).

  88. von Phedon Nicolaides

    Thank you for cour comments. @Julian Jarret: State aid rules are normally addressed to Member States. This does not mean that sub-national authorities are not allowed to grant aid. On the basis of the information required by the Commission and according to the approved measures so far, sub-national authorities are also allowed to grant aid to redress disruption caused by covid-19. However, it is never allowed to grant aid twice for the same costs. Therefore, a sub-national authority that wants to grant aid must ensure that the same company has not received similar aid for the same costs from the central government.

  89. von Phedon Nicolaides

    @ Carlos Oliveira: Good questions. A credit moratorium would not be state aid not only because would be general but also because, presumably, it would not involve transfer of state resources if it would simply suspend the repayment of loans. With respect to Art 349 regions, the MS concerned can already grant operating aid under the regional aid rules. But nothing prevents them from applying in those regions too covid-19-related measures.

  90. von Phedon Nicolaides

    Thank you for your comment! Indirect taxes support specific policy objectives such as discouraging consumption. The general indirect tax of VAT aims to raise revenue from consumption but the burden is borne by consumers, not companies. At any rate, it has long been established that VAT is a regressive tax. The important points here are the following two. First, the Member States that imposed the turnover taxes in specific sectors claimed that they did that in order to raise revenue according to ability to pay. EU courts have acknowledged that that is the prerogative of MS [although I wonder if the objective is to raise revenue why limit the tax to specific sectors. I suppose MS are free to “mess up” their economies]. Therefore, we need to determine the selectivity of that tax in relation to the objective it sets for itself. Second, EU courts have further acknowledged that due to economies of scale, larger turnover correlates with higher ability to pay. This is what they got wrong. The tax does not treat all companies equally, according to its own objective.

  91. von Benjamin Moha

    What should European Commission do with decisions that demand recovery of aid, which has already made, but the recovery deadline is somewhere april or during the crisis? I think Commission should suspend these decisions, because it affects these companies very badly. One problem right now is with Deggendorf principle, which means that undertakings that should pay illegal state aid back and has not yet done so, cannot rely on new state aid. What do you think about that?

  92. von Phedon Nicolaides

    Thank you for your comment! Recovery of incompatible aid aims to restore the situation before the aid was granted. If an undertaking that received incompatible aid is in a bad shape today, it would have been in an even worse shape without the aid. Despite its difficulties, it is still enjoying an advantage that should not have been granted to it. EU law recognises only one possibility for not pay back incompatible aid: reasonable expectations. The fact that repayment of incompatible may lead to bankruptcy has been accepted by EU courts as the natural consequence of recovery. The Commission is required [by the case law and Reg 2015/1589] to demand the repayment of incompatible aid. It may not ask for repayment only when it would be contrary to a general principle of EU law. So far it appears that the only principle recognised by EU courts is “absolute impossibility”. Therefore, unless the Commission can come up with an ingenious and novel justification, it cannot exempt non-repayment of incompatible aid.

  93. von Alexander Rose

    Thank you Phedon – insightful and directly relevant to the practical issues faced by public bodies distributing funding. Do you think the Commission might revise the definition of Undertaking in Difficulty in the next amendment to the Temporary Framework? It seems to be a sticking point for many Covid-19 measures and therefore the Commission might choose to depart from the definition used in other Regulations.

  94. von Phedon Nicolaides

    Good question. My friends in DG Competition did not mention anything in recent discussions we had. Of course, it does not mean that the Commission is not discussing this issue internally. My guess is that it would be difficult to adjust the definition for a short period. It would be very hard to predict the consequences of such an adjustment. Any change would be of a different nature; e.g. dropping the exclusion of undertakings in difficulty. I am more concerned about the cumulative impact of the state aid on the internal market.

  95. von Carlos Oliveira

    Dear Professor, You state that “the only explicit exclusion in the current Temporary Framework concerns undertakings that were in difficulty before 1 January 2020”. In fact, this provision is repeated in all sections of the Temporary Framework, except sections 3.9, “Aid in form of deferrals of tax and/or of social security contributions” and 3.10 “Aid in form of wage subsidies for employees to avoid lay-offs during the COVID-19”. Does this mean that aid under these two sections can be granted to undertakings in difficulty before January 1,2020? Thank you

  96. von James Jenkins

    So for clarity, an undertaking is in difficulty if it has lost over 50% of: 1) The total amount paid for various shares (e.g. €1m for €1 each and €1m for €5 each (a total of €2m); or 2) The amount last paid for a share * no. of outstanding shares (€6m in the above example)?

  97. von Phedon Nicolaides

    Since Art 2(18) of the GBER refers to “subscribed capital”, the right answer must be EUR 2m (as in your example).

  98. von Phedon Nicolaides

    I suppose your conclusion is what we must infer. I also suppose that the reason is that the aid aims to help employees rather than companies.

  99. von Marcos Araujo

    Interesting to see yet another case where the Court struggles to differentiate the facts from those of Preussen Elektra, despite the fact that there has not been any case since Preussen (over 18 yeas ago) where a State imposed payment has been considered not to entail State resources. One would think that the Court believes that at least some State-imposed obligations should remain outside the scope of State Aid rules, despite the angry winds to the contrary…