19.7.2016   

EN

Official Journal of the European Union

C 262/1

1.

2.

2.1.

2.2.

2.3.

2.4.

2.5.

2.6.

3.

3.1.

3.1.1.

3.1.2.

3.2.

3.2.1.

3.2.2.

3.2.3.

4.

4.1.

4.2.

4.2.1.

4.2.2.

4.2.3.

4.3.

5.

5.1.

5.2.

5.2.1.

5.2.2.

5.2.3.

5.3.

5.3.1.

5.3.2.

5.3.3.

5.4.

5.4.1.

5.4.2.

5.4.3.

5.4.4.

5.4.5.

5.4.6.

5.4.7.

5.4.8.

6.

6.1.

6.2.

6.3.

7.

7.1.

7.2.

7.2.1.

7.2.2.

7.2.3.

7.3.

7.4.

8.

1.

In the context of the State aid modernisation, the Commission wishes to provide further clarification on the key concepts relating to the notion of State aid as referred to in Article 107(1) of the Treaty on the Functioning of the European Union, with a view to contributing to an easier, more transparent and more consistent application of this notion across the Union.

2.

This Notice only concerns the notion of State aid as referred to in Article 107(1) of the Treaty, which both the Commission and national authorities (including national courts) have to apply in conjunction with the notification and standstill obligations provided for in Article 108(3) of the Treaty. It does not concern the compatibility of State aid with the internal market pursuant to Article 107(2) and (3) and Article 106(2) of the Treaty, which is for the Commission to assess.

3.

Given that the notion of State aid is an objective and legal concept defined directly by the Treaty, (1) this Notice clarifies the Commission's understanding of Article 107(1) of the Treaty, as interpreted by the Court of Justice and the General Court (‘the Union Courts’). On issues that have not yet been considered by the Union Courts, the Commission will set out how it considers that the notion of State aid should be construed. The views set out in this Notice are without prejudice to the interpretation of the notion of State aid by the Union Courts (2); the primary reference for interpreting the Treaty is always the case–law of the Union Courts.

4.

It should be stressed that the Commission is bound by this objective notion and enjoys only a limited margin of discretion in applying it, namely where the appraisals by the Commission are technical or complex in nature, in particular in situations involving complex economic assessments. (3)

5.

Article 107(1) of the Treaty defines State aid as ‘any aid granted by a Member State or through State resources in any form whatsoever which distorts or threatens to distort competition by favouring certain undertakings or the production of certain goods […], in so far as it affects trade between Member States’ (4). This Notice will clarify the different constituent elements of the notion of State aid: the existence of an undertaking, the imputability of the measure to the State, its financing through State resources, the granting of an advantage, the selectivity of the measure and its effect on competition and trade between Member States. In addition, given the need for specific guidance expressed by Member States, this Notice provides specific clarification with respect to public funding of infrastructure.

6.

The State aid rules only apply where the beneficiary of a measure is an ‘undertaking’.

7.

The Court of Justice has consistently defined undertakings as entities engaged in an economic activity, regardless of their legal status and the way in which they are financed. (5) The classification of a particular entity as an undertaking thus depends entirely on the nature of its activities. This general principle has three important consequences.

8.

First, the status of the entity under national law is not decisive. For example, an entity that is classified as an association or a sports club under national law may nevertheless have to be regarded as an undertaking within the meaning of Article 107(1) of the Treaty. The same applies to an entity that is formally part of the public administration. The only relevant criterion is whether it carries out an economic activity.

9.

Second, the application of the State aid rules does not depend on whether the entity is set up to generate profits. Non-profit entities can also offer goods and services on a market. (6) Where this is not the case, non-profit entities remain outside the scope of State aid control.

10.

Third, the classification of an entity as an undertaking is always relative to a specific activity. An entity that carries out both economic and non-economic activities is to be regarded as an undertaking only with regard to the former. (7)

11.

Several separate legal entities may be considered to form one economic unit for the purposes of the application of State aid rules. That economic unit is then considered to be the relevant undertaking. In this respect, the Court of Justice considers the existence of a controlling share and other functional, economic and organic links to be relevant. (8)

12.

To clarify the distinction between economic and non-economic activities, the Court of Justice has consistently held that any activity consisting in offering goods and services on a market is an economic activity. (9)

13.

The question whether a market exists for certain services may depend on the way those services are organised in the Member State concerned (10) and may thus vary from one Member State to another. Moreover, due to political choice or economic developments, the classification of a given activity can change over time. What is not an economic activity today may become one in the future, and vice versa.

14.

The decision of a public authority not to allow third parties to provide a certain service (for example, because it wishes to provide the service in-house) does not rule out the existence of an economic activity. In spite of such market closure, an economic activity can exist where other operators would be willing and able to provide the service in the market concerned. More generally, the fact that a particular service is provided in-house has no relevance for the economic nature of the activity. (11)

15.

Since the distinction between economic and non-economic activities depends to some extent on political choices and economic developments in a given Member State, it is not possible to draw up an exhaustive list of activities that a priori would never be economic. Such a list would not provide genuine legal certainty and would thus be of little use. Paragraphs 17 to 37 instead seek to clarify the distinction with respect to a number of important areas.

16.

The simple fact that an entity holds shares, even a majority shareholding, in an undertaking providing goods or services on a market does not mean that that entity should automatically be considered an undertaking for the purposes of Article 107(1) of the Treaty. Where that shareholding only gives rise to the exercise of rights attached to the status of shareholder as well as, if appropriate, the receipt of dividends, which are merely the fruits of the ownership of an asset, that entity will not be considered an undertaking if it does not itself provide goods or services on a market. (12)

17.

Article 107(1) of the Treaty does not apply where the State acts ‘by exercising public power’ (13) or where public entities act ‘in their capacity as public authorities’. (14) An entity may be deemed to act by exercising public power where the activity in question forms part of the essential functions of the State or is connected with those functions by its nature, its aim and the rules to which it is subject. (15) Generally speaking, unless the Member State concerned has decided to introduce market mechanisms, activities that intrinsically form part of the prerogatives of official authority and are performed by the State do not constitute economic activities. Examples of such activities are the following:

18.

In so far as a public entity exercises an economic activity which can be separated from the exercise of public powers, that entity acts as an undertaking in relation to that activity. In contrast, if that economic activity cannot be separated from the exercise of public powers, the activities exercised by that entity as a whole remain connected with the exercise of those public powers and therefore fall outside the notion of undertaking. (23)

19.

Whether schemes in the area of social security are to be classified as involving an economic activity depends on the way they are set up and structured. In essence, the case-law distinguishes between schemes based on the principle of solidarity and economic schemes.

20.

Solidarity-based social security schemes that do not involve an economic activity typically have the following characteristics:

21.

Such solidarity-based schemes must be distinguished from schemes that involve an economic activity. (30) The latter are regularly characterised by:

22.

Some schemes combine features of both categories. In such cases, the classification of the scheme depends on an analysis of different elements and their respective importance. (35)

23.

In the Union, health care systems differ significantly between Member States. Whether and to what degree different health care providers compete with each other depends on these national specificities.

24.

In some Member States, public hospitals are an integral part of a national health service and are almost entirely based on the principle of solidarity. (36) Such hospitals are directly funded from social security contributions and other State resources and provide their services free of charge on the basis of universal coverage. (37) The Union Courts have confirmed that, where such a structure exists, the relevant organisations do not act as undertakings. (38)

25.

Where that structure exists, even activities that in themselves could be of an economic nature, but are carried out merely for the purpose of providing another non-economic service, are not of an economic nature. An organisation that purchases goods — even in large quantities — for the purpose of offering a non-economic service does not act as an undertaking simply because it is a purchaser in a given market. (39)

26.

In many other Member States, hospitals and other health care providers offer their services for remuneration, be it directly from patients or from their insurance. (40) In such systems, there is a certain degree of competition between hospitals concerning the provision of health care services. Where this is the case, the fact that a health service is provided by a public hospital is not sufficient for the activity to be classified as non-economic.

27.

The Union Courts have also clarified that health care services which independent doctors and other private practitioners provide for remuneration at their own risk are to be regarded as an economic activity. (41) The same principles apply to pharmacies.

28.

Public education organised within the national educational system funded and supervised by the State may be considered as a non-economic activity. The Court of Justice held that the State: ‘by establishing and maintaining such a system of public education and financed entirely or mainly by public funds and not by pupils or their parents […] does not intend to become involved in activities for remuneration, but carries out its task towards its population in the social, cultural and educational areas’. (42)

29.

The non-economic nature of public education is in principle not affected by the fact that pupils or their parents sometimes have to pay tuition or enrolment fees which contribute to the operating expenses of the system. Such financial contributions often only cover a fraction of the true costs of the service and can thus not be considered as remuneration for the service provided. They therefore do not alter the non-economic nature of a general education service predominantly funded by the public purse. (43) These principles can cover public educational services such as vocational training, (44) private and public primary schools (45) and kindergartens, (46) secondary teaching activities in universities (47) and the provision of education in universities. (48)

30.

Such public education services must be distinguished from services financed predominantly by parents or pupils or commercial revenues. For example, higher education financed entirely by students clearly fall within the latter category. In certain Member States public entities can also offer educational services which, due to their nature, financing structure and the existence of competing private organisations, are to be regarded as economic.

31.

In the light of the principles set out in paragraphs 28, 29 and 30, the Commission considers that certain activities of universities and research organisations fall outside the scope of the State aid rules. This concerns their primary activities, namely:

32.

The Commission considers that knowledge transfer activities (licensing, creation of spin-off, or other forms of management of knowledge created by the research organisation or infrastructure) are non-economic where they are conducted either by the research organisation or research infrastructure (including their departments or subsidiaries) or jointly with, or on behalf of other such entities, and all income from those activities is reinvested in the primary activities of the research organisations or infrastructures concerned. (49)

33.

Culture is a vehicle of identities, values and meanings that mirror and shape the Union's societies. The area of culture and heritage conservation covers a vast array of purposes and activities, inter alia, museums, archives, libraries, artistic and cultural centres or spaces, theatres, opera houses, concert halls, archaeological sites, monuments, historical sites and buildings, traditional customs and crafts, festivals and exhibitions, as well as cultural and artistic education activities. Europe's rich natural heritage, including conservation of biodiversity, habitats and species further provides valuable benefits for societies in the Union.

34.

Taking into account their particular nature, certain activities related to culture, heritage and nature conservation may be organised in a non-commercial way and thus be non-economic in nature. Public funding thereof may therefore not constitute State aid. The Commission considers that public funding of a cultural or heritage conservation activity accessible to the general public free of charge fulfils a purely social and cultural purpose which is non-economic in nature. In the same vein, the fact that visitors of a cultural institution or participants in a cultural or heritage conservation activity, including nature conservation, open to the general public are required to pay a monetary contribution that only covers a fraction of the true costs does not alter the non-economic nature of that activity, as it cannot be considered genuine remuneration for the service provided.

35.

In contrast, cultural or heritage conservation activities (including nature conservation) predominantly financed by visitor or user fees or by other commercial means (for example, commercial exhibitions, cinemas, commercial music performances and festivals and arts schools predominantly financed from tuition fees) should be qualified as economic in nature. Similarly, heritage conservation or cultural activities benefitting exclusively certain undertakings rather than the general public (for example, the restoration of a historical building used by a private company) should normally be qualified as economic in nature.

36.

Moreover, many cultural or heritage conservation activities are objectively non-substitutable (for example, keeping public archives holding unique documents) and thus exclude the existence of a genuine market. In the Commission's view, such activities would also qualify as non-economic in nature.

37.

In cases where an entity carries out cultural or heritage conservation activities, some of which are non-economic activities as set out in paragraphs 34 and 36 and some of which are economic activities, public funding it receives will fall under the State aid rules only insofar as it covers the costs linked to the economic activities. (50)

38.

The granting of an advantage directly or indirectly through State resources and the imputability of such a measure to the State are two separate and cumulative conditions for State aid to exist. (51) However, they are often considered together when assessing a measure under Article 107(1) of the Treaty, as they both relate to the public origin of the measure in question.

39.

In cases where a public authority grants an advantage to a beneficiary, the measure is by definition imputable to the State, even if the authority in question enjoys legal autonomy from other public authorities. The same applies if a public authority designates a private or public body to administer a measure conferring an advantage. Indeed, Union law cannot permit the rules on State aid to be circumvented through the creation of autonomous institutions charged with allocating aid. (52)

40.

Imputability is less evident, however, if the advantage is granted through public undertakings. (53) In such cases, it is necessary to determine whether the public authorities can be regarded as having been involved, in one way or another, in adopting the measure. (54)

41.

The mere fact that a measure is taken by a public undertaking is not per se sufficient to consider it imputable to the State. (55) However, it does not need to be demonstrated that, in a particular case, the public authorities specifically incited the public undertaking to take the measure in question. (56) In fact, since relations between the State and public undertakings are necessarily close, there is a real risk that State aid may be granted through the intermediary of those undertakings in a non-transparent manner and in breach of the rules on State aid laid down by the Treaty. (57) Moreover, precisely because of the privileged relations that exist between the State and public undertakings, it will, as a general rule, be very difficult for a third party to demonstrate that measures taken by such an undertaking were in fact adopted on the instructions of the public authorities in a particular case. (58)

42.

For these reasons, the imputability to the State of a measure taken by a public undertaking may be inferred from a set of indicators arising from the circumstances of the case and the context in which the measure was taken. (59)

43.

Possible indicators to establish whether a measure is imputable include the following: (60)

44.

A measure is not imputable to a Member State if the Member State is under an obligation to implement it under Union law without any discretion. In that case, the measure stems from an act of the Union legislature and is not imputable to the State. (64)

45.

However, this is not the case in situations where Union law simply allows for certain national measures and the Member State enjoys discretion (i) as to whether to adopt the measures in question or (ii) in establishing the characteristics of the concrete measure which are relevant from a State aid perspective. (65)

46.

Measures that are adopted jointly by several Member States are imputable to all the Member States concerned pursuant to Article 107(1) of the Treaty. (66)

47.

Only advantages granted directly or indirectly through State resources can constitute State aid within the meaning of Article 107(1) of the Treaty. (67)

48.

State resources include all resources of the public sector, (68) including resources of intra-State entities (decentralised, federated, regional or other) (69) and, under certain circumstances, resources of private bodies (see paragraphs 57 and 58). It is irrelevant whether or not an institution within the public sector is autonomous. (70) Funds provided by the central bank of a Member State to specific credit institutions generally imply the transfer of State resources. (71)

49.

Resources of public undertakings also constitute State resources within the meaning of Article 107(1) of the Treaty because the State is capable of directing the use of these resources. (72) For the purposes of State aid law, transfers within a public group may also constitute State aid if, for example, resources are transferred from the parent company to its subsidiary (even if they constitute a single undertaking from an economic point of view). (73) The question of whether the transfer of such resources is imputable to the State is addressed in section 3.1. The fact that a public undertaking is a beneficiary of an aid measure does not mean it may not grant aid to another beneficiary by way of a different aid measure. (74)

50.

The fact that a measure granting an advantage is not financed directly by the State, but by a public or private body established or appointed by the State to administer the aid, does not necessarily mean that the measure is not financed through State resources. (75) A measure adopted by a public authority and favouring certain undertakings or products does not lose the character of a gratuitous advantage by virtue of the fact that it is wholly or partially financed by contributions imposed by the public authority and levied on the undertakings concerned. (76)

51.

The transfer of State resources may take many forms, such as direct grants, loans, guarantees, direct investment in the capital of companies and benefits in kind. A firm and concrete commitment to make State resources available at a later point in time is also considered a transfer of State resources. A positive transfer of funds does not have to occur; foregoing State revenue is sufficient. Waiving revenue which would otherwise have been paid to the State constitutes a transfer of State resources. (77) For example, a ‘shortfall’ in tax and social security revenue due to exemptions or reductions in taxes or social security contributions granted by the Member State, or exemptions from the obligation to pay fines or other pecuniary penalties, fulfils the State resources requirement of Article 107(1) of the Treaty. (78) The creation of a concrete risk of imposing an additional burden on the State in the future, by a guarantee or by a contractual offer, is sufficient for the purposes of Article 107(1). (79)

52.

If public authorities or public undertakings provide goods or services at a price below market rates, or invest in an undertaking in a manner that is inconsistent with the market economy operator test, as described from paragraph 73 onwards, this implies foregoing State resources (as well as the granting of an advantage).

53.

Granting access to a public domain or natural resources, or granting special or exclusive rights (80) without adequate remuneration in line with market rates, can constitute foregoing State revenues (as well as the granting of an advantage). (81)

54.

In these cases it needs to be established whether the State, in addition to its role of manager of the public assets in question, acts as a regulator that pursues policy objectives by making the selection process of the undertakings concerned subject to qualitative criteria (established ex ante in a transparent and non-discriminatory manner). (82) When the State acts as a regulator, it can decide legitimately not to maximise the revenues which could otherwise have been achieved without falling under the scope of State aid rules, provided that all the operators concerned are treated in line with the principle of non-discrimination, and that there is an inherent link between achieving the regulatory purpose and the foregoing of revenue. (83)

55.

In any event, a transfer of State resources is present if, in a given case, the public authorities do not charge the normal amount under their general system for access to the public domain or natural resources, or for granting certain special or exclusive rights.

56.

A negative indirect effect on State revenues stemming from regulatory measures does not constitute a transfer of State resources where it is an inherent feature of the measure. (84) For example, a derogation from employment law provisions altering the framework for contractual relations between undertakings and employees does not constitute a transfer of State resources, despite the fact that it may reduce social security contributions or taxes payable to the State. (85) Similarly, national regulation which sets a minimum price for certain goods does not entail the transfer of State resources. (86)

57.

The origin of the resources is not relevant provided that, before being directly or indirectly transferred to the beneficiaries, they come under public control and are therefore available to the national authorities, (87) even if the resources do not become the property of the public authority. (88)

58.

Thus, subsidies financed through parafiscal charges or compulsory contributions imposed by the State and managed and apportioned in accordance with the provisions of public rules imply a transfer of State resources, even if not administered by the public authorities. (89) Moreover, the mere fact that the subsidies are financed in part by voluntary private contributions is not sufficient to rule out the presence of State resources, since the relevant factor is not the origin of the resources but the degree of intervention of the public authority within the definition of the measure and its method of financing. (90) The transfer of State resources can only be ruled out in very specific circumstances, notably if resources from the members of a trade association are earmarked for funding a specific purpose in the interest of the members, are decided on by a private organisation and have a purely commercial purpose, and if the Member State is simply acting as a vehicle in order to make the contribution introduced by the trade organisation compulsory. (91)

59.

A transfer of State resources is also present if the resources are at the joint disposal of several Member States who decide jointly on the use of those resources. (92) This would be the case, for example, for funds from the European Stability Mechanism (ESM).

60.

Resources coming from the Union (for example from structural funds), from the European Investment Bank or the European Investment Fund, or from international financial institutions, such as the International Monetary Fund or the European Bank for Reconstruction and Development, are considered as State resources if national authorities have discretion as to the use of these resources (in particular the selection of beneficiaries). (93) By contrast, if such resources are awarded directly by the Union, by the European Investment Bank or by the European Investment Fund, with no discretion on the part of the national authorities, they do not constitute State resources (for example funding awarded in direct management under the Horizon 2020 framework programme, the EU programme for the Competitiveness of Enterprises and Small and Medium-sized Enterprises (COSME) or the Trans-European Transport Network (TEN-T) funds).

61.

Regulation that leads to financial redistribution from one private entity to another without any further involvement of the State does not, in principle, entail a transfer of State resources if the money flows directly from one private entity to another, without passing through a public or private body designated by the State to administer the transfer. (94)

62.

For example, an obligation imposed by a Member State on private electricity suppliers to purchase electricity produced from renewable energy sources at fixed minimum prices does not entail the direct or indirect transfer of State resources to undertakings which produce that type of electricity. (95) In this case, the undertakings concerned (that is to say the private electricity suppliers) are not appointed by the State to manage an aid scheme, but are only bound by an obligation to purchase a specific type of electricity with their own financial resources.

63.

However, a transfer of State resources is present where the charges paid by private persons transit through a public or private entity designated to channel them to the beneficiaries.

64.

For example, this is the case even where a private entity is appointed by law to collect such charges on behalf of the State and to channel them to the beneficiaries, but is not allowed to use the proceeds from the charges for purposes other than those provided for by the law. In this case, the sums in question remain under public control and are therefore available to the national authorities, which is sufficient reason for them to be considered State resources. (96) Since this principle applies both to public bodies and private entities appointed to collect the charges and process the payments, changing the status of the intermediary from a public to a private entity has no relevance for the State resources criterion if the State continues to strictly monitor that entity. (97)

65.

Moreover, a mechanism for offsetting in full the additional costs imposed on undertakings because of an obligation to purchase a product from certain providers at a price higher than the market price that is financed by all final consumers of the said product also constitutes an intervention through State resources, even when this mechanism is partly based on a direct transfer of resources between private entities. (98)

66.

An advantage, within the meaning of Article 107(1) of the Treaty, is any economic benefit which an undertaking could not have obtained under normal market conditions, that is to say in the absence of State intervention. (99) Section 4.2 of this Communication provides detailed guidance on the question whether a benefit can be considered to be obtained under normal market conditions.

67.

Only the effect of the measure on the undertaking is relevant, and not the cause or the objective of the State intervention. (100) Whenever the financial situation of an undertaking is improved as a result of State intervention (101) on terms differing from normal market conditions, an advantage is present. To assess this, the financial situation of the undertaking following the measure should be compared with its financial situation if the measure had not been taken. (102) Since only the effect of the measure on the undertaking matters, it is irrelevant whether the advantage is compulsory for the undertaking in that it could not avoid or refuse it. (103)

68.

The precise form of the measure is also irrelevant in establishing whether it confers an economic advantage on the undertaking. (104) Not only the granting of positive economic advantages is relevant for the notion of State aid, but relief from economic burdens (105) can also constitute an advantage. The latter is a broad category which comprises any mitigation of charges normally included in the budget of an undertaking. (106) This covers all situations in which economic operators are relieved of the inherent costs of their economic activities. (107) For instance, if a Member State pays part of the costs of the employees of a specific undertaking, it relieves that undertaking from costs that are inherent of its economic activities. An advantage also exists where public authorities pay a salary supplement to the workers of a specific undertaking, even if the undertaking was under no legal obligation to pay such a supplement. (108) It also covers situations where some operators do not have to bear costs that other comparable operators normally do under a given legal order, regardless of the non-economic nature of the activity to which the costs relate. (109)

69.

Costs arising from regulatory obligations imposed by the State (110) can in principle be considered to relate to the inherent costs of the economic activity, so that any compensation for these costs confers an advantage on the undertaking. (111) This means that the existence of an advantage is in principle not excluded by the fact that the benefit does not go beyond compensation for a cost stemming from the imposition of a regulatory obligation. The same applies to relief for costs that the undertaking would not have incurred had there been no incentive stemming from the State measure because without this incentive it would have structured its activities differently. (112) The existence of an advantage is also not excluded if a measure compensates charges of a different nature that are unconnected with that measure. (113)

70.

As regards compensation for costs incurred to provide a service of general economic interest, the Court made clear in the Altmark judgment that the granting of an advantage can be excluded if four cumulative conditions are met. (114) First, the recipient undertaking must actually have public service obligations to discharge, and the obligations must be clearly defined. Second, the parameters on the basis of which the compensation is calculated must be established in advance in an objective and transparent manner. Third, the compensation cannot exceed what is necessary to cover all or part of the costs incurred in the discharge of public service obligations, taking into account the relevant receipts and a reasonable profit. Fourth, where the undertaking that is to discharge public service obligations is not chosen following a public procurement procedure to select a tenderer capable of providing these services at the least cost to the community, the level of compensation needed must be determined on the basis of an analysis of the costs which a typical undertaking, well-run and adequately provided with means to meet the public service requirements, would have incurred in discharging those obligations, taking into account the relevant receipts and a reasonable profit for discharging the obligations. The Commission has further elaborated its understanding of these conditions in its Communication on the application of the European Union State aid rules to compensation granted for the provision of services of general economic interest. (115)

71.

The existence of an advantage is excluded in the case of a reimbursement of illegally levied taxes, (116) an obligation for the national authorities to compensate for damage they have caused to certain undertakings (117) or the payment of compensation for an expropriation. (118)

72.

The existence of an advantage is not ruled out by the mere fact that competing undertakings in other Member States are in a more favourable position, (119) because the notion of advantage is based on an analysis of the financial situation of an undertaking in its own legal and factual context with and without the particular measure.

73.

The Union legal order is neutral with regard to the system of property ownership (120) and does not in any way prejudice the right of Member States to act as economic operators. However, when public authorities directly or indirectly carry out economic transactions in any form, (121) they are subject to Union State aid rules.

74.

Economic transactions carried out by public bodies (including public undertakings) do not confer an advantage on its counterpart, and therefore do not constitute aid, if they are carried out in line with normal market conditions. (122) This principle has been developed with regard to different economic transactions. The Union courts have developed the ‘market economy investor principle’ to identify the presence of State aid in cases of public investment (in particular, capital injections): to determine whether a public body's investment constitutes State aid, it is necessary to assess whether, in similar circumstances, a private investor of a comparable size operating in normal conditions of a market economy could have been prompted to make the investment in question. (123) Similarly, the Union courts have developed the ‘private creditor test’ to examine whether debt renegotiations by public creditors involve State aid, comparing the behaviour of a public creditor to that of hypothetical private creditors that find themselves in a similar situation. (124) Finally, the Union courts have developed the ‘private vendor test’ to assess whether a sale carried out by a public body involves State aid, considering whether a private vendor, under normal market conditions, could have obtained the same or a better price. (125)

75.

Those tests are variations of the same basic concept that the behaviour of public bodies should be compared to that of similar private economic operators under normal market conditions to determine whether the economic transactions carried out by such bodies grant an advantage to their counterparts. In this Communication, the Commission will therefore refer, in general terms, to the ‘market economy operator’ (MEO) test as the relevant method to assess whether a range of economic transactions carried out by public bodies take place under normal market conditions and, therefore, whether they involve the granting of an advantage (which would not have occurred in normal market conditions) to their counterparts. The general principles and the relevant criteria for applying the MEO test are set out in sections 4.2.2. and 4.2.3.

76.

The purpose of the MEO test is to assess whether the State has granted an advantage to an undertaking by not acting like a market economy operator with regard to a certain transaction. In that respect, it is not relevant whether the intervention constitutes a rational means for the public bodies to pursue public policy (for example employment) considerations. Similarly, the profitability or unprofitability of the beneficiary is not in itself a decisive indicator for establishing whether or not the economic transaction in question is in line with market conditions. The decisive element is whether the public bodies acted as a market economy operator would have done in a similar situation. If this is not the case, the beneficiary undertaking has received an economic advantage which it would not have obtained under normal market conditions, (126) placing it in a more favourable position compared to that of its competitors. (127)

77.

For the purpose of the MEO test, only the benefits and obligations linked to the role of the State as an economic operator — to the exclusion of those linked to its role as a public authority — are to be taken into account. (128) Indeed, the MEO test is normally not applicable if the State acts as a public authority rather than as an economic operator. For example, if a State intervention is driven by public policy reasons (for instance, for reasons of social or regional development), the State's behaviour, while being rational from a public policy perspective, may at the same time include considerations which a market economy operator would normally not consider. Accordingly, the MEO test should be applied leaving aside all considerations which exclusively relate to a Member State's role as a public authority (for example social, regional or sectoral policy considerations). (129)

78.

Whether a State intervention is in line with market conditions must be examined on an ex-ante basis, having regard to the information available at the time the intervention was decided upon. (130) In fact, any prudent market economy operator would normally carry out its own ex-ante assessment of the strategy and financial prospects of a project, (131) for instance, by means of a business plan. It is not enough to rely on ex-post economic evaluations entailing a retrospective finding that the investment made by the Member State concerned was actually profitable. (132)

79.

If a Member State argues that it acted as a market economy operator it must, where there is doubt, provide evidence showing that the decision to carry out the transaction was taken on the basis of economic evaluations comparable to those which, in similar circumstances, a rational market economy operator (with characteristics similar to those of the public body concerned) would have had carried out to determine the profitability or economic advantages of the transaction. (133)

80.

Whether a transaction is in line with market conditions must be established through a global assessment of the effects of the transaction on the undertaking concerned without considering whether the specific means used to carry out that transaction would be available to market economy operators. For instance, the applicability of the MEO test cannot be ruled out simply because the means employed by the State are fiscal in nature. (134)

81.

In certain cases, several consecutive measures of State intervention may, for the purposes of Article 107(1) of the Treaty, be regarded as a single intervention. This could be the case, in particular, where consecutive interventions are so closely linked to each other, especially having regard to their chronology, their purpose and the circumstances of the undertaking at the time of those interventions, that they are inseparable. (135) For instance, a series of State interventions which take place in relation to the same undertaking in a relatively short period of time, are linked to each other, or were all planned or foreseeable at the time of the first intervention, may be assessed as one intervention. On the other hand, when the later intervention was a result of unforeseen events at the time of the earlier intervention (136) the two measures should normally be assessed separately.

82.

To assess whether certain transactions are in line with market conditions all the relevant circumstances of the particular case should be considered. For instance, there can be exceptional circumstances in which the purchase of goods or services by a public authority, even if carried out at market prices, may not be considered in line with market conditions. (137)

83.

When applying the MEO test, it is useful to distinguish between situations in which the transaction's compliance with market conditions can be directly established through transaction-specific market data and situations in which, due to the absence of such data, the transaction's compliance with market conditions has to be assessed on the basis of other available methods.

84.

A transaction's compliance with market conditions can be directly established through transaction-specific market information in the following situations

85.

In such cases, if the specific market information concerning the transaction shows that it does not comply with market conditions, it would not normally be appropriate to use other assessment methodologies to reach a different conclusion. (138)

86.

When a transaction is carried out under the same terms and conditions (and therefore with the same level of risk and rewards) by public bodies and private operators who are in a comparable situation (a ‘pari passu’ transaction), (139) as may occur in public private partnerships, it can normally be inferred that such a transaction is in line with market conditions. (140) In contrast, if a public body and private operators who are in a comparable situation take part in the same transaction at the same time but under different terms or conditions, this normally indicates that the intervention of the public body is not in line with market conditions. (141)

87.

In particular, to consider a transaction ‘pari passu’, the following criteria should be assessed:

88.

The ‘pari passu’ condition may not be applicable in some cases where the public involvement (in view of its unique nature or magnitude) is such that it could in practice not be replicated by a market economy operator.

89.

If the sale and purchase of assets, goods and services (or other comparable transactions (145)) are carried out following a competitive (146), transparent, non-discriminatory and unconditional tender procedure in line with the principles of the TFEU on public procurement (147) (see paragraphs 90 to 94), it can be presumed that those transactions are in line with market conditions, provided that the appropriate criteria for selecting the buyer or seller as set out in paragraphs 95 and 96 have been used. In contrast, if a Member States decides to provide support, for public policy reasons, to a certain activity and tenders out, for example, the amount of funding provided, such as in the case of support to the production of renewable energy or to the mere availability of electricity generation capacity, this would not fall in the scope of this sub-section (ii). In such a situation a tender can only minimize the amount granted but cannot exclude an advantage.

90.

A tender procedure has to be competitive to allow all interested and qualified bidders to participate in the process.

91.

The procedure has to be transparent to allow all interested tenderers to be equally and duly informed at each stage of the tender procedure. Accessibility of information, sufficient time for interested tenderers, and the clarity of the selection and award criteria are all crucial elements for a transparent selection procedure. A tender has to be sufficiently well-publicised, so that all potential bidders can take note of it. The degree of publicity needed to ensure sufficient publication in a given case depends on the characteristics of the assets, goods and services. Assets, goods and services which may attract bidders operating on a Europe-wide or international scale in view of their high value or other features should be publicised in such a manner as to attract potential bidders operating on a Europe-wide or international scale.

92.

Non-discriminatory treatment of all bidders at all stages of the procedure and objective selection and award criteria specified in advance of the process are indispensable conditions for ensuring that the resulting transaction is in line with market conditions. To guarantee equal treatment, the criteria for the award of the contract should enable tenders to be compared and assessed objectively.

93.

Using and complying with the procedures provided for in the Public Procurement Directives (148) can be considered sufficient to meet the requirements above provided that all the conditions for the use of the respective procedure are fulfilled. This does not apply in specific circumstances that make it impossible to establish a market price, such as the use of the negotiated procedure without publication of a contract notice. If only one bid is submitted, the procedure would not normally be sufficient to ensure a market price, unless either (i) there are particularly strong safeguards in the design of the procedure ensuring genuine and effective competition and it is not apparent that only one operator is realistically able to submit a credible bid or (ii) the public authorities verify through additional means that the outcome corresponds to the market price.

94.

A tender for the sale of assets, goods or services is unconditional when a potential buyer is generally free to acquire the assets, goods or services to be sold and to use them for its own purposes irrespective of whether or not it runs certain businesses. If there is a condition that the buyer is to assume special obligations for the benefit of the public authorities or in the general public interest, which a private seller would not have demanded — other than those arising from general domestic law or a decision of the planning authorities —, the tender cannot be considered unconditional.

95.

When public bodies sell assets, goods and services, the only relevant criterion for selecting the buyer should be the highest price, (149) also taking into account the requested contractual arrangements (for example the vendor's sales guarantee or other post-sale commitments). Only credible (150) and binding offers should be considered. (151)

96.

When public bodies buy assets, goods and services, any specific conditions attached to the tender should be non-discriminatory and closely and objectively related to the subject matter and to the specific economic objective of the contract. They should allow for the most economically advantageous offer to match the value of the market. The criteria therefore should be defined in such a way as to allow for an effectively competitive tendering procedure which leaves the successful bidder with a normal return, not more. In practice, this implies the use of tenders which put significant weight on the ‘price’ component of the bid or which are otherwise likely to achieve a competitive outcome (e.g. certain reverse tenders with sufficiently clear-cut award criteria).

97.

If a transaction has been realised through a tender or on ‘pari passu’ terms, this provides direct and specific evidence of compliance with market conditions. However, if a transaction has not been realised through a tender, or if the intervention of the public bodies is not ‘pari passu’ with that of private operators, this does not automatically mean that the transaction does not comply with market conditions. (152) In such cases compliance with market conditions can still be assessed through (i) benchmarking or (ii) other assessment methods. (153)

98.

To establish whether a transaction is in compliance with market conditions, that transaction can be assessed in the light of the terms under which comparable transactions carried out by comparable private operators have taken place in comparable situations (benchmarking).

99.

To identify an appropriate benchmark, it is necessary to pay particular attention to the kind of operator concerned (for example a group holding, a speculative fund, or a long-term investor seeking to secure profits in the longer run), the type of transaction at stake (for example equity participation or debt transaction) and the market or markets concerned (for example financial markets, fast-growing technology markets, utility or infrastructure markets). The timing of the transactions is also particularly relevant when significant economic developments have taken place. Where appropriate, the available market benchmarks may need to be adjusted according to the specific features of the State transaction (for instance, the situation of the beneficiary undertaking and of the relevant market). (154) Benchmarking may not be an appropriate method to establish market prices if the available benchmarks have not been defined with regard to market considerations or the existing prices are significantly distorted by public interventions.

100.

Benchmarking often does not establish one precise reference value but rather establishes a range of possible values by assessing a set of comparable transactions. Where the aim of the assessment is to consider whether the State intervention is in line with market conditions, it is normally appropriate to consider measures of central tendency such as the average or the median of the set of comparable transactions.

101.

Whether a transaction is in line with market conditions can also be established on the basis of a generally-accepted, standard assessment methodology. (155) Such a methodology must be based on the available objective, verifiable and reliable data, (156) which should be sufficiently detailed and should reflect the economic situation at the time at which the transaction was decided, taking into account the level of risk and future expectations. (157) Depending on the value of the transaction, the robustness of the evaluation should normally be corroborated by performing a sensitivity analysis, assessing different business scenarios, preparing contingency plans and comparing the results with alternative evaluation methodologies. A new (ex-ante) valuation may need to be carried out if the transaction is delayed and it is necessary to take into account recent changes in market conditions.

102.

A widely accepted standard methodology to determine the (annual) return on investments is to calculate the internal rate of return (IRR). (158) One can also evaluate the investment decision in terms of its net present value (NPV), (159) which produces results equivalent to the IRR in most cases. (160) To assess whether the investment is carried out on market terms, the return on the investment must be compared to the normal expected market return. A normal expected return (or cost of capital of the investment) can be defined as the average expected return that the market requires from the investment on the basis of generally accepted criteria, in particular the risk of the investment, taking into account the financial position of the company and the specific features of the sector, region or country. If this normal return cannot be reasonably expected, then the investment would most likely not be pursued on market terms. In general, the riskier the project, the higher the rate of return that fund providers will demand, that is to say the higher the cost of capital.

103.

The appropriate assessment methodology may depend on the market situation, (161) data availability or the type of transaction. For instance, whereas an investor seeks to generate a profit by investing in undertakings (in which case IRR or NPV are likely to be the most appropriate method), a creditor seeks to obtain payment of sums owed to it (the principal sum and any interest) by a debtor within the contractually and legally determined period (162) (in which case the evaluation of collateral, for example the asset value, could be more relevant). In the case of sales of land, an independent expert evaluation prior to the sale negotiations to establish the market value on the basis of generally accepted market indicators and valuation standards is in principle satisfactory. (163)

104.

Methods to establish the IRR or NPV of an investment do not typically result in one precise value that could be accepted, but rather in a range of possible values (depending on the economic, legal and other specific circumstances of the transaction inherent in the assessment method). Where the aim of the assessment is to consider whether the State intervention is in line with market conditions, it is normally appropriate to consider measures of central tendency, such as the average or the median of the set of comparable transactions.

105.

Prudent market economy operators typically assess their interventions by using several different methodologies to corroborate the estimates (for instance, NPV calculations are validated by benchmarking methods). The different methodologies converging at the same value will provide a further indication for establishing a genuine market price. Thus, the presence of complementary valuation methodologies corroborating each other's findings will be considered a positive indication when assessing whether a transaction is in line with market conditions.

106.

The fact that the public body concerned has prior economic exposure to an undertaking should be taken into consideration when examining whether a transaction is in line with market conditions, provided that a comparable private operator could have such prior exposure (for example in its capacity of shareholder of an undertaking). (164)

107.

Prior exposure must be considered in the framework of counterfactual scenarios for the purpose of the MEO test. For instance, in the case of an equity or debt intervention in a public undertaking in difficulty, the expected return on such an investment should be compared with the expected return in the counterfactual scenario of the liquidation of the company. In the event that liquidation provides higher gains or lower losses, a prudent market economy operator would choose that option. (165) For this purpose, the liquidation costs to be considered should not include costs linked to the responsibilities of the public authorities, but only costs that a rational market economy operator would incur, (166) also taking into account the evolution of the social, economic and environmental context in which it operates. (167)

108.

In the same way as any other transaction, loans and guarantees granted by public bodies (including public undertakings) may entail State aid if they are not in line with market terms.

109.

For guarantees, a triangular situation involving a public entity as a guarantor, a borrower and a lender normally has to be analysed. (168) In most cases, aid can only be present at the level of the borrower, as the public guarantee may grant it an advantage, by enabling it to borrow at a rate that would not have been obtainable on the market without the guarantee (169) (or to borrow in a situation where, exceptionally, no loan could have been obtained on the market at any rate). However, under certain specific circumstances, the granting of a public guarantee might also entail aid to the lender, in particular where the guarantee is given ex post on an existing obligation between lender and borrower, where a complete passing on of the advantage to the borrower is not ensured (170) or where a guaranteed loan is used to pay back a non-guaranteed one. (171)

110.

Any guarantee granted on terms that are more favourable than market conditions, taking into account the economic situation of the borrower, confers an advantage on the latter (who pays a fee that does not appropriately reflect the risk that the guarantor assumes). (172) In general, unlimited guarantees are not in line with normal market conditions. This also applies to implicit guarantees stemming from State liability for the debts of insolvent undertakings sheltered from ordinary bankruptcy rules. (173)

111.

In the absence of specific market information on a given debt transaction, the debt instrument's compliance with market conditions may be established on the basis of a comparison with comparable market transactions (that is to say through benchmarking). In the case of loans and guarantees, information on the financing costs of the undertaking may, for example, be obtained from other (recent) loans taken by the undertaking in question, from yields on bonds issued by the undertaking or from credit default swap spreads on that undertaking. Comparable market transactions may also be similar loan or guarantee transactions undertaken by a sample of comparator companies, bonds issued by a sample of comparator companies or credit default swap spreads on a sample of comparator companies. In the case of guarantees, if no corresponding price benchmark can be found on the financial markets, the total financing cost of the guaranteed loan, including the interest rate of the loan and the guarantee premium, should be compared to the market price of a similar non-guaranteed loan. Benchmarking methods may be complemented with assessment methods based on the return on capital. (174)

112.

To facilitate assessment of whether a measure complies with the MEO test, the Commission has developed proxies to determine the aid character of loans and guarantees.

113.

For loans, the methodology to calculate a reference rate, to act as a proxy for the market price in situations where comparable market transactions are not easy to identify (something that is more likely to apply to transactions involving limited amounts and/or transactions involving small and medium sized undertakings (SMEs)) is set out in the Reference Rate Communication. (175) It should be recalled that this reference rate is only a proxy. (176) If comparable transactions have typically taken place at a lower price than that indicated as a proxy by the reference rate, the Member State can consider this lower price to be the market price. If, on the other hand, the same company has carried out recent similar transactions at a higher price than the reference rate, and its financial situation and the market environment have remained substantially unchanged, the reference rate may not constitute a valid proxy of market rates for that specific case.

114.

The Commission has developed detailed guidance on proxies (and irrebuttable presumptions (‘safe harbours’) for SMEs) relating to guarantees in the Notice on Guarantees. (177) According to that Notice, in order to rule out the presence of aid, it is normally sufficient that the borrower is not in financial difficulty, that the guarantee is linked to a specific transaction, that the lender bears part of the risk and that the borrower pays a market-oriented price for the guarantee.

115.

An advantage can be conferred on undertakings other than those to which State resources are directly transferred (indirect advantage). (178) A measure can also constitute both a direct advantage to the recipient undertaking and an indirect advantage to other undertakings, for instance, undertakings operating at subsequent levels of activity. (179) The direct recipient of the advantage can be either an undertaking or an entity (natural or legal person) not engaged in any economic activity. (180)

116.

Such indirect advantages should be distinguished from mere secondary economic effects that are inherent in almost all State aid measures (for example through an increase of output). For this purpose, the foreseeable effects of the measure should be examined from an ex ante point of view. An indirect advantage is present if the measure is designed in such a way as to channel its secondary effects towards identifiable undertakings or groups of undertakings. This is the case, for example, if the direct aid is, de facto or de jure, made conditional on the purchase of goods or services produced by certain undertakings only (for example only undertakings established in certain areas). (181)

117.

To fall within the scope of Article 107(1) of the Treaty, a State measure must favour ‘certain undertakings or the production of certain goods’. Hence, not all measures which favour economic operators fall under the notion of aid, but only those which grant an advantage in a selective way to certain undertakings or categories of undertakings or to certain economic sectors.

118.

Measures of purely general application which do not favour certain undertakings only or the production of certain goods only do not fall within the scope of Article 107(1) of the Treaty. However, the case-law has made it clear that even interventions which, at first appearance, apply to undertakings in general may be selective to a certain extent and, accordingly, be regarded as measures designed to favour certain undertakings or the production of certain goods. (182) Neither a large number of eligible undertakings (which can even include all undertakings of a given sector), nor the diversity and size of the sectors to which they belong, provide grounds for concluding that a State measure constitutes a general measure of economic policy, if not all economic sectors can benefit from it. (183) The fact that the aid is not aimed at one or more specific recipients defined in advance, but that it is subject to a series of objective criteria according to which it may be granted, within the framework of a predetermined overall budget allocation, to an indefinite number of beneficiaries who are not initially individually identified, is insufficient to call into question the selective nature of the measure. (184)

119.

To clarify the notion of selectivity under State aid law, it is useful to distinguish between material and regional selectivity. Moreover, it is useful to provide further guidance on certain issues specific to tax (or similar) measures.

120.

The material selectivity of a measure implies that the measure applies only to certain (groups of) undertakings or certain sectors of the economy in a given Member State. Material selectivity can be established de jure or de facto.

121.

De jure selectivity results directly from the legal criteria for granting a measure that is formally reserved for certain undertakings only (for instance: those having a certain size, active in certain sectors or having a certain legal form (185); companies incorporated or newly listed on a regulated market during a particular period (186); companies belonging to a group having certain characteristics or entrusted with certain functions within a group (187); ailing companies (188); or export undertakings or undertakings performing export-related activities (189)). De facto selectivity can be established in cases where, although the formal criteria for the application of the measure are formulated in general and objective terms, the structure of the measure is such that its effects significantly favour a particular group of undertakings (as in the examples in the preceding sentence). (190)

122.

De facto selectivity may be the result of conditions or barriers imposed by Member States preventing certain undertakings from benefiting from the measure. For example, applying a tax measure (for example a tax credit) only to investments exceeding a certain threshold (other than a minor threshold for reasons of administrative expediency) may mean that the measure is de facto reserved for undertakings with significant financial resources. (191) A measure granting certain advantages for a brief period only may also be de facto selective. (192)

123.

General measures which prima facie apply to all undertakings but are limited by the discretionary power of the public administration are selective. (193) This is the case where meeting the given criteria does not automatically result in an entitlement to the measure.

124.

Public administrations have discretionary power in applying a measure, in particular, where the criteria for granting the aid are formulated in a very general or vague manner that necessarily involves a margin of discretion in the assessment. One example would be that the tax administration can vary the conditions for granting a tax concession according to the characteristics of the investment project submitted to it for assessment. Similarly, if the tax administration has broad discretion to determine the beneficiaries or the conditions under which a tax advantage is granted on the basis of criteria unrelated to the tax system, such as maintaining employment, the exercise of that discretion must then be regarded as favouring ‘certain undertakings or the production of certain goods’. (194)

125.

The fact that a tax relief requires prior administrative authorisation does not automatically mean that it constitutes a selective measure. This is not the case where a prior administrative authorisation is based on objective, non-discriminatory criteria which are known in advance, thus circumscribing the exercise of the public administrations' discretion. Such a prior administrative authorisation scheme must also be based on a procedural system which is easily accessible and capable of ensuring that a request for authorisation will be dealt with objectively and impartially within a reasonable time and refusals to grant authorisation must also be capable of being challenged in judicial or quasi-judicial proceedings. (195)

126.

When Member States adopt ad hoc positive measures benefiting one or more identified undertakings (for instance, granting money or assets to certain undertakings), it is normally easy to conclude that such measures have a selective character, as they reserve favourable treatment for one or a few undertakings. (196)

127.

The situation is usually less clear when Member States adopt broader measures applicable to all undertakings fulfilling certain criteria, which mitigate the charges that those undertakings would normally have to bear (for instance, tax or social security exemptions for undertakings fulfilling certain criteria).

128.

In such cases, the selectivity of the measures should normally be assessed by means of a three-step analysis. First, the system of reference must be identified. Second, it should be determined whether a given measure constitutes a derogation from that system insofar as it differentiates between economic operators who, in light of the objectives intrinsic to the system, are in a comparable factual and legal situation. Assessing whether a derogation exists is the key element of this part of the test and allows a conclusion to be drawn as to whether the measure is prima facie selective. If the measure in question does not constitute a derogation from the reference system, it is not selective. However, if it does (and therefore is prima facie selective), it needs to be established, in the third step of the test, whether the derogation is justified by the nature or the general scheme of the (reference) system. (197) If a prima facie selective measure is justified by the nature or the general scheme of the system, it will not be considered selective and will thus fall outside the scope of Article 107(1) of the Treaty. (198)

129.

However, the three-step analysis cannot be applied in certain cases, taking into account the practical effects of the measures concerned. It must be emphasised that Article 107(1) of the Treaty does not distinguish between measures of State intervention in terms of their causes or aims, but defines them in relation to their effects, independently of the techniques used. (199) This means that in certain exceptional cases it is not sufficient to examine whether a given measure derogates from the rules of the reference system as defined by the Member State concerned. It is also necessary to evaluate whether the boundaries of the system of reference have been designed in a consistent manner or, conversely, in a clearly arbitrary or biased way, so as to favour certain undertakings which are in a comparable situation with regard to the underlying logic of the system in question.

130.

Thus, in Joined Cases C-106/09 P and C-107/09 P (200) concerning the Gibraltar tax reform, the Court of Justice found that the reference system as defined by the Member State concerned, although founded on criteria that were of a general nature, discriminated in practice between companies which were in a comparable situation with regard to the objective of the tax reform, resulting in a selective advantage being conferred on offshore companies. (201) In this respect, the Court found that the fact that offshore companies were not taxed was not a random consequence of the regime, but the inevitable consequence of the fact that the bases of assessment were specifically designed so that offshore companies had no tax base. (202)

131.

Similar verification may also be necessary in certain cases concerning special-purpose levies, where there are elements indicating that the boundaries of the levy have been designed in a clearly arbitrary or biased way, so as to favour certain products or certain activities which are in a comparable situation with regard to the underlying logic of the levies in question. For instance, in Ferring, (203) the Court of Justice considered that a levy imposed on the direct sale of medicinal products by pharmaceutical laboratories but not on the sale by wholesalers was selective. In the light of the particular factual circumstances — such as the clear objective of the measure and its effects — the Court did not simply examine whether the measure in question would lead to a derogation from the reference system constituted by the levy. It also compared the situations of the pharmaceutical laboratories (subject to the levy) and of the wholesalers (excluded), concluding that the non-imposition of the tax on the direct sales by the wholesalers equated to granting them a prima facie selective tax exemption. (204)

132.

The reference system constitutes the benchmark against which the selectivity of a measure is assessed.

133.

The reference system is composed of a consistent set of rules that generally apply — on the basis of objective criteria — to all undertakings falling within its scope as defined by its objective. Typically, those rules define not only the scope of the system, but also the conditions under which the system applies, the rights and obligations of undertakings subject to it and the technicalities of the functioning of the system.

134.

In the case of taxes, the reference system is based on such elements as the tax base, the taxable persons, the taxable event and the tax rates. For example, a reference system could be identified with regard to the corporate income tax system, (205) the VAT system, (206) or the general system of taxation of insurance. (207) The same applies to special-purpose (stand-alone) levies, such as levies on certain products or activities having a negative impact on the environment or health, which do not really form part of a wider taxation system. As a result, and subject to special cases illustrated in paragraphs 129 to 131 above, the reference system is, in principle, the levy itself. (208)

135.

Once the reference system has been established, the next step of the analysis consists in examining whether a given measure differentiates between undertakings in derogation from that system. To do this, it is necessary to determine whether the measure is liable to favour certain undertakings or the production of certain goods as compared with other undertakings which are in a similar factual and legal situation, in the light of the intrinsic objective of the system of reference. (209) External policy objectives — such as regional, environmental or industrial policy objectives — cannot be relied upon by the Member State to justify the differentiated treatment of undertakings. (210)

136.

The structure of certain special-purpose levies (and in particular their tax bases), such as environmental and health taxes imposed to discourage certain activities or products that have an adverse effect on the environment or human health, will normally integrate the policy objectives pursued. In such cases, a differentiated treatment for activities or products whose situation is different from the situation of those activities or products which are subject to the tax as regards the intrinsic objective pursued, does not constitute a derogation. (211)

137.

If a measure favours certain undertakings or the production of certain goods which are in a comparable legal and factual situation, the measure is prima facie selective.

138.

A measure which derogates from the reference system (prima facie selectivity) is non-selective if it is justified by the nature or general scheme of that system. This is the case where a measure derives directly from the intrinsic basic or guiding principles of the reference system or where it is the result of inherent mechanisms necessary for the functioning and effectiveness of the system. (212) In contrast, it is not possible to rely on external policy objectives which are not inherent to the system. (213)

139.

The basis for a possible justification could, for instance, be the need to fight fraud or tax evasion, the need to take into account specific accounting requirements, administrative manageability, the principle of tax neutrality (214), the progressive nature of income tax and its redistributive purpose, the need to avoid double taxation, (215) or the objective of optimising the recovery of fiscal debts.

140.

Member States should, however, introduce and apply appropriate control and monitoring procedures to ensure that derogations are consistent with the logic and general scheme of the tax system. (216) For derogations to be justified by the nature or general scheme of the system, it is also necessary to ensure that those measures are proportionate and do not go beyond what is necessary to achieve the legitimate objective being pursued, in that the objective could not be attained by less far-reaching measures. (217)

141.

A Member State which introduces a differentiation between undertakings needs to be able to show that this differentiation is actually justified by the nature and general scheme of the system in question. (218)

142.

In principle, only measures that apply within the entire territory of the Member State escape the regional selectivity criterion laid down in Article 107(1) of the Treaty. However, as outlined below, the reference system does not necessarily have to be defined as the entire Member State. (219) It follows that not all measures that apply only to certain parts of the territory of a Member State are automatically selective.

143.

As established by the case-law, (220) measures with a regional or local scope of application may not be selective if certain requirements are fulfilled. This case-law has so far only dealt with tax measures. However, as regional selectivity is a general concept, the principles set out by the Union Courts as regards tax measures apply to other types of measures as well.

144.

In order to assess regional selectivity, three scenarios must be distinguished: (221)

145.

The existence of institutional autonomy can be established where the tax measure decision has been taken by a regional or local authority with its own constitutional, political and administrative status that is separate from that of the central government. In the Azores case, the Court observed that the Portuguese Constitution recognised the Azores as an autonomous region with its own political and administrative status and self-governing institutions, which also have their own fiscal competence, and the power to adapt national fiscal provisions to particular regional features. (226)

146.

The assessment of whether this criterion has been fulfilled in each individual case should include, in particular, examination of the constitution and other relevant laws of a given Member State so as to verify whether a given region indeed has its own separate political and administrative status and whether it has its own self-governing institutions which have the power to exercise their own fiscal competence.

147.

The existence of procedural autonomy can be established where a tax measure decision has been adopted without the central government being able to directly intervene in determining its content.

148.

The essential criterion for determining whether procedural autonomy exists is not the extent of the competence that the infra-State body is recognised as having, but the capability of that body, in view of its competence, to adopt a decision on a tax measure independently, that is to say without the central government being able to intervene directly as regards its content.

149.

The fact that a consultation or conciliation procedure exists between the central and regional (or local) authorities to avoid conflicts does not automatically mean that an infra-State body does not have procedural autonomy, provided that that body, and not the central government, has the final word on the adoption of the measure at stake. (227)

150.

The mere fact that the acts which an infra-State body adopts are subject to judicial review does not in itself mean that that body lacks procedural autonomy, since the existence of such review is an inherent feature of the rule of law. (228)

151.

A regional (or local) tax measure does not have to be completely separate from a more general tax system for it not to constitute State aid. In particular, it is not necessary that the tax system in question (bases of assessment, tax rates, tax recovery rules and exemptions) is fully devolved to the infra-State body. (229) For example, corporate tax devolution limited to the power to vary rates within a limited range, without devolving the power to change the bases of assessment (tax allowances and exemptions, etc.), could be considered as fulfilling the procedural autonomy condition if the pre-defined rate bracket allows the region concerned to exercise meaningful autonomous powers of taxation, without the central government being able to directly intervene as regards content.

152.

The existence of economic and financial autonomy can be established where an infra-State body assumes responsibility for the political and financial consequences of a tax reduction measure. This cannot be the case if the infra-State body is not responsible for managing a budget, that is to say when it does not have control of both revenue and expenditure.

153.

Therefore, in establishing the existence of economic and financial autonomy, the financial consequences of the tax measure in the region must not be offset by aid or subsidies from other regions or the central government. Hence, the existence of a direct causal link between the tax measure adopted by the infra-State body and the financial support from other regions or the central government of the Member State concerned rules out the existence of such autonomy. (230)

154.

The existence of economic and financial autonomy is not undermined by the fact that a shortfall in tax revenues as a result of the implementation of devolved tax powers (for example a lower tax rate) is offset by a parallel increase in the same revenues due to the arrival of new businesses attracted by the lower rates.

155.

The autonomy criteria do not require the rules governing tax collection to be devolved to the regional or local authorities, nor do they require the tax revenues to actually be collected by those authorities. The central government may continue to be responsible for collecting devolved taxes if the collection costs are borne by the infra-State authority.

156.

Member States are free to decide on the economic policy which they consider most appropriate and, in particular, to spread the tax burden as they see fit across the various factors of production. Nonetheless, Member States must exercise this competence in accordance with Union law. (231)

157.

In principle, genuine cooperative societies conform to operating principles which distinguish them from other economic operators. (232) In particular, they are subject to specific membership requirements and their activities are conducted for the mutual benefit of their members, (233) not in the interest of outside investors. In addition, reserves and assets are non-distributable and must be dedicated to the common interest of the members. Finally, cooperatives generally have limited access to equity markets and generate low profit margins.

158.

In the light of these particular features, cooperatives can be regarded as not being in a comparable factual and legal situation to that of commercial companies, so that preferential tax treatment for cooperatives may fall outside the scope of the State aid rules provided that: (234)

159.

If, however, the cooperative society under examination is found to be comparable to commercial companies, it should be included in the same reference framework as commercial companies and undergo the three-step analysis as set out in paragraphs 128 to 141. The third step of that analysis requires an analysis of whether the tax regime in question is justified by the logic of the tax system. (235)

160.

For this purpose, it should be noted that the measure needs to be in line with the basic or guiding principles of the Member State's tax system (by reference to the mechanisms inherent to that system). A derogation for cooperative societies in the sense that they are not taxed themselves as cooperatives can, for example, be justified by the fact that they distribute all their profits to their members and that tax is then levied on those individual members. In any event, the reduced taxation must be proportionate and not go beyond what is necessary. Moreover, appropriate control and monitoring procedures must be applied by the Member State concerned. (236)

161.

It is generally accepted that investment vehicles, such as undertakings for collective investment, (238) should be subject to an appropriate level of taxation since they basically operate as intermediary bodies between (third party) investors and the target companies that are the subject of investment. The absence of special tax rules governing investment funds or companies could result in an investment fund being treated as a separate taxpayer — with an additional layer of tax being imposed on any income or gains by the intermediary vehicle. In this context, Member States generally seek to reduce adverse taxation effects on investments through investment funds or companies compared to direct investments by individual investors and, as far as possible, to ensure that the overall final tax burden on the basket of various types of investments is about the same, irrespective of the vehicle used for the investment.

162.

Tax measures aimed at ensuring tax neutrality for investments in collective investment funds or companies should not be viewed as selective where those measures do not have the effect of favouring certain undertakings for collective investment or certain types of investments, (239) but rather of reducing or eliminating double economic taxation in accordance with the overall principles inherent to the tax system in question. For the purpose of this section, tax neutrality means that taxpayers are treated the same whether they invest in assets, such as government securities and the shares of joint-stock companies, directly or indirectly through investment funds. Accordingly, a tax regime for undertakings for collective investment respecting the purpose of fiscal transparency at the level of the intermediary vehicle may be justified by the logic of the tax system in question, provided that the prevention of double economic taxation constitutes a principle inherent to the tax system in question. By contrast, preferential tax treatment limited to well-defined investment vehicles which fulfil specific conditions (240) to the detriment of other investment vehicles that are in a comparable legal and factual situation should be viewed as selective, (241) for instance, where tax rules would provide for a favourable treatment of national venture, social impact or long-term investment funds and omit EU-harmonised EuVECA, (242) EuSEF (243) or ELTIF (244) funds.

163.

However, tax neutrality does not mean that such investment vehicles should be entirely exempt from any tax or that the fund managers should be exempt from tax on the fees charged by them for managing the underlying assets being invested into by the funds. (245) Nor does it justify a more beneficial tax treatment of a collective investment than of an individual investment for the tax regimes in question. (246) In such cases, the tax regime would be disproportionate and would go beyond what is necessary to achieve the objective of preventing double taxation and would therefore constitute a selective measure.

164.

Tax amnesties commonly involve immunity from criminal penalties, fines and (some or all) interest payments. While certain amnesties require payment in full of tax amounts due, (247) others entail a partial waiver of the amount of tax due. (248)

165.

In general, a tax amnesty measure that applies to undertakings can be considered a general measure if the conditions below are met. (249)

166.

First, the measure is effectively open to any undertaking of any sector or size that has outstanding tax liabilities due at the date set by the measure, without favouring any pre-defined group of undertakings. Second, it does not entail any de facto selectivity in favour of certain undertakings or sectors. Third, the tax administration's action is limited to administering the implementation of the tax amnesty without any discretionary power to intervene in the granting or intensity of the measure. Finally, the measure does not entail a waiver from verification.

167.

The limited temporal application of tax amnesties, which apply only for a short period (250) to tax liabilities which were due before a pre-defined date and which are still due at the time of the introduction of the tax amnesty, is inherent to the concept of a tax amnesty that aims to improve both the collection of taxes and taxpayers' compliance.

168.

Tax amnesty measures may also be considered as general measures if they follow the national legislature's objective of ensuring compliance with a general principle of law, such as the principle that a judgment must be given within a reasonable period of time. (251)

169.

The function of a tax ruling is to establish in advance the application of the ordinary tax system to a particular case in view of its specific facts and circumstances. For reasons of legal certainty, many national tax authorities provide prior administrative rulings on how specific transactions will be treated fiscally. (252) This may be done to establish in advance how the provisions of a bilateral tax treaty or national fiscal provisions will be applied to a particular case or how ‘arm's-length profits’ will be set for related party transactions where uncertainty justifies an advance ruling to ascertain whether certain intra-group transactions are priced at arm's length. (253) Member States can provide their taxpayers with legal certainty and predictability on the application of general tax rules, which is best ensured if its administrative ruling practice is transparent and the rulings are published.

170.

The grant of a tax ruling must, however, respect the State aid rules. Where a tax ruling endorses a result that does not reflect in a reliable manner what would result from a normal application of the ordinary tax system, that ruling may confer a selective advantage upon the addressee, in so far as that selective treatment results in a lowering of that addressee's tax liability in the Member State as compared to companies in a similar factual and legal situation.

171.

The Court of Justice has held that a reduction in the taxable base of an undertaking that results from a tax measure that enables a taxpayer to employ transfer prices in intra-group transactions that do not resemble prices which would be charged in conditions of free competition between independent undertakings negotiating under comparable circumstances at arm's length confers a selective advantage on that taxpayer, by virtue of the fact that its tax liability under the ordinary tax system is reduced as compared to independent companies which rely on their actually recorded profit to determine their taxable base. (254) Accordingly, a tax ruling which endorses a transfer pricing methodology for determining a corporate group entity's taxable profit that does not result in a reliable approximation of a market-based outcome in line with the arm's length principle confers a selective advantage upon its recipient. The search for a ‘reliable approximation of a market-based outcome’ means that any deviation from the best estimate of a market-based outcome must be limited and proportionate to the uncertainty inherent in the transfer pricing method chosen or the statistical tools employed for that approximation exercise.

172.

This arm's length principle necessarily forms part of the Commission's assessment of tax measures granted to group companies under Article 107(1) of the Treaty, independently of whether a Member State has incorporated this principle into its national legal system and in what form. It is used to establish whether the taxable profit of a group company for corporate income tax purposes has been determined on the basis of a methodology that produces a reliable approximation of a market-based outcome. A tax ruling endorsing such a methodology ensures that that company is not treated favourably under the ordinary rules of corporate taxation of profits in the Member State concerned as compared to standalone companies who are taxed on their accounting profit, which reflects prices determined on the market negotiated at arm's length. The arm's length principle the Commission applies in assessing transfer pricing rulings under the State aid rules is therefore an application of Article 107(1) of the Treaty, which prohibits unequal treatment in taxation of undertakings in a similar factual and legal situation. This principle binds the Member States and the national tax rules are not excluded from its scope. (255)

173.

When examining whether a transfer pricing ruling complies with the arm's length principle inherent in Article 107(1) of the Treaty, the Commission may have regard to the guidance provided by the Organisation for Economic Cooperation and Development (‘OECD’), in particular the ‘OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations’. Those guidelines do not deal with matters of State aid per se, but they capture the international consensus on transfer pricing and provide useful guidance to tax administrations and multinational enterprises on how to ensure that a transfer pricing methodology produces an outcome in line with market conditions. Consequently, if a transfer pricing arrangement complies with the guidance provided by the OECD Transfer Pricing Guidelines, including the guidance on the choice of the most appropriate method and leading to a reliable approximation of a market based outcome, a tax ruling endorsing that arrangement is unlikely to give rise to State aid.

174.

In sum, tax rulings confer a selective advantage on their addressees in particular where:

175.

Tax settlements generally occur in the context of disputes between a taxpayer and the tax authorities concerning the amount of tax owed. They constitute a common practice in a number of Member States. The conclusion of such tax settlements allows tax authorities to avoid long-standing legal disputes before national jurisdictions and to ensure quick recovery of the tax due. While the competence of Member States in this field is not in dispute, State aid may be involved in the conclusion of a tax settlement, in particular where it appears that the amount of tax due has been reduced without clear justification (such as optimising the recovery of debt) or in a disproportionate manner to the benefit of the taxpayer. (260)

176.

In this context, a transaction between the tax administration and a taxpayer may in particular entail a selective advantage where: (261)

177.

In general, tax measures of a purely technical nature such as depreciation/amortisation rules do not constitute State aid. The method of calculating asset depreciation varies from one Member State to another, but such methods may be inherent to the tax systems to which they belong.

178.

The difficulty in assessing possible selectivity with regard to the depreciation rate of certain assets lies in the requirement to establish a benchmark (from which a specific rate or depreciation method would possibly derogate). While in accounting terms, the purpose of this exercise is generally to reflect the economic depreciation of the assets with the aim of presenting a fair view of the financial situation of the company, the fiscal process follows different purposes such as allowing companies to spread deductible expenses over time.

179.

Depreciation incentives (such as a shorter term of depreciation, a more favourable depreciation method, (262) early depreciation, etc.) for certain types of assets or undertakings, which are not based on the guiding principles of the depreciation rules in question, may give rise to the existence of State aid. In contrast, accelerated and early depreciation rules for leased assets may be seen as general measures if the lease contracts in question are really accessible to companies of all sectors and sizes. (263)

180.

If the tax authority has discretionary power to set different depreciation periods or different valuation methods, firm by firm or sector by sector, there is obviously a presumption of selectivity. Likewise, prior authorisation from a tax administration as a condition for applying a depreciation scheme involves selectivity if the authorisation is not limited to the prior verification of the legal requirements. (264)

181.

Specific provisions that do not contain discretionary elements, allowing, for example, income tax to be determined on a fixed basis, may be justified by the nature and general scheme of the system where, for instance, they take account of specific accounting requirements or of the importance of land in assets which are specific to certain sectors.

182.

Such provisions are therefore not selective, if the following conditions are fulfilled:

183.

The provision of anti-abuse rules may be justified as measures to prevent tax avoidance by taxpayers. (265) However, such rules might be selective if they provide for a derogation (non-application of the anti-abuse rules) to specific undertakings or transactions, which would not be consistent with the logic underlying the anti-abuse rules in question. (266)

184.

Although excise duties are largely harmonised at Union level (which may affect the imputability criterion (267)), this does not automatically imply that any duty relief in these areas would fall outside the scope of the State aid rules. In fact, a reduced excise duty can grant a selective advantage to the undertakings which use the product in question as an input or sell it on the market. (268)

185.

Public support to undertakings only constitutes State aid under Article 107(1) of the Treaty if it ‘distorts or threatens to distort competition by favouring certain undertakings or the production of certain goods’ and only insofar as it ‘affects trade between Member States’.

186.

These are two distinct and necessary elements of the notion of aid. In practice, however, these criteria are often treated jointly in the assessment of State aid as they are, as a rule, considered inextricably linked. (269)

187.

A measure granted by the State is considered to distort or threaten to distort competition when it is liable to improve the competitive position of the recipient compared to other undertakings with which it competes. (270) For all practical purposes, a distortion of competition within the meaning of Article 107(1) of the Treaty is generally found to exist when the State grants a financial advantage to an undertaking in a liberalised sector where there is, or could be, competition. (271)

188.

The fact that the authorities assign a public service to an in-house provider (even if they were free to entrust that service to third parties) does not as such exclude a possible distortion of competition. However, a possible distortion of competition is excluded if the following cumulative conditions are met:

189.

Public support is liable to distort competition even if it does not help the recipient undertaking to expand and gain market share. It is enough that the aid allows it to maintain a stronger competitive position than it would have had if the aid had not been provided. In this context, for aid to be considered to distort competition, it is normally sufficient that the aid gives the beneficiary an advantage by relieving it of expenses it would otherwise have had to bear in the course of its day-to-day business operations. (274) The definition of State aid does not require that the distortion of competition or effect on trade is significant or material. The fact that the amount of aid is low or the recipient undertaking is small will not in itself rule out a distortion of competition or the threat thereof, (275) provided however that the likelihood of such a distortion is not merely hypothetical. (276)

190.

Public support to undertakings only constitutes State aid under Article 107(1) of the Treaty insofar as it ‘affects trade between Member States’. In that respect, it is not necessary to establish that the aid has an actual effect on trade between Member States but only whether the aid is liable to affect such trade. (277) In particular, the Union Courts have ruled that ‘where State financial aid strengthens the position of an undertaking as compared with other undertakings competing in intra-[Union] trade, the latter must be regarded as affected by the aid.’ (278)

191.

Public support can be considered capable of having an effect on trade between Member States even if the recipient is not directly involved in cross-border trade. For instance, the subsidy may make it more difficult for operators in other Member States to enter the market by maintaining or increasing local supply. (279)

192.

The relatively small amount of aid or the relatively small size of the undertaking which receives it does not as such exclude the possibility that trade between Member States might be affected. (280) A public subsidy granted to an undertaking which provides only local or regional services and does not provide any services outside its State of origin may nonetheless have an effect on trade between Member States where undertakings from other Member States could provide such services (also through the right of establishment) and that possibility is not merely hypothetical. For example, where a Member State grants a public subsidy to an undertaking for supplying transport services, the supply of those services may, by virtue of the subsidy, be maintained or increased with the result that undertakings established in other Member States have less of a chance of providing their transport services in the market in that Member State. (281) Such an effect may, however, be less likely where the scope of the economic activity is very small, as may for instance be evidenced by a very low turnover.

193.

In principle, an effect on trade can also occur even if the recipient exports all or most of its production outside the Union, but in such situations the effect is less immediate and cannot be assumed from the mere fact that the market is open to competition. (282)

194.

In establishing an effect on trade, it is not necessary to define the market or to investigate in detail the impact of the measure on the competitive position of the beneficiary and its competitors. (283)

195.

However, an effect on trade between Member States cannot be merely hypothetical or presumed. It must be established why the measure distorts or threatens to distort competition and is liable to have an effect on trade between Member States, based on the foreseeable effects of the measure. (284)

196.

The Commission has in a number of decisions considered, in view of the specific circumstances of the cases, that the measure had a purely local impact and consequently had no effect on trade between Member States. In those cases the Commission ascertained in particular that the beneficiary supplied goods or services to a limited area within a Member State and was unlikely to attract customers from other Member States, and that it could not be foreseen that the measure would have more than a marginal effect on the conditions of cross-border investments or establishment.

197.

While it is not possible to define general categories of measures that typically meet these criteria, past decisions provide examples of situations where the Commission found, in the light of the specific circumstances of the case, that public support was not liable to affect trade between Member States. Some examples of such cases are:

198.

Even if the circumstances in which the aid is granted are in most cases sufficient to show that the aid is capable of affecting trade between Member States and of distorting or threatening to distort competition, those circumstances should be appropriately set out. In the case of aid schemes, it is normally sufficient to examine the characteristics of the particular scheme. (295)

199.

The guidance regarding the notion of State aid as set out in this Notice applies to the public funding of infrastructure having an economic use, as it applies to any other public funding that favours an economic activity. (296) However, given the strategic importance of public funding of infrastructure, not least for the promotion of growth, and the questions which it often raises, it is appropriate to provide specific guidance on when public funding of infrastructure favours an undertaking, grants an advantage and has an effect on competition and on trade between Member States.

200.

Infrastructure projects often involve several categories of actors and any State aid involved may potentially benefit the construction (including extensions or improvements), the operation or the use of the infrastructure. (297) For the purposes of this section, it is, therefore, useful to distinguish between the developer and/or first owner (‘developer/owner’ (298)) of an infrastructure, the operators (that is to say undertakings who make direct use of the infrastructure to provide services to end-users, including undertakings which acquire the infrastructure from the developer/owner to exploit it economically or which obtain a concession or lease for the use and operation of the infrastructure), and the end-users of an infrastructure, although these functions may in some cases overlap.

201.

The public funding of much infrastructure was traditionally considered to fall outside the State aid rules since their construction and operation were considered to constitute general measures of public policy and not an economic activity. (299) More recently, several factors, such as liberalisation, privatisation, market integration and technological progress have, however, increased the scope for commercial exploitation of infrastructures.

202.

In the Aéroports de Paris judgment (300) the General Court acknowledged this evolution, clarifying that the operation of an airport had to be seen as an economic activity. More recently, the Leipzig/Halle judgment (301) confirmed that the construction of a commercial airport runway is an economic activity in itself. While these cases relate specifically to airports, the principles developed by the Union Courts appear to be of broader interpretation and thus applicable to the construction of other infrastructures that are indissociably linked to an economic activity. (302)

203.

On the other hand, the funding of infrastructure that is not meant to be commercially exploited is in principle excluded from the application of the State aid rules. This concerns, for instance, infrastructure that is used for activities that the State normally performs in the exercise of its public powers (for instance, military facilities, air traffic control in airports, lighthouses and other equipment for the needs of general navigation including on inland waterways, flood protection and low water management in the public interest, police and customs) or that is not used for offering goods or services on a market (for instance roads made available for free public use). Such activities are not of an economic nature and consequently fall outside the scope of the State aid rules, as does, accordingly, the public funding of the related infrastructure. (303)

204.

Where an infrastructure originally used for non-economic activities is later re-assigned to economic use (for example where a military airport is converted to civilian use), only the costs incurred for the conversion of the infrastructure to economic use will be taken into account for the assessment under the State aid rules. (304)

205.

If an infrastructure is used for both economic and non-economic activities, public funding for its construction will fall under the State aid rules only insofar as it covers the costs linked to the economic activities.

206.

If an entity is engaged in economic and non-economic activities, Member States have to ensure that the public funding provided for the non-economic activities cannot be used to cross-subsidize the economic activities. This can notably be ensured by limiting the public funding to the net cost (including the cost of capital) of the non-economic activities, to be identified on the basis of a clear separation of accounts.

207.

If, in cases of mixed use, the infrastructure is used almost exclusively for a non-economic activity, the Commission considers that its funding may fall outside the State aid rules in its entirety, provided the economic use remains purely ancillary, that is to say an activity which is directly related to and necessary for the operation of the infrastructure, or intrinsically linked to its main non-economic use. This should be considered to be the case when the economic activities consume the same inputs as the primary non-economic activities, for example material, equipment, labour or fixed capital. Ancillary economic activities must remain limited in scope, as regards the capacity of the infrastructure. (305) Examples of such ancillary economic activities may include a research organisation occasionally renting out its equipment and laboratories to industrial partners. (306) The Commission also considers that public financing provided to customary amenities (such as restaurants, shops or paid parking) of infrastructures that are almost exclusively used for a non-economic activity normally has no effect on trade between Member States since those customary amenities are unlikely to attract customers from other Member States and their financing is unlikely to have a more than marginal effect on cross-border investment or establishment.

208.

As the Court of Justice acknowledged in its Leipzig/Halle judgment, the construction of an infrastructure, or part of it, can fall within the State's exercise of public powers. (307) In such a case the public funding of the infrastructure (or the relevant part of the infrastructure) is not subject to State aid rules.

209.

Due to the uncertainty that existed prior to the Aéroports de Paris judgment, public authorities could legitimately consider that the public funding of infrastructure granted prior to that judgment did not constitute State aid and that, accordingly, such measures did not need to be notified to the Commission. It follows that the Commission cannot put such funding measures definitively adopted before the Aéroports de Paris judgment into question on the basis of State aid rules. (308) This does not imply any presumption as regards the presence or absence of State aid or legitimate expectations as regards funding measures not definitively adopted before the Aéroports de Paris judgment, which will have to be verified on a case by case basis. (309)

210.

The rationale underlying the cases in which the Commission considered that certain measures were not capable of affecting trade between Member States as set out in paragraphs 196 and 197 can also be relevant for certain public funding of infrastructure, particularly local or municipal infrastructure, even if it is commercially exploited. One pertinent characteristic of such cases would be a predominately local catchment area as well as evidence that cross-border investment is unlikely to be affected more than marginally. For example, the construction of local leisure installations, health care facilities, small airports or ports that predominately serve local users and for which the impact on cross-border investment is marginal are unlikely to affect trade. Evidence to demonstrate that there is no effect on trade could include data showing that there is only limited use of the infrastructure from outside the Member State and that cross-border investments in the market under consideration are minimal or unlikely to be adversely affected.

211.

There are circumstances in which certain infrastructures do not face direct competition from other infrastructures of the same kind or other infrastructures of a different kind offering services with a significant degree of substitutability, or with such services directly (310). The absence of direct competition between infrastructures is likely the case for comprehensive network infrastructures (311) that are natural monopolies, that is to say for which a replication would be uneconomical. Similarly, there may be sectors where private financing for the construction of infrastructures is insignificant. (312) The Commission considers that an effect on trade between Member States or a distortion of competition is normally excluded as regards the construction of the infrastructure in cases where at the same time (i) an infrastructure typically faces no direct competition, (ii) private financing is insignificant in the sector and Member State concerned and (iii) the infrastructure is not designed to selectively favour a specific undertaking or sector but provides benefits for society at large.

212.

In order for the entire public funding of a given project to fall outside State aid rules, Member States have to ensure that the funding provided for the construction of the infrastructures in the situations mentioned in paragraph 211 cannot be used to cross-subsidize or indirectly subsidize other economic activities, including the operation of the infrastructure. Cross-subsidization can be excluded by ensuring that the infrastructure owner does not engage in any other economic activity or — if the infrastructure owner is engaged in any other economic activity — by keeping separate accounts, allocating costs and revenues in an appropriate way and ensuring that any public funding does not benefit other activities. The absence of indirect aid, in particular to the operator of the infrastructure, can be ensured, for example, by tendering out the operation.

213.

This section provides an overview of how the Commission intends to assess the State aid nature of the funding of infrastructure in different sectors, having regard to the main features that public infrastructure financing typically and currently shows in the different sectors with respect to the conditions identified above. It is without prejudice to the outcome of the concrete case by case assessment of projects in the light of their specific characteristics, the way a given Member State has organised the provision of services linked to the use of the infrastructure and the development of the commercial services and of the internal market. It is not intended to replace an individual assessment of whether all the elements of the notion of State aid are fulfilled in respect of the concrete financing measure of a specific infrastructure. The Commission has also provided more detailed guidance on specific sectors in some of its Guidelines and Frameworks.

214.

Airport infrastructure consists of different types of infrastructure. Based on the case law of the Union courts, it is well established that most airport infrastructure (313) is intended for the provision of airport services to airlines against payment, (314) which qualify as economic activities, and that therefore their funding is subject to the State aid rules. Similarly, if an infrastructure is intended for non-aeronautical commercial services to other users, its public funding is subject to the State aid rules. (315) Since airports often compete with one another, the financing of airport infrastructure is also likely to affect trade between Member States. In contrast, public funding of infrastructures intended for activities that fall under the responsibility of the State in the exercise of its public powers does not fall under the State aid rules. Air traffic control, aircraft rescue and firefighting, police, customs and activities necessary to safeguard civil aviation at an airport against acts of unlawful interference are in general considered to be of such a non-economic nature.

215.

Similarly, as follows from the Commission's decision-making practice, (316) public funding of port infrastructure favours an economic activity and is hence in principle subject to State aid rules. As is the case with airports, ports may compete with one another and the financing of port infrastructure is therefore also likely to affect trade between Member States. However, investment for infrastructure that is necessary for activities that fall under the responsibility of the State in the exercise of its public powers are not subject to State aid control. Maritime traffic control, firefighting, police and customs are in general of such non-economic nature.

216.

Broadband infrastructure is used to enable the provision of telecommunication connectivity to end-users. Providing connectivity to end-users against payment is an economic activity. Broadband infrastructure is in many instances built by operators without any State funding, which is evidence of significant market financing, and in many geographical zones several networks of different operators compete. (317) Broadband infrastructures are part of large, interconnected and commercially exploited networks. For these reasons, public funding of broadband infrastructure for the provision of connectivity to end-users is subject to State aid rules, as set out in the Guidelines for the application of State aid rules in relation to the rapid deployment of broadband networks. (318) In contrast, connecting only public authorities is a non-economic activity and the public funding of so called ‘closed networks’ therefore does not constitute State aid. (319)

217.

Energy infrastructure  (320) is used for the provision of energy services against payment, which amounts to an economic activity. Energy infrastructure is, to a large extent, built by market actors, which is evidence of significant market financing, and financed through user tariffs. Public funding of energy infrastructure therefore favours an economic activity and is likely to have an effect on trade between Member States and is hence in principle subject to State aid rules. (321)

218.

Public funding of research infrastructures can favour an economic activity and is hence subject to State aid rules insofar as the infrastructure is in fact intended for the performance of economic activities (such as renting out equipment or laboratories to undertakings, supplying services to undertakings or performing contract research). Public funding of research infrastructures used for non-economic activities, such as independent research for increased knowledge and better understanding, in contrast, does not fall under the State aid rules. For more detailed guidance on the distinction between economic and non-economic activities in the area of research, see the explanations provided in the Framework for State aid for research and development and innovation. (322)

219.

While the operation of railway infrastructure  (323) may constitute an economic activity, (324) the construction of railway infrastructure which is made available to potential users on equal and non-discriminatory terms — as opposed to the operation of the infrastructure — typically fulfils the conditions set out in paragraph 211 and its financing therefore typically does not affect trade between Member States or distort competition. To ensure that the entire funding of a given project is not subject to State aid rules, Member States also have to ensure that the conditions set out in paragraph 212 are fulfilled. The same reasoning applies to investments in railway bridges, railway tunnels and urban transport infrastructure  (325).

220.

While roads made available for free public use are general infrastructures and their public funding does not fall under State aid rules, the operation of a toll-road constitutes in many instances an economic activity. However, the construction as such of road infrastructure (326), including toll-roads — as opposed to the operation of a toll-road and provided that it does not constitute dedicated infrastructure — typically fulfils the conditions set out in paragraph 211 and its financing therefore typically does not affect trade between Member States or distort competition. (327) To ensure that the entire public funding of a given project is not subject to State aid rules Member States also have to ensure that the conditions set out in paragraph 212 are fulfilled. The same reasoning applies to investments in bridges, tunnels and inland waterways (for example rivers and canals).

221.

While the operation of water supply and waste water networks  (328) constitutes an economic activity, the construction of a comprehensive water supply and waste water network as such typically fulfils the conditions set out in paragraph 211 and its financing therefore typically does not distort competition or affect trade between Member States. To ensure that the entire funding of a given project is not subject to State aid rules Member States also have to ensure that the conditions set out in paragraph 212 are fulfilled.

222.

Where all the elements of Article 107(1) of the Treaty are fulfilled as regards the developer/owner of an infrastructure, State aid to the developer/owner is present, irrespective of whether they make direct use of the infrastructure to provide goods or services themselves or make the infrastructure available to a third party operator who in turn provides services to end-users of the infrastructure (for example, where the owner of an airport grants a concession for the provision of services in the airport).

223.

Operators who make use of the aided infrastructure to provide services to end-users receive an advantage if the use of the infrastructure provides them with an economic benefit that they would not have obtained under normal market conditions. This normally applies if what they pay for the right to exploit the infrastructure is less than what they would pay for a comparable infrastructure under normal market conditions. Guidance on how to establish whether the terms of operation comply with market conditions is provided in section 4.2. In line with that section, the Commission considers that an economic advantage to the operator can in particular be excluded if the concession to operate the infrastructure (or parts of it) is assigned for a positive price through a tender that meets all the relevant conditions set out in paragraphs 90 to 96. (329)

224.

However, the Commission recalls that if a Member State does not comply with its notification obligation and there are doubts as to the compatibility of the aid to the developer/owner with the internal market, the Commission may issue an injunction requiring the Member State to suspend the implementation of the measure and to provisionally recover any money paid until it has taken a decision on its compatibility. In addition, national judges are under an obligation to do so as well at the request of competitors. Furthermore, if, following its assessment of the measure, the Commission adopts a decision declaring the aid to be incompatible with the internal market and orders its recovery, an impact on the operator of the infrastructure cannot be excluded.

225.

If the operator of an infrastructure has received State aid or if its resources constitute State resources, it is in a position to grant an advantage to the users of the infrastructure (if they are undertakings) unless the terms of use comply with the MEO test, that is to say the infrastructure is made available to the users on market terms.

226.

In accordance with the general principles explained in section 4.2, an advantage to users in such cases can be excluded where the fees for use of the infrastructure have been set through a tender that meets all the relevant conditions set out in paragraphs 90 to 96.

227.

As explained in section 4.2, where such specific evidence is not available, the question of whether a transaction is in line with market conditions can be assessed in the light of the terms and conditions under which the use of comparable infrastructure is granted by comparable private operators in comparable situations (benchmarking), provided such a comparison is possible.

228.

If none of the above assessment criteria can be applied, the fact that a transaction is in line with market conditions can be established on the basis of a generally accepted, standard assessment methodology. The Commission considers that the MEO test can be satisfied for public funding of open infrastructures not dedicated to any specific user(s) where their users incrementally contribute, from an ex ante viewpoint, to the profitability of the project/operator. This is the case where the operator of the infrastructure establishes commercial arrangements with individual users that allow covering all costs stemming from such arrangements, including a reasonable profit margin on the basis of sound medium-term prospect. This assessment should take into account all incremental revenues and expected incremental costs incurred by the operator in relation to the activity of the specific user. (330)

229.

This Communication replaces the following Commission Communications and Notices:

230.

The present Communication replaces any opposing statements relating to the notion of State aid included in any existing Commission Communications and Frameworks, save for statements pertaining to specific sectors and justified by their particular features.