1.

By its appeal, the European Commission asks the Court to set aside the judgment of the General Court in Case T‑386/14, ( 2 ) in which the General Court annulled Commission Decision 2014/884/EU on State aid granted by the Kingdom of Denmark to FIH. ( 3 )

2.

In essence, the General Court found that the Commission committed an error in law in having chosen an incorrect analytical framework for establishing the existence of State aid and, in particular, in not having applied the market economy creditor test, a variant of the ‘market economy operator’ (MEO) test. ( 4 )

3.

The principal question raised by the present appeal is the following: is the Commission to apply the market economy creditor test in order to take into account the financial exposure of the Member State, as a creditor, when that exposure stems from the previous State aid granted to the undertaking in question?

4.

Since this issue concerns the application of the MEO test to a series of consecutive aid measures, it is of particular relevance to the assessment of public support granted in order to address an evolving financial crisis. The Court has in the past addressed similar problems, ( 5 ) but the present case shows the need to consolidate and clarify the existing case-law.

5.

The interpretation of the concept of State aid given by the Court in the present case might be also important for the application of the Bank Recovery and Resolution Directive (2014/59/EU), which is related to the function of the banking union. ( 6 )

6.

The background to the dispute, as set out in the judgment under appeal, may be summarised as follows.

7.

FIH Erhvervsbank A/S (‘FIH’) is a Danish bank wholly owned by FIH Holding A/S (‘FIH Holding’).

8.

In 2009, like other banks, FIH benefited from certain measures adopted by the Kingdom of Denmark in order to stabilise its banking sector (‘the 2009 measures’).

9.

In June and July 2009, FIH received a hybrid tier 1 capital injection of 1.9 billion Danish krone (DKK) (approximately EUR 225 million) and was granted a State guarantee totalling DKK 50 billion (approximately EUR 6.31 billion). Both measures were adopted in the context of State aid schemes which had been approved by the Commission.

10.

FIH used the entire guarantee to issue State-guaranteed bonds, of a total amount of DKK 41.7 billion (approximately EUR 5.56 billion), due to mature in 2012 and 2013. Between 2009 and 2011, Moody’s ratings agency downgraded FIH’s rating from A2 to B1 with negative outlook.

11.

In 2012 the Danish Government envisaged the adoption of a second package of measures in relation to FIH, and notified them to the Commission (‘the 2012 measures’).

12.

In a nutshell, those measures envisaged the transfer of the FIH’s most problematic assets to NewCo, a newly created subsidiary of FIH Holding, which would be provided with funding from the Danish Financial Stability Company (‘the FSC’). Subsequently, the FSC was to buy the shares in NewCo and the latter would be wound up, while FIH Holding would give the FSC an unlimited loss guarantee.

13.

The Commission concluded that the 2012 measures constituted State aid to NewCo and the FIH Group, but approved them temporarily, while also initiating a formal investigation procedure. In the course of investigation, the Danish Government submitted a restructuring plan and proposed commitments designed to address the concerns expressed by the Commission.

14.

In the contested decision, the Commission declared that the measures in favour of FIH and FIH Holding constituted State aid. In particular, the Commission considered that the measures did not fulfil the MEO test, since the Kingdom of Denmark did not act in a manner comparable to a market economy operator. The share purchase agreement in relation to NewCo was likely to generate a loss for the FSC and the proposed equity remuneration on investment was insufficient.

15.

As regards the compatibility of the State aid, the Commission examined the measures on the basis of Article 107(3)(b) TFEU and in the light of the Impaired Assets Communication and the Restructuring Communication, ( 7 ) and declared them to be compatible, subject to the observance of the restructuring plan and of the commitments set out in the Annex to the decision.

16.

By application lodged at the registry of the General Court on 24 May 2014, FIH and FIH Holding brought an action seeking the annulment of the contested decision.

17.

The applicants at first instance put forward three pleas of law in support of annulment. The first plea concerned infringement of Article 107(1) TFEU in so far as the contested decision did not correctly apply the MEO test. ( 8 )

18.

In the judgment under appeal, the General Court upheld the first plea and annulled the contested decision.

19.

Having recalled the applicable legal principles (paragraphs 50 to 60 of the judgment under appeal), the General Court observed that the question before it was whether the Commission had applied the appropriate test for assessing the existence of aid. The applicants claimed that the Commission should have applied the market economy creditor test in order to take into account the risk of financial losses resulting from the pre-existing debts of FIH with regard to the Danish Government. The Commission contended that those pre-existing debts could not be taken into account since they were the consequence of the 2009 measures which themselves constituted State aid (paragraph 61 of the judgment under appeal).

20.

The General Court observed that an economic operator in a situation such as that in the present case, where it has previously granted a capital injection and a guarantee to the company concerned, is akin to a private creditor seeking to minimise its losses rather than a private investor seeking to maximise the profits. It can be rational for an economic operator, having invested capital in a company to which he has also granted a guarantee, to adopt measures which substantially reduce the risk of losing his capital and activating the guarantee. It may be equally economically rational for the Kingdom of Denmark to agree to measures such as a transfer of impaired assets, in so far as they have a limited cost and involve reduced risk and that, without such measures, it would be highly likely that it would have to bear losses in an amount greater than that cost (paragraphs 64 to 66 of the judgment under appeal).

21.

The General Court further concluded that the contested decision did not examine the cost that would have arisen had the Danish Government not adopted the 2012 measures and, in that respect, applied an incorrect legal test, namely, the market economy investor principle, instead of examining the measures in the light of the market economy creditor principle. The Kingdom of Denmark’s conduct, when it had adopted the measures at issue in 2012, could not be compared to that of an investor seeking to maximise its profit, but that of a creditor seeking to minimise the losses (paragraphs 68 and 69 of the judgment under appeal).

22.

As a consequence, the General Court rejected the Commission’s arguments (paragraphs 72 to 81 of the judgment under appeal) and upheld the first plea.

23.

The Commission claims that the Court should set aside the judgment under appeal, refer the case back to the General Court and reserve the costs.

24.

FIH and FIH Holding claim that the Court should dismiss the appeal and order the Commission to pay the costs.

25.

The Commission relies on a single ground of appeal, namely, that the General Court erred in law in finding that the Commission was required to apply the private creditor test and to take into account the cost which would have arisen for the Danish Government had it not adopted the 2012 measures.

26.

The Commission contends that that finding of the General Court is vitiated by an error of law because the cost in question is the direct consequence of earlier State aid, and in accordance with the case-law of the Court — in particular, judgments in Gröditzer Stahlwerke and Land Burgenland — the Commission cannot take such a cost into account when it applies the MEO test.

27.

FIH and FIH Holding contend that, in adopting the 2012 measures, the Kingdom of Denmark acted as a prudent and rational private creditor would have acted, and that it efficiently protected its economic interests as a creditor. They observe that prior to the adoption of the 2012 measures, the Kingdom of Denmark faced a significant risk of suffering a net loss of DKK 3.8 billion on the DKK 42 billion State guaranteed bonds and, therefore, a total net risk of loss of at least DKK 5.7 billion in case of FIH’s default. The 2012 measures significantly lowered the FSC’s exposure, both in terms of the degree and the amount at risk.

28.

According to FIH and FIH Holding, it would be inconsistent with the objective of State aid control if the State aid rules should prevent a rational rescheduling of a Member State’s exposure only because the risk facing it stemmed from previous State aid approved by the Commission. Such a result would also be contrary to the principle of neutrality in Article 345 TFEU, since public creditors would not have the same possibilities that private creditors have to protect their interests. In support of their argument, FIH and FIH Holding invoke the Court’s judgment in ING, which is also relied upon in the judgment under appeal.

29.

The present appeal presents the Court with the opportunity to explore the rationale behind the MEO test.

30.

According to the case-law of the Court, for a measure to be categorised as State aid within the meaning of the Treaties each of the four cumulative conditions laid down in Article 107(1) TFEU must be fulfilled. First, there must be an intervention by the State or through State resources; second, the intervention must be liable to affect trade between Member States; third, it must confer an advantage on the recipient; fourth, it must distort or threaten to distort competition. ( 9 )

31.

In the light of the settled case-law of the Court, the third of those conditions refers to the measures which, whatever their form, are likely directly or indirectly to favour certain undertakings or are to be regarded as an economic advantage which the recipient undertaking would not have obtained under normal market conditions. ( 10 )

32.

The concept of aid encompasses various forms of support, including those stemming from the transactions between the State and the undertakings operating on the market, if in the course of those transactions the recipient undertaking receives an economic advantage which it would not have obtained under normal market conditions. ( 11 )

33.

In order to determine whether a particular transaction confers such an economic advantage, the Commission has developed the MEO test. ( 12 ) That test applies in various forms, depending on whether the State acts as investor, creditor, vendor, guarantor, purchaser or lender. If similar conditions would have been obtained in the course of a comparable private market transaction, the measure does not confer any selective advantage on the undertaking, and therefore does not constitute State aid.

34.

That test is an expression of the principle of equal treatment of public and private undertakings, or of the wider principle of neutrality in relation to national law governing property ownership (Article 345 TFEU). When public authorities act on the market in circumstances which correspond to normal market conditions, their actions should not be regarded as State aid. ( 13 )

35.

The MEO test is applicable where the intervention of the State is of an economic nature, and not when the State acts as a public authority. That test, where applicable, is among the factors which the Commission is required to take into account for the purposes of establishing whether such aid exists. ( 14 )

36.

In two cases concerning the privatisation of public undertakings, the Court has excluded the application of the MEO test in relation to a transaction which is linked to previous State aid granted to the same undertaking.

37.

In Gröditzer Stahlwerke — concerning the privatisation of a public undertaking in difficulty — the German Government argued that the negative selling price was justified under the MEO test, since it would have cost more to wind up the undertaking than to privatise it for the negative price. The high potential cost of winding up was due to the fact that the undertaking had received in the past State guarantees and loans.

38.

The Court observed that a distinction must be drawn between the obligations which the State must assume as a shareholder and as a public authority. ( 15 ) The obligations in connection with State guaranteed loans arise from the grant of aid, thus, from the State acting as a public authority, and may not be included in the calculation of the normal cost of winding up. ( 16 ) Insofar as the shareholder loans were concerned, having regard to the undertaking’s difficulties and the conditions attaching to the loans, they constituted aid and, as a consequence, could not be taken into account in calculating the cost of the winding up. ( 17 )

39.

The Court confirmed this approach in Land Burgenland. The case concerned the privatisation of a bank which benefited from the ‘Ausfallhaftung’ — a statutory guarantee, which involved the obligation on the State to intervene in the event of insolvency. As a consequence, the State would have to act as guarantor for the bank, even after its privatisation, with regard to liabilities entered into before the date of the privatisation. The Austrian authorities argued that, given the risk of activating the State guarantee for the losses of the bank even after its privatisation, it was entitled to make sure that the buyer presented sufficient evidence of seriousness and of creditworthiness. According to the Austrian authorities, this consideration justified the acceptance of a considerably lower privatisation price, if that price was associated with the lower risk of having to discharge the guarantee obligation under the Ausfallhaftung. ( 18 )

40.

The Court recalled that, for the purposes of the MEO test, account must be taken only of the obligations linked to the situation of the State as shareholder, and not as a public authority. ( 19 ) The manner in which the advantage is provided and by which the State intervenes are irrelevant — provided that the Member State intervenes in its capacity as shareholder. ( 20 )

41.

The Court further observed that the General Court had examined whether the Ausfallhaftung had to be taken into account when implementing the private vendor test and had found that a private vendor would not have entered into such a guarantee. The Ausfallhaftung constituted State aid, and was therefore granted by the State exercising its prerogatives as a public authority. ( 21 )

42.

As a consequence, since the Ausfallhaftung itself was not entered into on normal market conditions, the financial risks associated with it could not be taken into account in order to justify the lower privatisation price under the MEO test. ( 22 )

43.

The Court’s judgments Gröditzer Stahlwerke and Land Burgenland have drawn some criticism in academic literature. It has been pointed out that an approach which precludes taking into account the financial burden resulting from lawful State aid limits the State’s power to minimize financial risk to its budget and to avoid further costs being incurred by taxpayers. ( 23 ) Similar arguments have been invoked by FIH and FIH Holding in the present appeal.

44.

A related issue was discussed in ING, in the context of the State measure consisting in the amendment to the repayment terms of the aid granted to a bank in difficulty.

45.

The measure in question consisted of a capital injection in exchange for securities, subject to repayment. Under the initial repayment terms, the securities were, on the initiative of ING Groep NV (‘ING’), either to be repurchased at the price of EUR 15 per security (with a redemption premium, as compared to the issue price of EUR 10) or, after three years, converted into ordinary shares, subject to the payment of accrued interest.

46.

Subsequently, the Kingdom of the Netherlands asked the Commission to approve the amendment to repayment terms, allowing ING to repurchase up to 50% of the securities at the issue price (EUR 10), plus the accrued interest and the early repayment penalty capped at a certain amount. The Commission concluded that the amended terms would result in an additional advantage for ING, thus, constituted State aid, and refused to assess whether the amendment satisfied the MEO test. ( 24 )

47.

In the context of an action for annulment brought by ING, the General Court found that the Commission had misinterpreted the concept of aid by not assessing whether, by accepting the amendment to the repayment terms, the Netherlands State acted as a private investor, inter alia because the Netherlands State could have been repaid early and because when the amendment occurred it obtained a greater certainty of being repaid in a satisfactory manner. ( 25 ) In consequence, the General Court annulled the Commission’s decision.

48.

The Court, confirming the judgment of the General Court, concluded that the Commission was wrong not to have assessed the economic rationality of the amendment to the repayment terms in the light of the MEO test. The fact that the prior capital injection itself constituted State aid was irrelevant. ( 26 )

49.

The Court observed that the application of the case-law related to the MEO test could not be compromised merely because the case concerned the applicability of the test to an amendment to the conditions for the redemption of securities acquired in return for State aid. A hypothetical private investor might also agree to renegotiate the conditions of the redemption of securities. The Commission erred in not having examined whether a private investor might have accepted the amendment as economically rational, in particular in order to increase the prospects of obtaining the repayment. ( 27 )

50.

In the meantime, the Commission re-examined the amendment to the repayment terms in the light of the private investor test, and found that the measure did not satisfy the test, since the initial repayment terms were financially more favourable to the Netherlands Government. ( 28 )

51.

The Land Burgenland and ING judgments both deal with measures adopted as a consequence of a previous State transaction with the beneficiary, but arrive at different outcomes as regards the application of the MEO test. In Land Burgenland, the MEO test could not be applied in order to assess the lower privatisation price, taking into account the financial risk resulting from earlier State aid. In ING the MEO test could justify the amendment to the repayment terms of previously granted aid, in particular, taking into account the earlier redemption date and the increased prospect of satisfactory repayment of the capital injection.

52.

Since both judgments deal with closely related situations, it is important to delimit their application.

53.

In the present appeal, the Commission interprets the judgments in Gröditzer Stahlwerke and Land Burgenland as meaning that the costs stemming from a previous State aid measure cannot be taken into account for the purposes of the MEO test, since those costs relate to the obligations of the Member State as a public authority.

54.

In my view, the principle stemming from that case-law is somewhat narrower. It precludes the taking into account of the financial risk stemming from the previous State aid measure, when the prospect of reducing that risk constitutes the economic justification of the subsequent State measure.

55.

Thus, in Gröditzer Stahlwerke the German Government argued that the negative sale price was justified by minimisation of the risk stemming from previous guarantees and loans. In Land Burgenland, the Austrian authorities argued that the lower privatisation price took into account the reduced risk of activating the State guarantee.

56.

In both situations, taking into account those objectives — as economically rational — would have resulted in a fallacious argument. When granting State aid, the Member State acted as a public authority, not as a market operator. When subsequently the State takes measures to minimise the financial risk resulting from its actions undertaken outside the competitive framework, such measures cannot be properly assessed in the light of competitive market conditions.

57.

A different interpretation might affect the effet utile of Article 107 TFEU. The aim of that provision is to prevent trade between Member States from being affected by advantages granted by public authorities which, in various forms, distort or threaten to distort competition by favouring certain undertakings or certain products. ( 29 ) In a situation where the State has lawfully granted aid, the risk of losses stemming from that aid measure cannot justify classifying further public support as neutral from the point of view of competition. Otherwise, the aid beneficiary would be treated more favourably than other undertakings, which have not received State aid.

58.

Not only would this limit the scope of the prohibition of State aid, which represents a fundamental principle of EU law of considerable importance for the functioning of the internal market, but it would also run counter to the rationale underlying that prohibition, which is to achieve a level playing field in terms of competition for all undertakings operating in the internal market. ( 30 )

59.

I would like to add that the principle stemming from Land Burgenland must be differentiated from other situations where the State finds itself in the position of a creditor vis-à-vis market operators. The Court has already confirmed that the market economy creditor test applies in relation to fiscal and the social security debts. ( 31 )

60.

The fact that a particular debt arises from the State’s actions as a public authority collecting taxes or social security contributions does not in itself preclude the application of the MEO test to subsequent measures aimed at recovering that debt. When a public creditor makes an arrangement with a debtor its actions should be assessed in relation to those of a circumspect private creditor in a market economy, since the State makes an attempt to achieve an economically profitable result by maximising its chances to recover the debt, whatever its origin — public or private. This scenario does not imply that potential losses stemming from earlier State aid measures should be taken into account.

61.

On the other hand, where, as in Land Burgenland, the State considers granting further support in order to reduce its financial exposure resulting from an earlier State aid measure, the origin of this financial exposure is relevant. State actions aimed at dealing with that financial exposure cannot be measured against the normal conditions of competition, insofar as the State has accepted that financial risk by acting outside the competitive framework.

62.

Can this interpretation be reconciled with the ING judgment? In my opinion, the answer depends on the reading of that judgment.

63.

The question raised in ING was to some extent similar to that in Land Burgenland: whether the State can be regarded as a rational market operator, when it accepts the amendment of the terms of a previous transaction, which constituted State aid, knowing that without the amendment the financial cost of the transaction might be higher. However, unlike Land Burgenland, the amendment to the terms was not exclusively justified by the risk of incurring further financial loss because of previously granted State aid. There was other, potential economic, justification of the amendment to the temporary capital injection in question: in some circumstances, an investor might accept a lower return on the invested capital, if the capital was repaid earlier. In ING, there was therefore at least a theoretical possibility that the amended terms would have been financially more favourable to the Dutch Government.

64.

If, on the other hand, the judgment in ING were interpreted, as indeed proposed by FIH and FIH Holding, as meaning that an amendment to pre-existing exposure from State aid can be justified under the MEO test insofar as it reduces the State’s exposure, then the judgments in Land Burgenland and ING would in my opinion be irreconcilable.

65.

To sum up, where an attempt to reduce the financial exposure resulting from an earlier State aid measure constitutes the only justification for further State action, there is no room for the MEO test. There may be some room for that test, when the measure is not exclusively directed at reducing losses, but can, overall, potentially constitute an attempt to achieve an economically justified result.

66.

I think that this interpretation is supported by the reasoning of the Land Burgenland judgment. The Austrian authorities claimed that, through the Ausfallhaftung, the Province of Burgenland was also seeking to make a profit or, at the very least, attempting to do so in addition to its other objectives. The Court did not outright reject that argument as invalid, but instead observed that it was for the Austrian authorities to prove that they indeed acted in that way. ( 32 ) However, given the circumstances of the case, it was difficult to see how the Ausfallhaftung could be regarded as a profit seeking measure.

67.

In ING, on the other hand, the repayment terms for the temporarily invested capital constituted a crucial element in assessing the existence of State aid from the point of view of the State as investor. Therefore, the amendment of those terms had to be also analysed against that yardstick. The fact that the initial transaction was found not to be on market conditions did not in itself exclude the possibility of analysing the amended terms from the market perspective.

68.

In line with this interpretation, the judgment in ING requires the application of the MEO test, but does not imply that, in the course of that application, it is admissible to take into account the State’s exposure stemming from an earlier State aid measure. ( 33 ) Thus, ING does not contradict the reasoning of Land Burgenland.

69.

In the light of my observations, the Land Burgenland and ING judgments both concern a series of consecutive aid measures, but in relation to two different scenarios.

70.

In Land Burgenland, the only justification for the lower privatisation price referred to the minimisation of risk resulting from an earlier State intervention, and therefore there was no room for the application of the MEO test. In ING, even though the amendment to repayment terms could help to reduce the risk to which the State was exposed by reason of an earlier State aid, the amendment could also pursue other, economic, objectives, and therefore — to that latter extent — had to be assessed under the MEO test.

71.

The particularity of the present case lies in the fact that it concerns a package of measures potentially connecting to both scenarios.

72.

On the one hand, in the context of the 2012 measures, the Kingdom of Denmark invested in FIH, and since that investment could potentially be in line with normal market conditions, it had to be checked under the MEO test. It appears from the decision contested before the General Court that the Commission has indeed examined those measures under the private investor test, against the expected return on investment of a hypothetical private investor. This test was applied — in my view, in line with the ING judgment — despite the fact that the 2012 measures came on top of earlier rescue aid.

73.

Secondly, FIH and FIH Holding argue that, regardless of that economic justification, the 2012 measures could also be justified by the need to reduce the financial exposure stemming from the previously granted State guarantee. In my view, this second justification could not have led to the application of the MEO test — in this case, private creditor test — because of the principle stated in the Land Burgenland judgment.

74.

In paragraphs 61 to 69 of the judgment under appeal, the General Court examined the first plea in law, which raised the question whether the Commission had to apply the private creditor test, in order to examine whether the 2012 measures could be rationally justified in view of the risk of financial losses resulting from the 2009 measures.

75.

The General Court stated that a rational economic operator would have taken into account its exposure arising from a capital injection and a guarantee granted to a company in subsequent financial difficulty and would have envisaged the adoption of measures to prevent the risk from materialising (paragraph 63 of the judgment under appeal). According to the General Court, it can be rational for an economic operator, having invested capital in a company to which he has also granted a guarantee, to adopt measures involving loss where they substantially reduce, if not eliminate, the risk of losing his capital and activating the guarantee (paragraph 65 of the judgment under appeal) and, therefore, it may be economically rational for the Danish Government to agree to the 2012 measures, such as a transfer of impaired assets, in so far as they have a limited cost and involve reduced risk and that, without such measures, it would be highly likely that it would have to bear losses in an amount greater than that cost (paragraph 66 of the judgment under appeal).

76.

The reasoning in those paragraphs essentially is economically sound, except in one respect. The economic rationale, as described, is entirely aimed at addressing the risk stemming from an earlier State aid measure, which was adopted in pursuance of non-economic objectives, was not adopted under market conditions, and — in accordance with the Land Burgenland judgment — could not constitute the premiss for the application of the MEO test in relation to subsequent State intervention. The State guarantee in question was not entered into on normal market conditions and the associated risk could not be taken into account, as economic risk, in order to justify subsequent public intervention. ( 34 ) In other words, contrary to what the General Court held in paragraphs 63 to 66 of the judgment under appeal, there was no room for the private creditor test, since the only justification that could be invoked under that test would be the need to reduce financial exposure stemming from an earlier State aid, and such justification is inadmissible under the Land Burgenland judgment.

77.

Moreover, I am not convinced by the interpretation of the Land Burgenland judgment, as explained by the General Court in paragraphs 79 to 82 of the judgment under appeal.

78.

The General Court correctly observes that, under that judgment, the Commission, in assessing the aid measure, could not take account of an earlier State guarantee, since by granting that guarantee the State had pursued non-economic objectives and acted as a public authority. ( 35 ) However, the General Court further holds — in my view, incorrectly — that the Land Burgenland judgment was merely concerned with the specific application of the private vendor test and with the elements to be taken into account as part of that test (paragraphs 79 and 80 of the judgment under appeal).

79.

This inconsistency is apparent in the General Court’s conclusion. The General Court concludes that the Commission will have to give, if necessary, due effect to the judgment in Land Burgenland, when applying the correct legal test (the private creditor test) to the 2012 measures (paragraph 81 of the judgment under appeal). However, if the Commission followed the Land Burgenland judgment, the exposure stemming from an earlier State aid measure could not be taken into account. Since, in the present case, reducing that financial exposure constitutes the only justification for invoking the private creditor test, that test would in any case be inapplicable.

80.

More generally, I do not agree with the General Court’s position, shared by FIH and FIH Holding in their written observations, that the interpretation of the scope of the MEO test adopted by the General Court is necessary, since it would be illogical if the Member State had to incur considerable financial loss to its budget in order to respect State aid rules (paragraph 67 of the judgment under appeal). This position is based on the premiss that the rationale, which consists in minimising the risk of financial loss to a State’s budget, cannot be otherwise incorporated in the framework of State aid control.

81.

This premiss does not seem to me correct. As the Commission observed at the hearing, in assessing the compatibility of State measures with the internal market, the Commission could take into account the proportionality of the State intervention, in particular, with regard to the potential losses that would be incurred by the budget, had the measures not been adopted. The issue is moot in the present case, since the Commission has, in any case, approved the 2012 measures.

82.

For the same reason, I am not convinced by the argument raised by FIH and FIH Holding invoking the principle of neutrality (Article 345 TFEU). ( 36 ) The principle stated in the Land Burgenland judgment does not create any risk of discrimination between the public and private sectors, but simply delimits the scope of the MEO test in relation to the consequences of the State’s actions in its capacity as a public authority.

83.

Moreover, in my opinion, the rationality of State actions with regard to the potential losses to its budget resulting from earlier State aid should be examined within the framework of State aid control (when assessing the compatibility of the measure) — and should not lead to putting the measure outside that framework (classifying it as not constituting State aid).

84.

Finally, if such financial support measures — which aim at minimising the losses resulting from earlier State aid — could escape the boundaries of the State aid definition and therefore the framework of State aid control, this would lead to wider unintended consequences.

85.

First, this would mean that, once the State has granted a guarantee, further support could be continued outside the State aid framework as defined in the Treaty, on the condition that the beneficiary undertaking is in difficulty and that there is a risk of activating the guarantee.

86.

Secondly, as the Commission has argued at the hearing, such an interpretation would not only remove some financial support measures from the scope of State aid control, but would also affect the scope of the recovery and resolution framework under Directive 2014/59, which is related to the functioning of the banking union. The application of those measures is inter alia based on the criterion referring to the question whether ‘extraordinary public financial support’ is required (Article 32(4)(d)), which in turn refers to the concept of State aid (Article 2(28)).

87.

For all those reasons, I consider that the General Court erred in law in concluding, in paragraph 69 of the judgment under appeal, that the Commission had applied an incorrect legal test in the contested decision, and that it should have examined the 2012 measures in the light of the market economy creditor principle in order to take into account the financial exposure resulting from earlier State aid measures.

88.

In the light of all those observations, I am of the view that the judgment under appeal should be set aside and — in particular, since the General Court has not examined the second plea in law raised at first instance — the case should be referred back to the General Court.

89.

In the light of the foregoing, I propose that the Court should uphold the appeal, set aside the judgment under appeal and refer the case back to the General Court, reserving the costs.