Language of document : ECLI:EU:C:2017:1017

OPINION OF ADVOCATE GENERAL

WAHL

delivered on 20 December 2017(1)

Case C203/16 P

Dirk Andres (administrator of Heitkamp BauHolding GmbH), previously Heitkamp BauHolding GmbH

v

European Commission

(Appeal — State aid — German tax legislation concerning the possibility of carrying losses forward to future tax years — Decision declaring aid incompatible with the internal market — Action for annulment — Admissibility — Article 263, fourth paragraph, TFEU — Individual concern — Concept of State aid — Selectivity — Reference system — Comparison — Justification)






1.        By its appeal, the administrator of Heitkamp BauHolding GmbH (‘the appellant’ or ‘HBH’) asks the Court to set aside the judgment of the General Court in Case T‑287/11. (2) By that judgment, the General Court dismissed the appellant’s action for annulment of Commission Decision 2011/527/EU (3) on State aid implemented by Germany on the basis of a scheme that allows losses to be carried forward in the case of restructuring of ailing companies.

2.        The appeal raises two issues that touch upon the very core of EU State aid law.

3.        The first issue is procedural. It concerns the standing requirements for private applicants set out in Article 263, fourth paragraph, TFEU: is the appellant individually concerned by the decision at issue within the meaning of the line of authority devolving from the Court’s judgment in Plaumann? (4) More particularly, where a tax saving arising from the impugned national measure has not been established in a final tax decision, can an undertaking nonetheless satisfy the criterion of individual concern? In that regard, this case gives the Court the opportunity to consider, in the context of tax measures, the arguably ambiguous demarcation between undertakings that have standing and those that do not.

4.        The second issue is substantive. It concerns one of the constituent elements of State aid. How should the concept of selectivity be construed in the particular context of direct tax measures? As a follow-up to World Duty Free, (5) the Court can in this case give guidance on the parameters that ought to inform the definition of the normal tax system (the reference system). That question is of particular significance, bearing in mind that the reference system constitutes the benchmark against which the selectivity of a tax measure is to be assessed. (6)

I.      Background to the proceedings

5.        Regarding the background to the present proceedings, the following is apparent from the decision at issue and the judgment under appeal.

A.      National legal framework

6.        In Germany, corporate taxation is based primarily on the Einkommensteuergesetz (Law on income tax) and the Körperschaftsteuergesetz (Law on corporation tax, ‘the KStG’).

7.        In accordance with Paragraph 10d(2) of the Law on income tax, losses incurred in the course of a tax year may be carried forward to later tax years. This means that taxable income in future tax years may be reduced by setting off losses (‘the loss carry-forward rule’). Under Paragraph 8(1) of the KStG, the loss carry-forward rule applies also to undertakings subject to corporation tax.

8.        The possibility of carrying losses forward led to the acquisition of ‘empty-shell companies’ for the sole purpose of reducing tax liabilities. Those companies had long ceased any economic activity, but still retained losses that had been carried forward.

9.        To prevent the trade in such companies, Paragraph 8(4) of the KStG was introduced in 1997. Under that rule, the possibility of carrying forward losses was restricted to corporate entities that were legally and economically identical to the entity that had incurred losses (‘the former rule governing the forfeiture of losses’). The rule did not contain a definition of ‘economically identical’, but gave one negative and two positive examples:

‘(a) a corporate entity is not economically identical if more than half of its shares are transferred and if the entity then continues its economic activity or starts it again with predominantly new assets;

(b) a corporate entity is economically identical, however, if the injection of new assets is solely for the purpose of restructuring the loss-making entity and if the activity which gave rise to the unrelieved loss carry-forward continues on a comparable scale for the following five years;

(c) a corporate entity is also economically identical if, rather than injecting new assets, the acquiring entity covers the losses that have accrued at the loss-making entity.’

10.      The exception to the forfeiture of losses (points b and c) was generally referred to as the Sanierungsklausel, a clause allowing restructuring of ailing companies.

11.      In January 2008, the former rule governing the forfeiture of losses laid down in Article 8(4) of the KStG was repealed. A new Paragraph 8c(1) was inserted into the KStG (‘the rule governing the forfeiture of losses’). That provision limited the possibility of carrying losses forward if 25% or more of the shares in a company are acquired (‘the prejudicial acquisition of a shareholding’). More specifically, if between 25% and 50% of the share capital, membership rights, ownership rights or voting rights are transferred, unused losses are forfeited on a pro rata basis. If more than 50% of the share capital, membership rights, ownership rights or voting rights are transferred to an acquirer, losses are forfeited entirely.

12.      At the time when the rule governing the forfeiture of losses came into force, it did not allow for exceptions. However, the tax authorities could, in the case of the prejudicial acquisition of a shareholding relating to the restructuring of an undertaking in difficulty, grant tax exemptions based on considerations of equity, pursuant to the decree on restructuring of 27 March 2003 issued by the German Federal Ministry of Finance.

13.      In June 2009, another provision was inserted into the KStG, namely Paragraph 8c(1a). In accordance with that provision, it was again possible to carry forward losses where an ailing company is acquired for restructuring purposes (‘the restructuring clause’ or ‘the measure at issue’).

14.      More specifically, under that clause, a corporate entity that has been acquired may carry losses forward also in the case of the prejudicial acquisition of a shareholding, provided that: (i) the acquisition serves the purpose of restructuring the corporate entity, (ii) the company is, or is likely to be, insolvent or over-indebted at the time of the acquisition, (iii) the company’s fundamental business structures are preserved, (iv) the company does not change its sector of activity during the five years following the acquisition, and (v) the company had not ceased operation at the time of the acquisition.

15.      The restructuring clause entered into force on 10 July 2009. It applies retroactively from 1 January 2008, which is the same date as that on which the rule governing the forfeiture of losses entered into force.

B.      The decision at issue

16.      The decision at issue was adopted on 26 January 2011. In accordance with Article 1 thereof, the State aid granted on the basis of Paragraph 8c(1a) of the KStG, unlawfully put into effect by Germany in breach of Article 108(3) TFEU, is incompatible with the internal market.

17.      Regarding the existence of State aid within the meaning of Article 107 TFEU, the Commission considered in particular that the restructuring clause constituted an exception to the general rule providing for the forfeiture of losses which had not been used by companies that had undergone a change of ownership. The clause in question was in the Commission’s view liable to confer a selective advantage on companies that met the requirements for benefiting from it. That difference in treatment was not justified by the nature or general scheme of the system. Further, according to the decision at issue, the purpose of the restructuring clause was to tackle the problems caused by the economic and financial crisis, an objective which was extrinsic to the tax system.

18.      In Articles 2 and 3 of the decision at issue, limited individual aid granted to certain beneficiaries under the restructuring scheme pursuant to Paragraph 8c(1a) was, nevertheless, declared to be compatible with the internal market provided that certain conditions were met.

19.      In Articles 4 and 6 of the decision at issue, the Commission ordered Germany to recover the incompatible aid granted under the measure at issue and to submit to the Commission a list of the beneficiaries of the aid scheme.

C.      Facts underlying the appeal

20.      At the time when the decision at issue was adopted, HBH was at risk of insolvency. On 20 February 2009, HBH’s parent company, Heitkamp KG, bought the shares in HBH with a view to a merger between the two companies. On the date of the transaction, the appellant met the conditions for application of the restructuring clause. That is clear from the binding information from the Finanzamt Herne (Herne tax office, Germany) of 11 November 2009 (‘the binding information’). On 29 April 2010, the appellant also received from the tax authorities a notice of advance payment relating to corporation tax for the tax year 2009, which took account of the losses carried forward pursuant to the restructuring clause.

21.      On 24 February 2010 the Commission informed the Federal Republic of Germany that it had decided to initiate the procedure laid down in Article 108(2) TFEU in respect of the measure at issue. By letter of 30 April 2010, the German Ministry of Finance ordered the tax authorities not to apply the restructuring clause.

22.      On 27 December 2010, the notice of advance payment of 29 April 2010 was replaced by a new notice relating to corporation tax for the 2009 tax year. That notice did not take account of the restructuring clause.

23.      On 1 April 2011, HBH received the tax decisions relating to corporation tax and basic trade tax for the 2009 tax year. Because the restructuring clause was not applied, the appellant could not carry forward the losses existing on 31 December 2008.

24.      On 19 April 2011, the tax authorities annulled the binding information.

25.      On 22 July 2011, the Federal Republic of Germany sent the Commission, as required by the decision at issue, the list of companies which had benefited from the measure at issue. It also sent to the Commission a list of companies for which binding information concerning the application of the restructuring clause had been annulled. HBH was mentioned in that latter list.

26.      The appellant challenged the abovementioned notices of advance payment and tax decision before the tax authorities and before the competent national courts. By order of 1 August 2011, the Finanzgericht Münster (Finance Court, Münster, Germany) suspended the application of those notices.

II.    Proceedings before the General Court and the judgment under appeal

27.      By application lodged on 6 June 2011, HBH brought an action before the General Court seeking annulment of the decision at issue.

28.      By a separate document lodged on 16 September 2011, the Commission raised an objection of inadmissibility.

29.      By document lodged on 29 August 2011, the Federal Republic of Germany applied for leave to intervene in support of the form of order sought by the applicant. By order of 5 October 2011, the President of the Second Chamber of the Court granted that leave.

30.      By order of 21 May 2014, consideration of the objection of inadmissibility was reserved for the final judgment.

31.      In support of its action, HBH relied on two grounds: on the one hand, that the measure at issue was not selective and, on the other hand, that it was justified by the nature or general scheme of the system.

32.      As regards the objection of inadmissibility, the Commission argued that HBH lacked locus standi within the meaning of Article 263, fourth paragraph, TFEU. More particularly, it claimed that HBH was not individually concerned by the decision at issue and that the decision at issue entails implementing measures. In addition, the Commission claimed that because HBH was not a beneficiary of aid, it had no interest in bringing proceedings.

33.      In the judgment under appeal, the General Court first rejected the objection of inadmissibility. In that respect, it considered that HBH was directly and individually concerned by the decision at issue. The General Court then went on to dismiss the action brought by HBH as unfounded.

III. Procedure before the Court and forms of order sought

34.      By its appeal, HBH claims that the Court should:

–        set aside the judgment under appeal to the extent that it dismissed the action as unfounded (points 2 and 3 of the operative part of the judgment) and annul the decision at issue;

–        in the alternative, set aside the judgment under appeal to the extent that it dismissed the action as unfounded (points 2 and 3 of the operative part of the judgment) and refer the case back to the General Court;

–        order the Commission to pay the costs.

35.      The Commission claims that the Court should dismiss the appeal and order HBH to pay the costs.

36.      By its cross-appeal, the Commission claims that the Court should:

–        set aside point 1 of the operative part of the judgment under appeal;

–        dismiss the action brought at first instance as inadmissible;

–        dismiss the appeal;

–        set aside point 3 of the operative part of the judgment under appeal ordering the Commission to pay a third of its costs;

–        order HBH to pay the costs incurred in the proceedings before the General Court and the Court.

37.      HBH claims that the Court should dismiss the cross-appeal as unfounded and order the Commission to pay the costs.

38.      HBH, the Commission and the German Government presented oral argument at the hearing held on 19 October 2017.

IV.    Analysis

39.      HBH relies on two grounds in support of its appeal. The first ground of appeal alleges that in the judgment under appeal, the General Court failed to comply with the obligation to state reasons. The judgment under appeal is allegedly vitiated by insufficient or contradictory reasoning. The second ground of appeal alleges that the General Court misconstrued Article 107(1) TFEU. It identifies several errors of law in the assessment of the selectivity of the measure at issue.

40.      The Commission claims that both grounds of appeal should be dismissed as either inadmissible or unfounded.

41.      The Commission has also brought a cross-appeal against the judgment under appeal. In its single ground of appeal, it alleges that the General Court erred in law in allowing the action for annulment brought by HBH. That is because HBH supposedly lacks standing: it is not individually concerned by the decision at issue. Consequently, the action brought by HBH at first instance should have been rejected as inadmissible.

42.      HBH considers that the cross-appeal should be rejected as unfounded.

43.      For procedural reasons, I shall first deal with the cross-appeal.

A.      The cross-appeal: Is the appellant individually concerned by the decision at issue?

44.      The Commission submits that, in the judgment under appeal, the General Court misconstrued the concept of individual concern within the meaning of Article 263, fourth paragraph, TFEU. (7) In particular, by granting HBH standing, the judgment under appeal departs from the Court’s settled case-law on that criterion in the specific context of actions for annulment brought against decisions adopted by the Commission declaring aid schemes incompatible with the internal market.

45.      The Commission submits that the judgment under appeal confuses the clear distinction, established in the case-law, between actual beneficiaries of aid that have standing and potential, future beneficiaries that do not. More specifically, it takes issue with the fact that in its assessment, as regards the criterion of individual concern, the General Court relied on the assertion that HBH had an ‘acquired right’ to a tax saving.

46.      HBH disagrees.

47.      To explain why I consider HBH to have standing, I shall begin by some introductory observations regarding the criterion of individual concern in the specific context of State aid law.

1.      Preliminary remarks on the criterion of individual concern in State aid law

48.      Access to courts and — by extension — judicial review constitute the bedrock of a legal system based on the principle of accountability and the rule of law. Not without reason, therefore, the question of standing of private applicants in EU law has captured the imagination of the EU legal community since (if not before) the Court’s seminal judgment in Plaumann. (8) That judgment laid down the foundations that still inform the interpretation of the standing requirements set out in Article 263, fourth paragraph, TFEU.

49.      As a general rule, for the purposes of bringing proceedings against an act of the EU institutions, a private applicant that is not an addressee of the act in question must show that it is directly and individually concerned by that act. That is so apart from in the particular case of a regulatory act that does not entail implementing measures referred to in Article 263, fourth paragraph, TFEU. In the case of such regulatory acts, it is sufficient that direct concern is established.

50.      As concerns decisions related to State aid, however, direct and individual concern must be shown if an undertaking wishes to bring an action for annulment against a decision adopted by the Commission declaring aid incompatible with the internal market. (9)

51.      The Court has adopted a strict approach to standing requirements that concern private applicants. The criterion of individual concern is particularly difficult to meet.

52.      In Plaumann, the Court held that persons other than those to whom a decision is addressed may claim to be individually concerned only if that decision affects them by reason of certain attributes which are specific to them or by reason of circumstances in which they are differentiated from all other persons and, by virtue of those factors, distinguishes them individually just as in the case of the person addressed. (10)

53.      The Court has remained unwavering in its approach, despite invitations to reconsider that test and proposals outlining alternatives thereto. (11)

54.      The test has also been adapted to the specific context of State aid law, where the decisions adopted by the Commission are addressed only to the Member State concerned.

55.      In that particular context, an undertaking that wishes to bring an action against a Commission decision prohibiting an aid scheme will not meet the criterion of individual concern on the sole basis that that undertaking belongs to the sector in question and that it is a potential beneficiary of the scheme. (12) In that regard, the Court has explained that the possibility of determining more or less precisely the persons to whom a measure applies does not automatically imply that that measure is of individual concern to them. (13)

56.      More specifically, individual concern requires adherence to a closed, identifiable group at the time when the decision at issue was adopted. (14) That is most notably the case of actual beneficiaries of aid, that is to say, those undertakings that have received a positive benefit. (15) That is not the only such case however. The Court has also allowed actions, under certain specific circumstances, brought by competitors of beneficiaries of State aid measures. (16)

57.      The Court’s approach in the field of State aid law shows that the criterion of individual concern as set out in Plaumann is fulfilled where the private applicant can be individualised on the basis of certain specific attributes. Those attributes can relate, inter alia, to the substantive impact of the aid on the position of the competitor on the market or to the fact that the undertaking has effectively received a positive benefit from State resources.

58.      In the judgment under appeal, the General Court considered that, in the light of the specific factual and legal situation applicable to HBH, that undertaking was individually concerned by the decision at issue within the meaning of Plaumann. The General Court based that conclusion on the fact that HBH would have benefited from the restructuring clause by the end of the fiscal year (2009). The German authorities did not retain any leeway with regard to the application of the clause. In that regard, the General Court placed particular emphasis on HBH’s ‘acquired right’, which was certified by the German authorities by way of the binding information. This sets it apart from any other undertaking that met the conditions for benefiting from the restructuring clause. (17)

59.      The Commission considers that in so doing the General Court erred in law. However, contrary to what the Commission suggests, determining whether a private applicant is individually concerned by a Commission decision in the field of State aid is by no means an arithmetic exercise: the case-law is not premised on a binary logic based on the distinction between potential and actual beneficiaries of aid. As I shall illustrate in the following, what is decisive under the Plaumann test — as applied in the context of State aid law — is that the applicant can be distinguished from other undertakings on the basis of special characteristics. Those characteristics may vary from case to case.

2.      Determining individual concern in the context of tax measures: the present case

60.      As the present case illustrates, it is a matter of contention how the concept of an actual beneficiary of aid ought to be construed in the particular context of taxation. Indeed, drawing a clear-cut distinction between potential and actual beneficiaries in that context may prove particularly difficult. That is because tax measures rarely involve a concrete payment of aid.

61.      In essence, the Commission argues that that distinction is crucial because only actual beneficiaries having effectively received aid meet the criterion of individual concern. Only those beneficiaries may, in its submission, bring an action for annulment against a decision declaring aid incompatible with the internal market.

62.      The case-law does not lend support to the Commission’s position. In fact, it seems to me that the Commission attempts artificially to extrapolate from the Court’s case-law a rule of general application that simply is not there.

63.      To support its position, the Commission relies essentially on two lines of cases. It bases its interpretation on the line of authority devolving from the Court’s judgments in Italy and Sardegna Lines (18) and Comitato ‘Venezia vuole vivere’. (19) In both cases, the Court accepted that beneficiaries of individual aids granted under an aid scheme of which the Commission has ordered recovery are, by that fact, individually concerned within the meaning of Article 263, fourth paragraph, TFEU. In addition, it follows from Italy and Sardegna Lines that, in contrast to such beneficiaries, potential beneficiaries that simply belong to the sector concerned do not have standing. (20)

64.      In my view, the lessons that can be drawn from those cases are limited. On the one hand, those cases confirm that actual beneficiaries of aid (that is, those who have effectively received aid that has been ordered to be recovered) should be allowed to bring an action against a decision declaring aid incompatible with the internal market. On the other hand, that right does not extend to undertakings that are simply potential future beneficiaries of an aid scheme.

65.      The inherent logic here is that the group of actual beneficiaries can be distinguished — within the meaning of Plaumann — from undertakings that have not benefited from aid. To be sure, this distinction is particularly helpful in the context of aid schemes that involve a transfer of funds from the State to the undertakings concerned.

66.      However, no definitive conclusion can in my view be drawn from those cases with regard to other undertakings that may, in specific circumstances, be individually concerned by a decision declaring aid incompatible with the internal market.

67.      Fundamentally, the distinction between actual and potential beneficiaries is simply a terminological tool employed to distinguish, in the abstract, certain categories of undertakings that are, or are not, individually concerned by such a decision.

68.      As I see it, the relevant legal test remains that laid down in Plaumann: does the applicant belong to a closed group that can be identified in the light of specific attributes that distinguish it from others?

69.      Seen from that angle, it is clear that attributes other than those related to the applicant’s status as an effective beneficiary of aid may be of relevance in that assessment. Obviously, what those attributes are is a question that cannot be resolved in the abstract. Rather, that assessment depends heavily upon the circumstances.

70.      Against that backdrop, I have difficulty in accepting the Commission’s argument that Belgium and Forum 187 (21) and FrieslandCampina (22) cannot be relied upon in the present circumstances. Those cases, which arose in the context of transitional measures regarding aid declared incompatible with the internal market, illustrate the Court’s willingness to grant standing to applicants that have taken necessary steps to benefit from the impugned national measure without effectively receiving a benefit. (23)

71.      Contrary to what is argued by the Commission, I also fail to see why Stichting Woonpunt (24) and Stichting Woonlinie (25) are of no relevance here. In those cases, the Court considered that individual concern was established by the circumstance that, before the decision in question was adopted, the applicants had acquired a right to make use of the fiscal advantage subsequently declared incompatible with the internal market. (26)

72.      Given the nature of the measure at issue here, it stands to reason that the General Court sought inspiration from the abovementioned cases. In fact, when the decision at issue was adopted, HBH did not only meet the general conditions for the application of the restructuring clause in the abstract. It had also received binding information and a notice of advance payment relating to corporation tax for the tax year 2009, which took account of the losses carried forward pursuant to the restructuring clause. That is what the General Court described as an ‘acquired right’.

73.      Precisely because of the binding information and the notice of advance payment, HBH’s position is fundamentally different from undertakings that simply meet the general conditions for the application of the restructuring clause. (27) That is why the appellant meets the criterion of individual concern.

74.      To be sure, the General Court’s choice of wording is an unhappy one. As the Commission’s arguments illustrate, that choice may lead to unintended parallels being drawn to the principle of legitimate expectations that arises from EU law. Yet, as I understand it, the reference to an ‘acquired right’ simply attempts to describe the factual and legal circumstances that set HBH apart from other undertakings for the purposes of individual concern within the meaning of Plaumann in this particular case.

75.      I would point out that, in the context of taxation, it may be particularly difficult to determine the decisive point at which an undertaking has effectively received aid. Pinning down the relevant point in time is, to a certain extent, arbitrary. Is it the moment at which binding information was issued or the moment when HBH received the notice for advance payment? Or is it, as the Commission argues, the moment at which the tax saving is established in a final tax decision (bearing in mind that such a decision may be established several years after the end of the tax year in question)? Or is it some other point in time?

76.      Clearly, in determining whether an undertaking is individually concerned by a decision declaring aid incompatible with the internal market, the uncertainties and arbitrariness involved in choosing any of those alternatives, or indeed, another one, are beyond doubt.

77.      Given those uncertainties, the question of whether the applicant has effectively received aid should be of only secondary importance here. Rather, as the case-law makes clear, the criterion of individual concern is met when the applicant can, on the basis of specific attributes, be distinguished from other undertakings. (28) That is precisely the case of HBH.

78.      For all those reasons, I find no fault in the judgment under appeal as regards the assessment of individual concern within the meaning of Article 263, fourth paragraph, TFEU. Accordingly, the cross-appeal brought by the Commission ought to be dismissed as unfounded.

B.      The appeal: Did the General Court err in considering that the measure at issue confers a selective advantage on ailing companies?

79.      HBH has put forward two grounds of appeal. Those grounds are intrinsically linked.

80.      In the first ground of appeal, the appellant takes issue with the reasoning of the General Court. It alleges that the judgment under appeal is vitiated by procedural errors related to the obligation to state reasons because the General Court’s reasoning is insufficient or contradictory as regards (1) the determination of the reference system, (2) the assessment of the legal and factual situation of the undertakings requiring restructuring and of the restructuring clause as a ‘general measure’, and (3) the justification of the measure at issue.

81.      The second ground of appeal takes issue with the same aspects of the judgment, but from a substantive viewpoint. In that ground of appeal, HBH claims that the judgment under appeal breaches Article 107(1) TFEU, because of legal errors or distortion of national law in (1) the determination of the reference system, (2) the assessment of the legal and factual situation of the undertakings requiring restructuring and the restructuring clause as a ‘general measure’, and (3) the justification of the measure at issue.

82.      The Commission objects to the appellant’s arguments. As regards the first ground of appeal pertaining to reasoning, that institution considers, in essence, that that ground is based on a flawed reading of the judgment under appeal. As regards, on the other hand, the second ground of appeal, the Commission maintains that that ground is inadmissible. In the alternative, the Commission argues that the second ground of appeal is unfounded.

83.      Given the overlap between the two grounds of appeal, I shall deal with them jointly in steps. That is necessary, in particular, because rather than alleging an absence of a statement of reasons, the first ground of appeal indirectly calls into question the substantive assessment undertaken by the General Court.

84.      First, I shall assess the arguments regarding the determination of the reference system. For the reasons described below, I take the view that the first part of the second ground of appeal is well founded. For that reason, the judgment under appeal, as well as the decision at issue, ought to be set aside to the extent that the reference system for assessing the selectivity of the measure at issue was determined erroneously.

85.      In case the Court disagrees with my assessment of the first part of the second ground of appeal, I shall also examine briefly the other arguments put forward by the appellant.

86.      Second, therefore, I shall examine the arguments regarding the assessment of the legal and factual situation of the undertakings requiring restructuring and the restructuring clause as a ‘general measure’. Third, and last, I shall consider the arguments presented concerning the justification of the measure at issue.

87.      Before doing that, however, some introductory remarks regarding the concept of selectivity in the particular context of taxation are in order.

1.      Preliminary remarks on the assessment of selectivity in the context of direct tax measures

(a)    The concept of selectivity — an assessment in three steps

88.      First of all, it is useful to make some observations regarding the rationale underlying the concept of selectivity and the purpose it serves. It is also helpful to call to mind the analytical framework that informs the assessment of selectivity and the difficulties that arise in applying that framework in the context of direct tax measures.

89.      Under Article 107(1) TFEU, 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 is, in so far as it affects trade between Member States, incompatible with the internal market. Accordingly, four cumulative criteria must be met for a Member State measure to fall within the scope of that provision. First, there must be an advantage. Second, that advantage must be selective. Third, there must be an intervention by the State or through State resources. Fourth, the intervention must be liable to affect trade between the Member States.

90.      In the context of taxation, the question of what constitutes a selective advantage is arguably the most contentious one.

91.      In general terms, the criterion of selectivity serves to identify measures which favour certain undertakings (taxpayers) or the production of certain goods over others.

92.      A tax advantage resulting from a ‘general measure’ applicable without distinction to all economic operators does not constitute State aid. That is because such a measure is not selective. By contrast, a measure that places the recipients in a more favourable position than other taxpayers is capable of procuring a selective advantage for the recipients and, consequently, of constituting State aid, within the meaning of Article 107(1) TFEU. (29)

93.      Therefore, the crux of the assessment of selectivity lies in the comparison of undertakings. As the Court has held, the question to be determined is whether a measure favours certain undertakings or the production of certain goods within the meaning of Article 107(1) TFEU in comparison with other undertakings which are in a comparable legal and factual situation. (30)

94.      However, before it can be assessed whether recipients and other undertakings are in a comparable situation, the reference system must be established.

95.      The Court’s recent judgment in World Duty Free (31) sets out the parameters on the basis of which the selectivity of a tax measure ought to be assessed. The Court has distinguished three steps in the assessment of selectivity of tax measures.

96.      According to the Court,in order to classify a tax measure as selective, the ordinary or normal tax system applicable in the Member State concerned must be identified (step one). Next, it must be shown that the tax measure at issue derogates from that ordinary system, in so far as it differentiates between operators who, in the light of the objective pursued by that system, are in a comparable factual and legal situation (step two). (32) If that is so, it must be verified whether the differentiation is justified to the extent that it derives from the nature or general structure of the system of which the measure forms part (step three). (33)

97.      More specifically, the Court explained that the condition of selectivity is satisfied where the Commission can demonstrate that the impugned measure derogates from the ordinary or ‘normal’ tax system applicable in the Member State concerned, thereby introducing, through its actual effects, differences in the treatment of operators, although the operators who qualify for the tax advantage and those who do not are in a comparable factual and legal situation in the light of the objective pursued by that Member State’s tax system. (34)

98.      The present appeal illustrates the difficulties involved in determining the (material) selectivity of a tax measure on the basis of those parameters. In particular, it highlights the difficulties involved in determining, on the basis of an objective set of criteria, the reference system under step one. (35) That is why it is also useful to make certain introductory observations on the criteria that ought to inform the determination of the reference system before dealing with the appeal in more detail.

(b)    The crucial importance of the reference system for the assessment of selectivity and the criteria to be employed in determining that system

99.      The criterion of selectivity is employed to identify measures that entail unjustified differential treatment of undertakings that are in a legally and factually comparable situation. Yet such a comparison only makes sense against a benchmark. That is why the proper identification of a reference system is of crucial importance for the assessment of selectivity. (36)

100. Nonetheless, the Court’s case-law remains silent on how the relevant reference system ought to be determined. The Court has simply explained that the reference system is the common or ‘normal’ tax regime applicable in the Member State concerned. (37) As a criterion of assessment that statement is remarkably unhelpful.

101. Nevertheless, the Court’s reluctance to devise exact criteria is not surprising. That is because the determination of the reference system entails that the general level of taxation to which undertakings are subject under a national tax system must be identified. In contrast to other types of aid schemes, determining with precision such a common, generally applicable system is replete with uncertainty in the context of taxation. Keeping in mind the complexities of any tax system and the variables involved in determining the tax burden of undertakings, it seems impossible to know with certainty what the ‘normal situation’ is.

102. That is not so with positive benefits. For example, in the case of beneficial loan conditions or mining authorisations that only benefit a limited number of undertakings, identifying the ‘normal situation’ that existed before the impugned measure was adopted remains a relatively straightforward exercise.

103. The Commission describes the reference system as being comprised of a consistent set of rules that apply generally on the basis of objective criteria to all undertakings falling within its scope as defined by its objective: 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. Regarding tax measures in particular, the reference system is based on such elements as the tax base, the taxable persons, the taxable event and the tax rates. (38)

104. Arguably, any number of fiscal provisions, or the combination thereof, can fit that description.

105. In that regard, a question was put to the Commission at the hearing. Questioned on the criteria to be employed in determining the reference system, the Commission was unable to explain on what basis it determines the reference system. It described that process as a search for logic in the system. If anything, the Commission’s response seems to confirm that the determination of the reference system in a particular case is not, in reality, based on an objective set of criteria.

106. That being said, certain conclusions can be drawn from the Court’s case-law. A perusal of that case-law suggests that a broad approach is to be favoured in determining the reference system. Such an approach takes into account all relevant legislative provisions as a whole, or the broadest possible reference point. (39) In addition, it emerges from the case-law that the determination of the reference system ought not to be a formalistic exercise. (40)

107. I observe, for example, that in World Duty Free the Court endorsed the Commission’s view that the relevant benchmark was not the rules governing investments abroad, but rather the Spanish corporate tax system as a whole. Against that benchmark, the Commission had concluded that a selective advantage had been conferred on a particular category of undertakings. Undertakings taxable in Spain that acquired shareholdings of at least 5% in foreign undertakings were treated more favourably as compared to undertakings taxable in Spain making identical investments in domestic companies even though those two categories of undertakings were in comparable situations in the light of the objective pursued by the general Spanish system for the taxation of companies. (41)

108. On the other hand, in the judgment in Gibraltar, the Court confirmed that the reference system can be comprised of a number of different tax rules. It was on such a basis that the Commission had concluded that certain undertakings (offshore companies) had been given a selective advantage. That was so even though, formally speaking, those undertakings were liable to the same tax burden as other undertakings. In that case, the Court also confirmed that regulatory technique ought not to be afforded importance in determining the reference system. (42)

109. To put it simply, it can be inferred from the case-law that the Court has endorsed an approach that seeks to identify the entire body of rules that influence the tax burden weighing on undertakings. In my view, such an approach is warranted. It ensures that the selectivity of a tax measure is assessed against a framework that includes all relevant provisions, and not against provisions that have been carved out artificially from a broader legislative framework. It also ensures that steps one and two are not conflated in the assessment of selectivity. That is because a narrower approach would require the identification of the undertakings that are in a comparable legal and factual situation. Indeed, it should not be forgotten that the determination of the reference system should precede the comparison of undertakings.

110. I shall examine the present appeal in the light of those considerations.

2.      Step one: determining the reference system

(a)    Arguments of the parties

111. In the first part of the first ground of appeal, HBH claims that the reasoning of the General Court is insufficient, or contradictory, as regards the determination of the reference system. (43) That is in the appellant’s view so because the selectivity of the restructuring clause is assessed in the judgment under appeal in relation to the rule governing the forfeiture of losses, even though the General Court acknowledged that the measure at issue should be assessed in the light of a more general rule; namely the loss carry-forward rule. (44)

112. The Commission considers that the reasoning of the General Court is sound and by no means contradictory.

113. In the first part of the second ground of appeal, HBH claims that the General Court misconstrued Article 107(1) TFEU because of the errors related to the determination of the reference system. (45) The appellant also argues that the General Court incorrectly conflated steps one and two of the assessment of selectivity in the judgment under appeal. (46)

114. As its primary contention, the Commission claims that the arguments presented by the appellant relate to findings of fact and should be rejected as inadmissible. In the alternative, the Commission argues that those arguments are unfounded. Nothing in the case file lends support to HBH’s position that the loss carry-forward rule ought to have been considered as the relevant rule for assessing the selectivity of the restructuring clause.

115. I shall begin by addressing the issue of admissibility.

(b)    Assessment

(1)    The determination of the reference system is a question of law reviewable on appeal

116. As has become customary in appeals before the Court, the Commission contests the admissibility of the appellant’s arguments. It argues that the appellant’s arguments relating to the determination of the reference system are not amenable to review by the Court. That is because they concern findings of fact.

117. The Commission’s arguments should be rejected outright.

118. True, the General Court has, by dint of Article 58 of the Statute of the Court of Justice and Article 256 TFEU, exclusive jurisdiction to find the facts and assess those facts. Indeed, appeals before the Court are limited to points of law. In that regard, it is settled case-law that the determination of the content of national law forms part of the assessment of facts not amenable to review on appeal. (47)

119. Nonetheless, the Court may review the legal characterisation of those facts and the legal conclusions the General Court has drawn from them. (48) It may also determine whether national law was distorted where the alleged distortion can be detected without a fresh assessment of the facts or the evidence. (49)

120. The arguments put forward by HBH regarding the determination of the reference system cannot escape review on appeal.

121. Here, a point that should not be overlooked is that the parties agree on the content of national law. Yet they disagree on the objectives of that law. Disagreement also persists regarding the legal characterisation of that law in the light of EU State aid rules. More specifically, as regards step one, at issue is the legal assessment of selectivity that the General Court made on the basis of the relevant facts.

122. Clearly, the proper construction of the concept of selectivity is a question of law, not fact.

123. Adopting the rigid approach advocated by the Commission would mean that a question of fundamental importance for the assessment of selectivity would systematically fall outside the Court’s jurisdiction. Considering that the determination of the reference system has a decisive impact on the following two steps in the assessment of selectivity, such an approach should not, in my view, be endorsed. As I have explained, the reference system constitutes the very benchmark against which the selectivity of a tax measure is to be assessed.

(2)    The General Court erred in determining the reference system

(i)    The legal framework in which the measure at issue operates and the reasoning of the General Court

124. The legal framework of which the restructuring clause forms part encompasses three sets of rules.

125.  First, the loss carry-forward rule applies to all undertakings in accordance with Article 8(1) of the KStG. It reflects the principle that taxpayers are taxed on the basis of their ability to pay. Second, the rule governing the forfeiture of losses, as laid down in Article 8c(1) of the KStG, is an exception to that rule. This is because it excludes the prejudicial acquisition of shareholdings (25% or over) from the scope of the general rule. Third, the restructuring clause, as set out in Article 8c(1a) of the KStG, carves out from the scope of the exception (that is to say, from the rule governing the forfeiture of losses) certain specific situations. By dint of that measure, the situations defined therein (namely the restructuring of ailing companies) are no longer covered by the rule governing the forfeiture of losses. Those situations fall back under the more general rule which allows an undertaking to carry forward losses.

126. Given the chosen legislative technique, the selectivity of the measure at issue is heavily dependent on perspective. In fact, depending on whether the general rule on loss carry-forward or the exception to that rule, namely the rule governing the forfeiture of losses, is employed as the benchmark for assessing the selectivity of the restructuring clause, the conclusion on selectivity will look very different. (50)

127. More specifically, if the rule governing the forfeiture of losses is carved out from the broader legislative context of the loss carry-forward rule, the restructuring clause becomes an exception to the rule governing the forfeiture of losses. Contrariwise, if the loss carry-forward rule is included in the reference system, the restructuring clause no longer constitutes an obvious derogation from the reference system liable to confer a selective advantage to certain undertakings. Rather, it becomes an intrinsic part of the reference system itself.

128. In that regard, the judgment under appeal is not a model of clarity.

129. In particular, the following statement could be a source of confusion: ‘… it is clear that the rule governing the forfeiture of losses, like the loss carry-forward rule, is part of the legislative framework in the context of which the measure at issue arises. In other words, the general loss carry-forward rule, as limited by the rule governing the forfeiture of losses, constitutes the relevant legal framework in the present case, and it is precisely within that framework that it is appropriate to check whether the measure at issue differentiates between operators in a comparable factual and legal situation …’. (51)

130. Considered in isolation, that statement could be interpreted (as the appellant does) as meaning that the General Court considered that the loss carry-forward rule and the rule governing the forfeiture of losses together constitute the reference system. Nonetheless, the appellant’s arguments alleging insufficient or contradictory reasoning are in my view based on a flawed reading of the judgment under appeal.

131. A closer look at the judgment under appeal reveals that the General Court took as a starting point that the rule governing the forfeiture of losses constitutes the reference system.

132. True, the General Court acknowledged the existence of a more general rule (the loss carry-forward rule). It also noted that the reference system as determined by the Commission constitutes an exception to that general rule. Nevertheless, it went on to explain the reasons why it considered that the rule governing the forfeiture of losses constitutes the relevant reference system for assessing the selectivity of the restructuring clause.

133. Specifically, the General Court explained that the rule governing the forfeiture of losses restricts the use of the loss carry-forward rule in the event of the acquisition of a shareholding equal to or greater than 25% of the share capital and withdraws it in the event of the acquisition of more than 50% of the share capital. Therefore, that rule applies systematically to all cases of a change of ownership of 25% or more of a company’s share capital, without drawing any distinction on the basis of the nature or characteristics of the undertakings concerned. (52) The General Court further observed that the restructuring clause was worded in the form of an exception to the rule governing the forfeiture of losses and applies only to those well-defined situations which are subject to that rule. (53) On that basis, it held that the Commission had correctly defined the reference system as being comprised of the rule governing the forfeiture of losses. (54)

134. In that regard, I am unable to identify any error pertaining to the obligation to state reasons in the judgment under appeal. However, I am of the opinion that the judgment under appeal is vitiated by a substantive error in law with regard to the determination of the reference system. That error amounts to a misapplication of Article 107(1) TFEU.

135. To understand why, it is necessary to look also at the reasons set out in the decision at issue. That is so because the General Court reaffirmed the Commission’s assessment of the legal framework in which the restructuring clause operates.

(ii) The General Court determined the reference system in the light of the legislative technique employed and, in so doing, conflated steps one and two

136. In the decision at issue, after setting out the national legal framework and the changes introduced thereto in chronological order, the Commission explained that the restructuring clause laid down in Article 8c(1a) of the KStG differs from the former rule in one significant aspect. That aspect was, according to the Commission, of crucial importance for State aid purposes. (55)

137. More specifically, the Commission explained that under Article 8c(1) of the KStG, a company forfeits its loss carry-forward entirely where more than half of its shares are transferred, unless the restructuring clause falls to be applied. The general rule is therefore the forfeiture of losses on significant changes in ownership and the restructuring clause laid down in Article 8c(1a) of the KStG constitutes the exception to the general rule. (56)

138. By contrast, under the former rule governing the forfeiture of losses (laid down in the repealed Article 8(4) of the KStG), the general rule was continuation of loss carry-forwards in the case of significant changes in ownership, provided that the company was economically identical. That exception was intended to prevent abuse, for example in the form of trading in empty-shell companies. (57)

139. On the face of it, the explanation why the rule governing the forfeiture of losses should be regarded as the reference system is appealing. Yet, on closer inspection, it becomes apparent that that explanation is based on a distinction without a difference.

140. That is because the only difference between the old and the new system lies in the form of the measure. The alleged difference is conditioned on the legislative technique employed by the Member State concerned. Keeping in mind the Court’s Gibraltar case-law, such an approach is far from satisfactory. (58)

141. It should not be forgotten that, in the judgment under appeal, the rule governing the forfeiture of losses is considered as the relevant reference system because the restructuring clause constitutes an exception to that rule. Under the former rule governing the forfeiture of losses that was not so.

142. A comparison between the former rule governing the forfeiture of losses and the new one reveals the flaws in the reasoning based on legislative technique. It shows that the two rules simply approach the issue of restricting loss carry-forward from different angles.

143. Under the former rule governing the forfeiture of losses, emphasis was placed on the economic identity of the acquired undertaking. While economically identical entities were allowed to carry forward losses, those that had changed economic identity as a result of a change of ownership were not. Expressed in terms of an example of ‘economically identical’, the restructuring clause formed an inherent part of the former rule governing the forfeiture of losses itself. (59)

144. While the new rule on the forfeiture of losses is arguably more precise (and so is the restructuring clause), I fail to see how the change of legislative technique — and enhanced precision in the rules in question — could be held to be of crucial importance for State aid purposes. Indeed, like the former rule governing the forfeiture of losses, the (new) rule governing the forfeiture of losses merely limits the possibility of deducting previously incurred losses in the future in strictly defined situations pertaining to change of ownership. Barring that exception, the possibility of carrying forward losses remains possible.

145. Materially, therefore, the restructuring clause simply limits the scope of the rule governing the forfeiture of losses. (60) In that sense, the restructuring clause forms an inseparable part of the general rule, namely the loss carry-forward rule.

146. Apart from the legislative technique employed, there is nothing in the case file that would help explain why, under the regime under consideration, the loss carry-forward rule should not form part of the reference system.

147. In that regard, a corollary problem arises in the judgment under appeal. By adopting a narrow approach based on legislative technique and by carving the rule governing the forfeiture of losses out of its broader legislative framework, the judgment under appeal also conflates steps one and two in the assessment of selectivity. In fact, the General Court relied on the comparability of undertakings that undergo a change of ownership in order to determine the reference system. A closer look reveals that the General Court began its reasoning with the assumption that all undertakings that have undergone a substantial change in ownership are necessarily in a legally and factually comparable situation, rather than identifying the reference system first. (61) Indeed, it is only possible to conclude that the loss carry-forward rule does not form part of the reference system if it is assumed that undertakings that undergo a change of ownership are in a comparable situation.

148. I therefore consider that the General Court erred in determining the reference system. Specifically, by basing its conclusion on the legislative technique employed and on the comparability of undertakings that have undergone a substantial change in ownership, the General Court artificially limited the reference system to exclude the loss carry-forward rule therefrom.

149. On that basis, I am of the opinion that the first part of the second ground of appeal ought to be upheld.

150. To conclude on this point, I observe that in the judgment under appeal the rule governing the forfeiture of losses is employed as the benchmark for assessing the selectivity of the measure at issue (steps two and three). If the Court were to uphold the first part of the second ground of appeal, as I suggest, there would be no need to consider the remainder of the arguments that concern the assessment of selectivity with regard to steps two and three. That is because the fate of the measure at issue depends, in the final analysis, on the definition of the reference system.

151. Nevertheless, in case the Court disagrees with me, I shall make the following observations regarding that assessment.

3.      Step two: comparing the legal and factual situation of undertakings

(a)    Arguments of the parties

152. In the second part of the first ground of appeal, HBH claims that the reasoning of the General Court is insufficient or contradictory. More specifically, in the first limb of the second part of the first ground of appeal, it claims that — bearing in mind how the reference system was defined in the judgment under appeal — the General Court did not adequately explain why it was possible to compare the legal and factual situation of undertakings requiring restructuring with that of healthy undertakings. (62) In the second limb, the appellant takes issue with the reasons set out in the judgment under appeal to explain why the restructuring clause was not to be regarded as a ‘general measure’ potentially open to all undertakings. (63)

153. The Commission maintains that the reasoning of the General Court is sound and sufficient.

154. In the second part of the second ground of appeal, HBH alleges a breach of Article 107(1) TFEU. It argues that the General Court erred in its assessment of the (prima facie) selectivity of the measure at issue. By considering that ailing undertakings in need of restructuring and healthy undertakings are in a comparable situation in the light of the objective of the tax system, the General Court erred in law. (64)

155. More specifically, HBH argues that the General Court erred in defining the objective of the rule governing the forfeiture of losses (first limb). (65) Moreover, the appellant argues that in holding that the measure at issue was prima facie selective and not a ‘general measure’ potentially open to all undertakings, the General Court erred in law (second limb). (66) That is, in essence, because the judgment under appeal departed from the case-law established in the General Court’s judgment in Autogrill. (67)

156. The Commission finds no fault in the judgment under appeal regarding the assessment of selectivity.

157. As its primary contention, however, the Commission maintains that the second part of the second ground of appeal is inadmissible. On the one hand, the arguments put forward in the first limb concern the finding and assessment of facts. On the other hand, the second limb deals with an issue which did not form part of the subject matter of the dispute at first instance. Therefore, by allowing the appellant’s arguments on the question of the characterisation of the measure at issue as a general measure, the General Court ruled ultra petita. (68)

(b)    Assessment

(1)    The appellant’s arguments pertaining to the definition of the objective of the reference system

158. On the one hand, HBH’s arguments concerning insufficient or contradictory reasoning in the first limb of the second part of the first ground of appeal relate to a point examined by the General Court for the sake of completeness. They are also based on the same (flawed) premiss as the first part of the first ground of appeal; namely that the General Court considered both the loss carry-forward rule and the rule governing the forfeiture of losses to constitute the reference system. As illustrated above, the General Court found in the judgment under appeal that the rule governing the forfeiture of losses constitutes the relevant reference system. (69) Accordingly, those arguments should be rejected.

159. On the other hand, the appellant’s arguments relating to the breach of Article 107(1) TFEU in the first limb of the second part of the second ground of appeal are also bound to fail. Those arguments are partly inadmissible and partly ineffective.

160. In the first place, those arguments reiterate the position that the General Court considered that the loss carry-forward rule and the rule governing the forfeiture of losses together constitute the reference system. As already explained, that position is based on an erroneous reading of the judgment under appeal.

161. In the second place, by its arguments, the appellant is calling into question the General Court’s finding that the rule governing the forfeiture of losses is designed to ensure that losses are not carried forward in the case of a substantial change in the ownership of an undertaking that has accumulated losses. (70) In the appellant’s view, that rule simply aims to combat abuse by preventing the trade in empty-shell companies.

162. Certainly, any tax regime may pursue a plethora of different objectives. Those objectives include, but are not limited to, generating revenue for the State budget, steering the behaviour of consumers and undertakings and combating problems such as tax avoidance. A tax regime may also be designed to redress the consequences of an economic downturn.

163. Similarly to the determination of the reference system, the comparability of taxpayers in the light of the objective pursued by the tax system under step two of the assessment of selectivity is, in my view, a question of law. That question concerns the legal characterisation of the relevant national provisions under EU State aid rules. (71)

164. Here, however, the appellant asks the Court to (re-)determine the objective of the rule governing the forfeiture of losses. That is not, in my view, a point the Court is competent to deal with on appeal. I agree with the Commission that the Court’s jurisdiction does not extend to reassessing the objective of the reference system. That is purely a question of fact pertaining to the assessment of the content of national law. Such questions are not open to review, save where a distortion of national law is evident from the case file. That is not the case here.

165. Lastly, and in the third place, in the first limb of the second part of the second ground of appeal, HBH also takes issue with a point the General Court examined for the sake of completeness. In the judgment under appeal, the General Court also examined the hypothesis put forward by HBH that the relevant objective of the reference system is to prevent abuse of the loss carry-forward rule by deterring the purchase of empty-shell companies. It concluded that, even on that hypothesis, healthy and ailing companies having undergone a substantial change in ownership were in a comparable situation. (72)

166. Given that the appellant has not been successful in showing that the General Court erred in law in its primary assessment, the arguments put forward in relation to that alternative assessment remain ineffective.

(2)    The appellant’s arguments pertaining to the General Court’s conclusion that the restructuring clause is not a general measure

167. The appellant’s arguments pertaining to the General Court’s conclusion that the restructuring clause is not a ‘general measure’ potentially open to all undertakings can be disposed of swiftly.

168. First of all, as concerns the second limb of the second part of the first ground of appeal (the issue of insufficient reasoning), in the judgment under appeal, the reasons why the measure at issue is not a general measure that is potentially available to all undertakings are exposed briefly, but clearly: it covers only one category of undertakings that are in a specific situation, namely undertakings in difficulty. (73)

169. More fundamentally still, as concerns the second limb of the second part of the second ground of appeal, the General Court correctly applied the relevant case-law.

170. After the written procedure in the present case was closed, the judgment of the General Court that the appellant relies upon here was overturned on appeal. (74) Contrary to what HBH argues, it does not therefore follow from the case-law that selectivity requires the impugned measure to be available only to a particular category of undertakings.

171. In World Duty Free, the Court confirmed that point. It held that for a tax measure to be selective it must not necessarily affect a particular category of undertakings that can be distinguished by reason of specific properties which are common to them and which are characteristic of them. Rather, determining the selectivity of a measure essentially involves ascertaining whether the exclusion of certain operators from the benefit of a tax advantage that arises from a measure derogating from an ordinary tax system constitutes discrimination with respect to those operators. (75)

172. That is the analytical method employed by the General Court in the judgment under appeal. (76) Therefore, it cannot be maintained that the relevant case-law regarding the assessment of selectivity was misapplied in the judgment under appeal.

173. On the above basis, I conclude that the second part of the first and second grounds of appeal ought to be dismissed as partly inadmissible and partly unfounded. (77)

4.      Step three: assessing the existence of a justification intrinsic to the tax system

(a)    Arguments of the parties

174. In the third part of the first ground of appeal, the appellant claims that the General Court did not sufficiently explain the reasons why it rejected the argument put forward by the appellant at first instance regarding the justification of the restructuring clause; namely that the measure at issue ensures that the principle of taxation according to the ability to pay is respected. (78)

175. The Commission does not find fault in the General Court’s reasoning.

176. In the third part of the second ground of appeal, HBH claims that the General Court misconstrued Article 107(1) TFEU by considering that the restructuring clause is designed to promote the restructuring of undertakings in difficulty and it therefore pursues an objective extrinsic to the tax system. (79)

177. The Commission considers that the General Court correctly identified the objective of the measure at issue on the basis of its exclusive jurisdiction to find the facts. As for the arguments pertaining to the principle of taxation according to the ability to pay, those arguments are in the Commission’s submission ineffective, or, in the alternative, unfounded.

(b)    Assessment

178. In the third part of the first and second grounds of appeal, the appellant takes particular issue with the General Court’s conclusion that the objective of the restructuring clause is not intrinsic to the tax system, that is to say, to ensure that taxation is based on the taxpayer’s ability to pay.

179. First, the argument that the General Court erred in law by incorrectly identifying the objective of the measure at issue should be rejected as inadmissible. Similarly to the determination of the objective of the reference system, the identification of the objective of the restructuring clause is, in my view, a matter of fact. (80) Barring obvious distortion, the Court is not competent to review such findings. In the present case, distortion is not apparent from the case file.

180. Second, the remainder of the appellant’s arguments are ineffective. In fact, the arguments pertaining to the alleged errors in the reasoning in the judgment under appeal and those concerning errors in the assessment of HBH’s argument that the restructuring clause is designed to ensure that taxation is based on the taxpayer’s ability to pay relate to an issue which the General Court examined for the sake of completeness.

181. In other words, I agree with the Commission that the appellant’s arguments in the third part of the first and second grounds of appeal are partly inadmissible and partly ineffective.

182. By way of conclusion, I observe that it may be particularly arduous to save a tax measure at this third step of the assessment of selectivity.

183. In fact, the Court has adopted a strict approach to justification. Only a justification stemming from the basic or guiding principles of the tax system are allowed. That is to say, a measure which has been found to be a priori selective can be justified only by reasons related to the nature or general scheme of the tax system. In that regard, the Court has drawn a distinction between, on the one hand, the objectives attributed to a particular tax scheme, which are extrinsic to it and, on the other, the mechanisms inherent in the tax system itself which are necessary for the achievement of such objectives. (81)

184. However clear cut that distinction may seem in theory, the reality of the matter is far more complex.

185. Firstly, that distinction starts on the basis that only reasons related to maintaining the tax base (that is to say, to the need to ensure revenue to the State budget) can validly be relied upon. That is illustrated by the examples given by the Commission: for example, the need to fight tax evasion, the need to take into account specific accounting requirements, administrative manageability and the principle of tax neutrality as well as the need to avoid double taxation could form the basis for a possible justification to save a tax measure. (82) Yet, I am not persuaded that those reasons can be separated meaningfully from other objectives that the State pursues through taxation. A point that should not be overlooked is that, in today’s world, taxation is also a tool employed by the State to steer behaviour. In other words, reasons that in the Court’s taxonomy are intrinsic to the tax system are inseparably linked to objectives that have a broader societal context. I have in mind, in particular, objectives such as the need to maintain employment, to protect the environment and to ensure regional development or the equal treatment of taxpayers.

186. Secondly, and even more importantly, any tax scheme that has been found to afford a (a priori) selective advantage to certain undertakings erodes the tax base. That is because a selective fiscal advantage alleviates the tax burden weighing on certain undertakings.

187. It is therefore not surprising that, to my knowledge, the Court has yet to accept the reasons relied upon by Member States under the third step of the assessment of selectivity. From that it could be inferred that we are dealing with a de facto irrefutable assumption that tax measures found to be a priori selective are, in reality, selective.

C.      Consequences of the assessment

188. Under the first paragraph of Article 61 of the Statute of the Court of Justice, the Court is to set aside the judgment of the General Court if the appeal is well founded. Where the proceedings so permit, it may itself give final judgment in the matter. It may also refer the case back to the General Court.

189. I have concluded that the first part of the second ground of appeal ought to be upheld. If the Court agrees with my assessment, I would advise the Court to give final judgment in the matter.

190. The legal error identified in the determination of the reference system entails that the judgment under appeal ought to be set aside to the extent that it dismissed the appellant’s action as unfounded. That is because the assessment of selectivity of the measure at issue is skewed by that legal error. That legal error led the General Court to confirm that the Commission had correctly defined the reference system in the decision at issue. The selectivity of the restructuring clause has therefore been assessed against a benchmark (the rule governing the forfeiture of losses) that I have found to be legally erroneous. In other words, the assessment of selectivity in the decision at issue is based on an erroneous premiss. That is why that decision must be annulled too.

191. Should the Court, instead, disagree with my assessment of that issue, the appeal should then be dismissed in its entirety.

D.      Costs

192. Under Article 138(1) of the Rules of Procedure of the Court, the unsuccessful party is to be ordered to pay the costs if they have been applied for in the pleadings of the successful party.

193. If the Court agrees with my assessment of the appeal, then, in accordance with Articles 137, 138 and 184 of the Rules of Procedure, the Commission should be ordered to pay the costs of the present proceedings, both at first instance and on appeal.

V.      Conclusion

194. In the light of the foregoing considerations, I propose that the Court:

–        dismiss the cross-appeal brought by the Commission;

–        set aside the judgment of the General Court of 4 February 2016 in Case T‑287/11, Heitkamp BauHolding v Commission, to the extent that it dismissed the action as unfounded;

–        annul Commission Decision 2011/527/EU of 26 January 2011 on State aid C 7/10 (ex CP 250/09 and NN 5/10) implemented by Germany — Scheme for the carry-forward of tax losses in the case of restructuring of companies in difficulty (Sanierungsklausel);

–        order the Commission to pay the costs.


1      Original language: English.


2      Judgment of 4 February 2016, Heitkamp BauHolding v CommissionHeitkamp BauHolding v CommissionHeitkamp BauHolding v Commission, T‑287/11, EU:T:2016:60 (‘the judgment under appeal’).


3      Decision of 26 January 2011 on State aid C 7/10 (ex CP 250/09 and NN 5/10) implemented by Germany — Scheme for the carry-forward of tax losses in the case of restructuring of companies in difficulty (Sanierungsklausel) (OJ 2011 L 235, p. 26, ‘the decision at issue’).


4      Judgment of 15 July 1963, Plaumann v CommissionPlaumann v CommissionPlaumann v Commission, 25/62, EU:C:1963:17.


5      Judgment of 21 December 2016, Commission v World Duty Free Group and OthersCommission v World Duty Free Group and OthersCommission v World Duty Free Group and Others, C‑20/15 P and C‑21/15 P, EU:C:2016:981.


6      Judgments of 6 September 2006, Portugal v CommissionPortugal v CommissionPortugal v Commission, C‑88/03, EU:C:2006:511, paragraph 56, and of 21 December 2016, Commission v Hansestadt LübeckCommission v Hansestadt LübeckCommission v Hansestadt Lübeck, C‑524/14 P, EU:C:2016:971, paragraph 55.


7      Paragraphs 50 to 79 of the judgment under appeal.


8      Judgment of 15 July 1963, Plaumann v CommissionPlaumann v CommissionPlaumann v Commission, 25/62, EU:C:1963:17.


9      I observe that the General Court has recently considered that a decision adopted by the Commission — a decision in part finding no State aid and in part declaring aid incompatible with the internal market but not ordering recovery thereof — constitutes a regulatory act that does not entail implementing measures within the meaning of Article 263, fourth paragraph, TFEU. See judgments of 15 September 2016, Ferracci v CommissionFerracci v CommissionFerracci v Commission, T‑219/13, EU:T:2016:485, paragraphs 50 to 55, and Scuola Elementare Maria Montessori v CommissionScuola Elementare Maria Montessori v CommissionScuola Elementare Maria Montessori v Commission, T‑220/13, not published, EU:T:2016:484. However, the Court has yet to decide on that issue on appeal (Scuola Elementare Maria Montessori v Commission, C‑622/16; Commission v Scuola Elementare Maria Montessori,C‑623/16; and Commission v Ferracci, C‑624/16, pending).


10      Judgment of 15 July 1963, Plaumann v CommissionPlaumann v CommissionPlaumann v Commission, 25/62, EU:C:1963:17, p. 107.


11      Most notably Opinion of Advocate General Jacobs in Unión de Pequeños Agricultores v CouncilUnión de Pequeños Agricultores v CouncilUnión de Pequeños Agricultores v Council, C‑50/00 P, EU:C:2002:197, point 59 et seq., and judgment of 3 May 2002, Jégo-Quéré v CommissionJégo-Quéré v CommissionJégo-Quéré v Commission, T‑177/01, EU:T:2002:112, paragraph 49. The legislature has, by contrast, been somewhat more open to the possibility of relaxing the standing requirements of private applicants. That openness is attested by the introduction of the category of ‘regulatory acts’ to Article 263, fourth paragraph, TFEU, by the Lisbon Treaty.


12      See, among many, judgments of 2 February 1988, Kwekerij van der Kooy and Others v CommissionKwekerij van der Kooy and Others v CommissionKwekerij van der Kooy and Others v Commission, 67/85, 68/85 and 70/85, EU:C:1988:38, paragraph15; of 19 October 2000, Italy and Sardegna Lines v CommissionItaly and Sardegna Lines v CommissionItaly and Sardegna Lines v Commission, C‑15/98 and C‑105/99, EU:C:2000:570, paragraph 33; and of 19 December 2013, Telefónica v CommissionTelefónica v CommissionTelefónica v Commission, C‑274/12 P, EU:C:2013:852, paragraph 49.


13      Judgment of 22 November 2001, Antillean Rice Mills v CouncilAntillean Rice Mills v CouncilAntillean Rice Mills v Council, C‑451/98, EU:C:2001:622, paragraph 52.


14      Judgments of 27 February 2014, Stichting Woonpunt and Others v CommissionStichting Woonpunt and Others v CommissionStichting Woonpunt and Others v Commission, C‑132/12 P, EU:C:2014:100, paragraphs 59 to 62, and of 27 February 2014, Stichting Woonlinie and Others v CommissionStichting Woonlinie and Others v CommissionStichting Woonlinie and Others v Commission, C‑133/12 P, EU:C:2014:105, paragraphs 46 to 49.


15      Judgment of 9 June 2011, Comitato ‘Venezia vuole vivere’ and Others v CommissionComitato ‘Venezia vuole vivere’ and Others v CommissionComitato ‘Venezia vuole vivere’ and Others v Commission, C‑71/09 P, C‑73/09 P and C‑76/09 P, EU:C:2011:368, paragraph 53.


16      See, for example, judgments of 28 January 1986, Cofaz and Others v CommissionCofaz and Others v CommissionCofaz and Others v Commission, 169/84, EU:C:1986:42, paragraph 25; of 19 May 1993, Cook v CommissionCook v CommissionCook v Commission, C‑198/91, EU:C:1993:197, paragraph 23; of 13 December 2005, Commission v Aktionsgemeinschaft Recht und EigentumCommission v Aktionsgemeinschaft Recht und EigentumCommission v Aktionsgemeinschaft Recht und Eigentum, C‑78/03 P, EU:C:2005:761, paragraph 37 and the case-law cited; and of 9 July 2009, 3F, C‑319/07 P, EU:C:2009:435, paragraph 34 and the case-law cited.


17      Paragraphs 66 to 79 of the judgment under appeal.


18      Judgment of 19 October 2000, Italy and Sardegna Lines v CommissionItaly and Sardegna Lines v CommissionItaly and Sardegna Lines v Commission, C‑15/98 and C‑105/99, EU:C:2000:570.


19      Judgment of 9 June 2011, Comitato ‘Venezia vuole vivere’ and Others v CommissionComitato ‘Venezia vuole vivere’ and Others v CommissionComitato ‘Venezia vuole vivere’ and Others v Commission, C‑71/09 P, C‑73/09 P and C‑76/09 P, EU:C:2011:368.


20      Judgments of 9 June 2011, Comitato ‘Venezia vuole vivere’ and Others v CommissionComitato ‘Venezia vuole vivere’ and Others v CommissionComitato ‘Venezia vuole vivere’ and Others v Commission, C‑71/09 P, C‑73/09 P and C‑76/09 P, EU:C:2011:368, paragraph 53, and of 19 October 2000, Italy and Sardegna Lines v CommissionItaly and Sardegna Lines v CommissionItaly and Sardegna Lines v Commission, C‑15/98 and C‑105/99, EU:C:2000:570, paragraphs 33 and 34.


21      Judgment of 22 June 2006, Belgium and Forum 187 v CommissionBelgium and Forum 187 v CommissionBelgium and Forum 187 v Commission, C‑182/03 and C‑217/03, EU:C:2006:416.


22      Judgment of 17 September 2009, Commission v Koninklijke FrieslandCampinaCommission v Koninklijke FrieslandCampinaCommission v Koninklijke FrieslandCampina, C‑519/07 P, EU:C:2009:556.


23      Judgments of 22 June 2006, Belgium and Forum 187 v CommissionBelgium and Forum 187 v CommissionBelgium and Forum 187 v Commission, C‑182/03 and C‑217/03, EU:C:2006:416, paragraphs 60 to 63, and of 17 September 2009, Commission v Koninklijke FrieslandCampinaCommission v Koninklijke FrieslandCampinaCommission v Koninklijke FrieslandCampina, C‑519/07 P, EU:C:2009:556, paragraphs 56 to 58. See also judgment of 17 January 1985, Piraiki-Patraiki and Others v CommissionPiraiki-Patraiki and Others v CommissionPiraiki-Patraiki and Others v Commission, 11/82, EU:C:1985:18, paragraph 19.


24      Judgment of 27 February 2014, Stichting Woonpunt and Others v CommissionStichting Woonpunt and Others v CommissionStichting Woonpunt and Others v Commission, C‑132/12 P, EU:C:2014:100.


25      Judgment of 27 February 2014, Stichting Woonlinie and Others v CommissionStichting Woonlinie and Others v CommissionStichting Woonlinie and Others v Commission, C‑133/12 P, EU:C:2014:105.


26      Judgments of 27 February 2014, Stichting Woonpunt and Others v CommissionStichting Woonpunt and Others v CommissionStichting Woonpunt and Others v Commission, C‑132/12 P, EU:C:2014:100, paragraphs 59 to 62, and of 27 February 2014, Stichting Woonlinie and Others v CommissionStichting Woonlinie and Others v CommissionStichting Woonlinie and Others v Commission, C‑133/12 P, EU:C:2014:105, paragraphs 46 to 49.


27      As a brief parenthesis, I would also point out that the facts underlying the judgment of 19 December 2013, Telefónica v CommissionTelefónica v CommissionTelefónica v Commission, C‑274/12 P, EU:C:2013:852, are also fundamentally different from the present case. In that case, the applicant had made investments on the basis of the national measure that had later been declared incompatible with the internal market by the Commission. It had thus benefited from the impugned national measure. However, the similarities stop there. The investments made by the applicant had been made before the relevant cut-off date: the Commission’s decision specifically allowed the impugned national measure to continue to apply to investments made before the decision to open a formal investigation was adopted. See, in that regard, paragraphs 47 to 50. See also order of 21 March 2012, Telefónica v CommissionTelefónica v CommissionTelefónica v Commission, T‑228/10, not published, EU:T:2012:140, paragraphs 36 to 40.


28      For that reason, the Commission’s objection that the appellant had not claimed to have received aid before the General Court has no bearing here.


29      Judgment of 21 December 2016, Commission v World Duty Free Group and OthersCommission v World Duty Free Group and OthersCommission v World Duty Free Group and Others, C‑20/15 P and C‑21/15 P, EU:C:2016:981, paragraph 56 and the case-law cited.


30      There is some discrepancy in the case-law as regards the question of whether the comparability of undertakings ought to be established in the light of the objective of the tax regime as a whole or the impugned national measure. For example, in Adria-Wien Pipeline the Court held that undertakings were to be compared in the light of the objective pursued by the measure at issue (judgment of 8 November 2001, Adria-Wien Pipeline and Wietersdorfer & Peggauer Zementwerke, C‑143/99, EU:C:2001:598, paragraph 41 and the case-law cited). By contrast, more recently, the Court held in World Duty Free that undertakings are to be compared in the light of the objective pursued by the ordinary tax system (judgment of 21 December 2016, Commission v World Duty Free Group and OthersCommission v World Duty Free Group and OthersCommission v World Duty Free Group and Others, C‑20/15 P and C‑21/15 P, EU:C:2016:981, paragraph 57 and the case-law cited).


31      Judgment of 21 December 2016, Commission v World Duty Free Group and OthersCommission v World Duty Free Group and OthersCommission v World Duty Free Group and Others, C‑20/15 P and C‑21/15 P, EU:C:2016:981.


32      Judgment of 21 December 2016, Commission v World Duty Free Group and OthersCommission v World Duty Free Group and OthersCommission v World Duty Free Group and Others, C‑20/15 P and C‑21/15 P, EU:C:2016:981, paragraph 57 and the case-law cited. See also judgment of 18 July 2013, P, C‑6/12, EU:C:2013:525, paragraph 19.


33      Judgment of 21 December 2016, Commission v World Duty Free Group and OthersCommission v World Duty Free Group and OthersCommission v World Duty Free Group and Others, C‑20/15 P and C‑21/15 P, EU:C:2016:981, paragraph 58 and the case-law cited.


34      Judgment of 21 December 2016, Commission v World Duty Free Group and OthersCommission v World Duty Free Group and OthersCommission v World Duty Free Group and Others, C‑20/15 P and C‑21/15 P, EU:C:2016:981, paragraphs 57 and 67.


35      For those difficulties, see, for example, O. Peiffert, ‘Comparaison n’est pas raison: Pour une clarification du critère de sélectivité d’une aide d’État’, Concurrences, n 3, 2017, 52 at 60.


36      For a different view regarding the importance of the identification of the ordinary tax system, see Opinion of Advocate General Kokott in ANGED, C‑233/16, EU:C:2017:852, point 88.


37      Judgments of 8 September 2011, Paint Graphos, C‑78/08 to C‑80/08, EU:C:2011:550, paragraph 49 and the case-law cited; of 18 July 2013, P, C‑6/12, EU:C:2013:525, paragraph 19; and of 21 December 2016, Commission v World Duty Free Group and OthersCommission v World Duty Free Group and OthersCommission v World Duty Free Group and Others, C‑20/15 P and C‑21/15 P, EU:C:2016:981, paragraph 57.


38      Commission Notice on the notion of State aid as referred to in Article 107(1) of the Treaty on the Functioning of the European Union, C/2016/2946 (OJ 2016 C 262, p. 1), available at: http://eur-lex.europa.eu/legal-content/EN/TXT/?uri=uriserv:OJ.C_.2016.262.01.0001.01.ENG&toc=OJ:C:2016:262:TOC#ntc205-C_2016262EN.01000101-E0205, points 133 and 134.


39      See, to that effect, judgments of 22 June 2006, Belgium and Forum 187 v CommissionBelgium and Forum 187 v CommissionBelgium and Forum 187 v Commission, C‑182/03 and C‑217/03, EU:C:2006:416, paragraphs 95, 104, 107 and 122; of 6 September 2006, Portugal v CommissionPortugal v CommissionPortugal v Commission, C‑88/03, EU:C:2006:511, paragraph 56; of 17 September 2009, Commission v Koninklijke FrieslandCampinaCommission v Koninklijke FrieslandCampinaCommission v Koninklijke FrieslandCampina, C‑519/07 P, EU:C:2009:556, paragraphs 2 to 7; and of 21 December 2016, Commission v World Duty Free Group and OthersCommission v World Duty Free Group and OthersCommission v World Duty Free Group and Others, C‑20/15 P and C‑21/15 P, EU:C:2016:981, paragraph 68.


40      Judgment of 15 November 2011, Commission and Spain v Government of Gibraltar and United KingdomCommission and Spain v Government of Gibraltar and United KingdomCommission and Spain v Government of Gibraltar and United Kingdom, C‑106/09 P and C‑107/09 P, EU:C:2011:732, paragraphs 92 and 93.


41      Judgment of 21 December 2016, Commission v World Duty Free Group and OthersCommission v World Duty Free Group and OthersCommission v World Duty Free Group and Others, C‑20/15 P and C‑21/15 P, EU:C:2016:981, paragraphs 22 and 67 to 69.


42      Judgment of 15 November 2011, Commission and Spain v Government of Gibraltar and United KingdomCommission and Spain v Government of Gibraltar and United KingdomCommission and Spain v Government of Gibraltar and United Kingdom, C‑106/09 P and C‑107/09 P, EU:C:2011:732, paragraphs 92 to 95.


43      Paragraphs 103 to 106 of the judgment under appeal.


44      Paragraphs 107 to 109 of the judgment under appeal.


45      Paragraphs 103 to 106 of the judgment under appeal.


46      Paragraph 104 of the judgment under appeal.


47      See, for example, judgments of 24 October 2002, Aéroports de Paris v CommissionAéroports de Paris v CommissionAéroports de Paris v Commission, C‑82/01 P, EU:C:2002:617, paragraph 63; of 21 December 2011, A2A v CommissionA2A v CommissionA2A v Commission, C‑318/09 P, not published, EU:C:2011:856, paragraph 125; and of 3 April 2014, France v CommissionFrance v CommissionFrance v Commission, C‑559/12 P, EU:C:2014:217, paragraphs 78 and 79 and the case-law cited.


48      See, among many, judgments of 17 December 1998, Baustahlgewebe v CommissionBaustahlgewebe v CommissionBaustahlgewebe v Commission, C‑185/95 P, EU:C:1998:608, paragraph 23 and the case-law cited, and of 3 April 2014, France v CommissionFrance v CommissionFrance v Commission, C‑559/12 P, EU:C:2014:217, paragraph 78 and the case-law cited.


49      Judgment of 3 April 2014, France v CommissionFrance v CommissionFrance v Commission, C‑559/12 P, EU:C:2014:217, paragraphs 79 and 80 and the case-law cited.


50      See similarly judgment of 18 July 2013,P, C‑6/12, EU:C:2013:525, paragraph 13, in which the Court paraphrases the position of the referring court regarding the alternatives available for determining the reference system.


51      Paragraph 106 of the judgment under appeal.


52      Paragraph 104 of the judgment under appeal.


53      Paragraph 105 of the judgment under appeal.


54      Paragraph 107 of the judgment under appeal.


55      Recital 21 of the decision at issue.


56      Recital 22 of the decision at issue.


57      Recital 23 of the decision at issue.


58      Indeed, the assessment of selectivity should not depend on form. See, similarly, judgment of 15 November 2011, Commission and Spain v Government of Gibraltar and United KingdomCommission and Spain v Government of Gibraltar and United KingdomCommission and Spain v Government of Gibraltar and United Kingdom, C‑106/09 P and C‑107/09 P, EU:C:2011:732, paragraphs 92 and 93.


59      See point 9 above.


60      Here, it is also useful to note that the restructuring clause is not the only provision limiting the scope of the rule governing the forfeiture of losses. In December 2009, two new exceptions were introduced to that rule. On the one hand, regarding all restructurings carried out exclusively within a group of undertakings at the head of which is a single individual or company holding 100% of the shares, the carry-forward of losses was to be maintained. On the other hand, that was also so if at the time of the prejudicial acquisition of a shareholding, the losses corresponded to hidden reserves of the company’s assets.


61      See point 133 above on the reasoning of the General Court.


62      Paragraphs 133 and 134 of the judgment under appeal.


63      Paragraph 141 of the judgment under appeal.


64      Paragraphs 126 to 133 of the judgment under appeal.


65      Paragraphs 128 to 131 of the judgment under appeal.


66      Paragraph 141 of the judgment under appeal.


67      Judgment of 7 November 2014, Autogrill España v CommissionAutogrill España v CommissionAutogrill España v Commission, T‑219/10, EU:T:2014:939.


68      Paragraph 122 of the judgment under appeal.


69      See point 131 et seq. above.


70      Paragraph 128 of the judgment under appeal.


71      See point 118 above.


72      Paragraphs 132 to 134 of the judgment under appeal.


73      Paragraph 141 of the judgment under appeal.


74      Judgments of 7 November 2014, Autogrill España v CommissionAutogrill España v CommissionAutogrill España v Commission, T‑219/10, EU:T:2014:939, and of 21 December 2016, Commission v World Duty Free Group and OthersCommission v World Duty Free Group and OthersCommission v World Duty Free Group and Others, C‑20/15 P and C‑21/15 P, EU:C:2016:981.


75      Judgment of 21 December 2016, Commission v World Duty Free Group and OthersCommission v World Duty Free Group and OthersCommission v World Duty Free Group and Others, C‑20/15 P and C‑21/15 P, EU:C:2016:981, paragraphs 69 to 71 and the case-law cited.


76      Paragraphs 140 and 141 of the judgment under appeal.


77      Given that the second limb of the second part of the second ground of appeal is unfounded, it is not necessary to deal with the Commission’s argument that the General Court ruled ultra petita.


78      Paragraphs 165 and 166 of the judgment under appeal.


79      Paragraphs 158 to 160 and 164 to 170 of the judgment under appeal.


80      See point 164 above.


81      See, for example, judgments of 6 September 2006, Portugal v CommissionPortugal v CommissionPortugal v Commission, C‑88/03, EU:C:2006:511, paragraphs 81 and 82; of 8 September 2011, Paint Graphos, C‑78/08 to C‑80/08, EU:C:2011:550, paragraph 69; of 9 October 2014, Ministerio de Defensa and Navantia, C‑522/13, EU:C:2014:2262, paragraph 42; and of 18 July 2013, P, C‑6/12, EU:C:2013:525, paragraph 29.


82      For those and other examples, Commission Notice, op. cit., point 138. It can further be inferred from the Court’s judgment in Paint Graphos that the avoidance of double taxation may be regarded as a reason intrinsic to the tax system (judgment of 8 September 2011, Paint Graphos, C‑78/08 to C‑80/08, EU:C:2011:550, paragraph 71).