The European Courts’ Jurisprudence After Altmark; Evolution or Devolution?

Abstract

This chapter examines the constitutional importance of the ruling in Altmark set in the context of the judicial application of the ruling.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

The judgment in Altmark was groundbreaking, but in many ways it can be seen in the light of its prequel, the judgment in Ferring. Many of the points that resulted in the critical reception of the latter judgment were addressed in Altmark. In this respect, the chronology coincides with an evolution and most of the cases since Altmark continue this evolutionary line.¹ However, there were also some cases that are rather more difficult to reconcile with Altmark. This chapter will review that jurisprudence since Altmark. Yet, in order to understand Altmark, we will first identify the genesis of that jurisprudence. In this regard, we will review both the pre- and post-Altmark case law from a constitutional and judicial protection perspective. This will expose the positions of the EU legislator, the Member States and Commission as well as that of national and the EU Courts in this area. Most importantly, it will also enable us to identify the role of private parties in the process of creating jurisprudence on this issue. Much like evolution, where (semi)external factors determine changes over generations leading to diversification, private parties with their widely differing backgrounds and reasons for starting actions can be seen as the natural environment that resulted in Altmark.² Evolution, however, has no purpose whereas its creationist counterpart devolution does presuppose such a motive. In this regard, reductions in complexity are often presented as evidence of devolution.³ This chapter will identify whether or not there is a purpose to Altmark and if so, what it is, particularly in the light of law’s objective of ensuring legal certainty in an increasingly complex world. Such a purpose could be the increase of efficiency in services of general economic interest or, on a more meta-legal level, the creation of an effective possibility to challenge the modalities governing such services or simply the creation of more legal certainty. Such legal certainty is all the more important in view of the importance of the services of general economic interest.

2 The Genesis of Altmark

In this section, we will study the genesis of the Altmark ruling from a constitutional and judicial protection perspective, as the substantive perspective, basically the jurisprudence from ADBHU, FFSA and SIC over Ferring, has been extensively documented.⁴ In a nutshell, following ADBHU, the compensation of costs arising from a service of general economic interest was initially not regarded as State aid by the Commission. This compensation approach was rejected by the General Court in FFSA and SIC ⁵ in what has been dubbed the State aid approach.⁶ The State aid approach entails a classification of the compensation as aid within the meaning of Article 107(1) TFEU. Such aid could then be justified on the basis of Article 106(2) TFEU. In the compensation approach, the government funding is considered outside the scope of Article 107(1), as that provision requires an advantage that does not exist if there is compensation of costs only. This line of jurisprudence was overturned by the Court in Ferring basically on the reasoning that mere compensation of costs does not confer the advantage that Article 107(1) TFEU requires for a presence of State aids.⁷ This approach was then refined in Altmark.⁸

2.1 The Constitutional Framework for a Genesis

The relatively limited number of cases concerning services of general economic interest in the first four decades of European integration on the basis of the Treaty of Rome should not detract from the obvious importance attached to these services by the Member States. Such services have always featured prominently in the Treaties with a wide-ranging justification clause in the form of Article 106(2) TFEU. However, the simple reading of that provision as one that may justify the disapplication of the entire Treaty to services of general economic interest ignores the fact that the Treaty applies to myriad forms of state and private interventions in the market. Services of general economic interest may be connected to exclusive or special rights, as Article 106(1) TFEU indicates, but they may also involve financial compensation. In relation to the former Article 106(3), TFEU clearly puts the Commission in the driving seat insofar as enforcement is concerned. Moreover, in view of the fact that most if not all exclusive rights will automatically translate into dominant positions, the Commission’s role in enforcing Article 102 TFEU means that it can effectively steer enforcement in this regard.⁹ Concerning the financing of services of general economic interest, the EU framework becomes considerably more complicated, as the Commission is the foremost enforcement body when transfers of state resources are involved insofar as such transfers amount to State aids and need to be declared compatible with the internal market on the basis of Article 107(3) TFEU. However, Article 108(2), third paragraph, TFEU allows the Council to declare State aid measures compatible with the internal market.¹⁰

The constitutional perspective on the genesis of Altmark can be seen in the opinion of Léger AG in Altmark, where he notes that the consequence of the compensation approach is to deprive Article 106(2) TFEU of a substantial part of its effect. This is problematic as it is exactly this provision that allows for a balancing of the Member States’ and EU interests between the services of general economic interest, on the one hand, and undistorted competition and the creation of an internal market on the other.¹¹ Further to the constitutional perspective, Léger notes that the central position of the Commission in State aid supervision is undermined.¹² In a nutshell, the traditional antagonists involved in the tension between undistorted competition and the internal market, on the one hand, (the Commission) and the need to have national room for manoeuvre on the other (the Member States) both face constitutional issues in defining the exact legal framework for the financing of services of general economic interest.

2.2 The Judicial Protection Framework for a Genesis

Closely connected to the constitutional issues is the judicial protection aspect. This relates essentially to the influence on access to justice of the choice for the compensation or State aid approach. In view of the fact that both the Commission and the Member States have privileged standing under Article 263 TFEU, a choice for the State aid or compensation approach should not affect their possibilities of obtaining a judicial review of a decision. This is rather more different for the private parties involved. This element of the judicial protection perspective is noted by Jacobs AG in GEMO, where he reiterates that the standstill provision that attaches to State aids means that ‘national courts must offer to individuals the certain prospect that all the appropriate conclusions will be drawn from the infringement of the last sentence of Article [108(3) TFEU]’.¹³ This judicial perspective is particularly relevant in the light of the absence of a cost-effectiveness test in Ferring and the effects of compensation on the position of competitors of the undertaking entrusted with the service of general economic interest.¹⁴ Finally, the importance of the judicial protection aspect is evidenced by the fact that most post-Altmark cases were brought by competitors of the undertakings in charge of the service of general economic interest.¹⁵

2.3 The Altmark Judgment

Altmark clearly bears the signs of its own genesis, with the Court referring to Ferring to uphold the compensation approach and the interventions, following Ferring and the opinion of Jacobs AG in GEMO, arguing in favour of the State aid or the quid pro quo approach.¹⁶

In Altmark, the Court adopted a refined compensation approach according to which a state measure can be seen as compensation for public services, meaning that the undertakings providing those services do not enjoy a real financial advantage and the measure thus does not have the effect of putting them in a more favourable competitive position than the undertakings competing with them. Such a measure would fall outside the scope of Article 107(1) TFEU if four cumulative conditions are met.

First, the recipient undertaking is required to discharge public service obligations and those obligations have been clearly defined. Second, the parameters on the basis of which the compensation is calculated must have been established beforehand in an objective and transparent manner. Third, the compensation must not exceed what is necessary to cover all or part of the costs incurred in the discharge of public service obligations, taking into account the relevant receipts and a reasonable profit for discharging those obligations. Finally, where the undertaking which is to discharge public service obligations is not chosen in a public procurement procedure, the level of compensation needed must be determined on the basis of an analysis of the costs which a typical undertaking, well-run and adequately provided with means of transport so as to be able to meet the necessary public service requirements, would have incurred in discharging those obligations, taking into account the relevant receipts and a reasonable profit for discharging the obligations.

The refinement consists of the compromise or even hybrid character involving the compensation and State aid approaches.¹⁷ This principally maintains the compensation approach in Ferring, but complements it with two additional criteria to address concerns relating to the overly broad discretion for Member States in financing services of general economic interest, thus bringing it more in line with the limited discretion that Member States enjoy under the State aid rules.¹⁸ This reduction of the Member State discretion to decide on the financing of services of general economic interest was a widely anticipated result of Altmark.¹⁹ Moreover, the reduction in Member State discretion would effectively address the concerns identified above in relation to the judicial protection perspective, provided that the national and EU judiciary would apply a stringent test in this regard.

Moreover, addressing the constitutional issues, the parallel between the first and third Altmark criteria and Article 106(2) TFEU may be identified.²⁰ The first Altmark criterion, which requires the existence of a clearly defined framework for the entrustment of services of general economic interest, is consistent with Article 106(2) case law according to which there needs to be a clear public law framework entrusting the service of general economic interest. The third criterion according to which there must be no overcompensation compares to the proportionality test applied in Article 106(2).²¹ This may be contrasted with the second and fourth Altmark conditions that introduce new standards compared to that prescribed by Article 106(2). The result of this appears to be a restriction of the Member State room for manoeuvre.²²

By and large the result of Altmark is a strict framework within which the Member States can escape State aid scrutiny only under stringent conditions that may also be relied upon by private parties both before the national and the EU judiciary. The strictness of Altmark, however, also results in problems because the increasing complexity of societies requires ever more creativity on the part of the Member States. Connected to a need to increase efficiency in societies in general because of international competitiveness, the result is a drive to come to new mechanisms that will increase efficiency and competitiveness whilst protecting the interests underlying the service of general economic interest in an ever more fine-tuned balancing act between these interests.

3 Jurisprudence After Altmark

Following Altmark, almost 20 judgments have been handed down that apply the rule laid down in that judgment. The majority of these judgments contain what can be called a simple and straightforward application of Altmark. There are also a number of judgments that point at the difficulties of applying Altmark in practice. This section will analyse these judgments from the perspective of the constitutional and judicial protection aspects identified above in order to determine the purpose of Altmark.

The first case to be mentioned in this regard is Valmont.²³ This essentially entails the appeal by a beneficiary of aid against the Commission's decision finding it to have received State aid for the construction of a car park that was incompatible with the internal market. One of Valmont’s arguments was that in fact it was only compensated for the burden that resulted from a gentlemen’s agreement with the municipality requiring it to also allow others to use the car park. The General Court read this in the light of the Altmark exception and dismissed the Commission’s approach to qualify only 50 % of the financing by the municipality as State aids.²⁴ According to the General Court the Commission should have applied the Altmark test, even though the Commission argued that the first Altmark criterion was not met in this case.²⁵ This shows that the undertakings responsible for—albeit ill-defined—services of general economic interest may rely on Altmark in order to be shielded from the EU State aid rules and avoid, for example, having to repay illegally received aids.

A further example of this can be seen in TV2/Danmark.²⁶ Here, the Commission argued that Denmark had not conducted any analysis pertaining to the fourth Altmark criterion.²⁷ This, the General Court held, would only suffice if the Commission could show that Denmark had indeed done nothing that could be construed as complying with that criterion or when these measures would have been manifestly inadequate or inappropriate for that purpose.²⁸ This, however, was not the case and again the Commission was essentially ordered to investigate whether the Altmark conditions were met. The basic message from TV2/Danmark is that the Commission can confine itself to a statement that the Altmark conditions were not met in cases where these conditions are manifestly not met,²⁹ but in all other cases a serious scrutiny of the applicability of Altmark is required.

On a similar note, the appraisals of the applicability of Altmark will affect standing for competitors under the State aid rules in the Commission procedure. In this regard, the decision to open the Phase II (or Article 108(2) TFEU) investigation turns on whether or not the compatibility with the internal market of the state measure presents serious difficulties. The message in Deutsche Post is that the appraisal of state measures in the light of Altmark will often entail a complex analysis that will not allow the Commission to come to a finding that a State aid measure presents no serious difficulties. As a result, a Phase II investigation on the basis of Article 108(2) will have to be opened.³⁰ This in turn offers competitors of the undertaking administering the service of general economic interest extra possibilities for judicial review of the Commission’s decisions.

Such judicial review will then have to be sufficiently detailed to allow for an in-depth appraisal of the applicability of all four criteria. This is where the EU Courts have shown different degrees of deference. The first case to be discussed in this regard is BUPA.³¹ This judgment resulted from the appeal by BUPA, a provider of medical insurance services, against a Commission decision declaring an Irish scheme for medical risk equalisation compatible with EU State aid law. Under the risk equalisation scheme, insurers with a better risk profile than the average market risk profile had to pay a charge to the Irish Health Insurance Authority. Corresponding payments were then made by the Health Insurance Authority to insurers with a risk profile worse than the average market risk profile. The aim of this scheme was to compensate the relatively bad risk profiles and thus level the playing field.³² Interestingly, BUPA was the main competitor of the incumbent insurance company, VHI and as a newcomer to the market, BUPA had primarily young and healthy customers, whereas VHI insured mostly older people and thus had a correspondingly worse risk profile.³³

BUPA argued that the Commission had misapplied Article 107(1) TFEU because the four Altmark conditions were not satisfied. In this regard, the General Court’s approach to the Commission’s application of the first and fourth criterion is particularly interesting.

Concerning the first criterion, BUPA argues essentially that there is a parallel between the service of general economic interest within the meaning of Article 106(2) TFEU and the public service obligation contained in the first Altmark criterion. From this, BUPA infers that the service must be universal and that its provision must be obligatory. Moreover, the obligation must be precise and limited and interpreted as a concept of EU law.³⁴ Relying on the Commission’s Communication on SGEIs, earlier case law of the General Court and Article 14 TFEU, the General Court comes to the conclusion that its review of the first criterion is limited to manifest errors of appraisal.³⁵ The full and unrestricted review asked for by BUPA was therefore not applicable. This deference on the part of the General Court is continued when the applicability of the first criterion to the Irish scheme is investigated.

Regarding the fourth criterion, BUPA argued that the absence of a comparison with an efficient operator ruled out the applicability of Altmark.³⁶ According to BUPA, the Commission did not compare VHI’s costs in administering the service to those incurred by an efficient operator. Furthermore, the Irish scheme did provide a reference point for assessing the efficiency or a benchmark for comparing decisions with those of an efficient operator.

The General Court’s answer was that the efficiency criterion could not be applied strictly in BUPA.³⁷ It based this on the neutrality of the compensation mechanism under the risk equalisation scheme by reference to the receipts and profits of the insurers and to the particular nature of the additional costs linked with a negative risk profile on the part of those insurers. The General Court noted that the payments under the Irish scheme were not determined solely by reference to the payments made by the insurer receiving compensation—which would correspond to the third and fourth Altmark criterion—but also by reference to the payments made by the contributing insurance company, which reflected the risk profile differentials of those two companies with the average market risk profile.³⁸ The level of compensation was determined by reference to the costs incurred by both the contributing and receiving company.

The General Court further held that the Commission was unable to identify the potential beneficiaries of payments under the Irish scheme and to compare these to an efficient operator because the risk equalisation scheme had not been activated when the contested decision was adopted.³⁹ At the time of the decision there was no undertaking whose efficiency could be judged against that of the benchmark.

The General Court then pointed to the purpose of the fourth Altmark criterion and held that the Commission was none the less required to satisfy itself that the compensation provided for by the Irish scheme did not entail the possibility of offsetting any costs that might result from inefficiency on the part of companies involved.⁴⁰ Here, the General Court stated that the Commission had found that the scheme allowed the insurers to keep the benefit of their own efficiencies. As the calculation of the compensation under the risk equalisation scheme depended solely on the costs not linked with the efficiency of the operators in question, that compensation was not capable of leading to the sharing of any costs resulting from their lack of efficiency.⁴¹

The judgment in BUPA can be seen as a modification of⁴² and withdrawal from the strict efficiency approach taken in Altmark.⁴³ However, it has also been argued that BUPA must be seen as evidence of the flexibility offered by the Altmark exception to services of general economic interest.⁴⁴ Buendia Sierra, however, argues that the exceptional nature of the scheme at hand in BUPA made the efficiency test less relevant. This in turn means that for non-exceptional services of general economic interest, the Altmark criteria apply in full.⁴⁵

This points to the fact that the biggest message coming from BUPA may well be that defining hard rules for services of general economic interest is well-nigh impossible in view of the diversity and complexity of services involved. Indeed, the distinction suggested by Buendia Sierra between special and normal services of general economic interest, only begs the next question: how to determine whether a specific service is normal or special? The approach by the General Court in BUPA looks at the purpose underlying the various Altmark criteria and what we can say is that this purpose functions as a teleological tool guiding the application of the Altmark test. In relation to the efficiency criterion, this shows that apart from competitive tendering and comparison to a benchmark efficient undertaking⁴⁶ there may also be other ways to ensure efficiency in the provision of services of general economic interest.

This idea of flexibility in the application of Altmark can also be found in the judgment in Chronopost.⁴⁷ The judgment in this case is the result of lengthy proceedings by UFEX et al., against the Commission decision declaring various measures undertaken by La Poste for the benefit of its daughter undertaking SFMI-Chronopost not to be State aid.⁴⁸

In this regard, we must first look at the concept of ‘normal market conditions’ used in SFEI ⁴⁹ to determine the circumstances in which the provision of logistical and commercial assistance by a public undertaking to its subsidiaries carrying on an activity open to competition constitutes State aid. UFEX et al., argued that the General Court, in referring to a private undertaking not operating in a reserved sector, had erred in basing its comparison on an undertaking that was structurally different from La Poste, instead of comparing the conduct of the latter with that of an undertaking in the same position thus with a reserved sector at its disposal.⁵⁰

The Court held that the General Court had failed to take account of the fact that an undertaking such as La Poste was in a situation very different from that of a private undertaking acting under normal market conditions.⁵¹ In this regard, the Court referred to the fact that La Poste was entrusted with a service of general economic interest, and thus had at its disposal substantial infrastructures and resources.⁵² In the absence of any possibility of comparing the situation of La Poste with that of a private group of undertakings not operating in a reserved sector, normal market conditions, which are necessarily hypothetical, had to be assessed by reference to the objective and verifiable elements which were available.⁵³ The Court stated that the costs borne by La Poste in respect of the provision to its subsidiary of logistical and commercial assistance could constitute such objective and verifiable elements.⁵⁴ There was no State aid to SFMI-Chronopost if, first, it was established that the price charged properly covered all the additional, variable costs incurred in providing the logistical and commercial assistance, an appropriate contribution to the fixed costs arising from use of the postal network and an adequate return on the capital investment in so far as it was used for SFMI-Chronopost’s competitive activity and if, second, there was nothing to suggest that those elements had been underestimated or fixed in an arbitrary fashion.⁵⁵

This effectively omits the efficiency test prescribed by the fourth Altmark criterion. Moreover, it was repeated in Chronopost II, where the Court held that the Commission should not, at first sight, be criticised for having based the contested decision on the only data available at the time, from which it was possible to reconstruct the costs incurred by La Poste.⁵⁶ The use of those data could be open to criticism only if it was established that they were based on manifestly incorrect considerations. The test laid down in Chronopost is very general in nature, prescribing the approach to be taken in order to assess whether the provision of commercial and logistical assistance involves State aid, without, however, specifying the economic, accounting or financial standards to be applied.⁵⁷ The exact definition of the variable costs to be included, as well as the ‘appropriate contribution’ and the ‘adequate return on the capital investment’ remain equally elusive. Apart from the discussion on the place of the Chronopost rulings in the grand scheme of Altmark,⁵⁸ this judgment points to the fact that cost-based standards are inherently complicated by the absence of precise cost allocation standards.

4 Financing and Costs of Services of General Economic Interest Outside the Altmark Context

The conclusion must be that an efficiency test like that prescribed in the fourth Altmark criterion may be difficult to implement in practice. Whereas a tendering procedure can be envisaged relatively easily in practice, the benchmark option appears to be a predominantly theoretical exercise. As a result, the underlying objective of ascertaining that services of general economic interest are provided at the least costs to society may also not be attained. Where, however, the public undertaking in charge of a service of general economic interest is accused of abusive practices in the meaning of Article 102 TFEU, the efficiency criterion is very much relevant again. This is because efficiencies derive from costs and costs are central to establishing many forms of abuse.

In this regard, we may point at the recent judgment in Post Danmark.⁵⁹ This is just one of a series of judgments dealing with undertakings delivering services of general economic interest in a reserved sector as well as being active in non-reserved sectors.⁶⁰ Basically, this concerned a decision by the Danish competition authority on exclusionary abuse undertaken by Post Danmark vis-à-vis its main competitor on the market for unaddressed mail, Forbruger-Kontakt.⁶¹ The decision found that Post Danmark had abused its dominant position by engaging in price discrimination with regard to former customers of Forbruger-Kontakt. Whether or not such price discrimination amounts to abuse depends on the relation of the prices to the costs.

Setting the scene for its reasoning, the Court first reiterated the ‘Michelin special responsibility’. This refers to the Court’s consistent case law that holds that dominance in itself is not a ground of criticism on the basis of Article 102 TFEU, but it does put upon that undertaking ‘a special responsibility not to allow its behaviour to impair genuine, undistorted competition on the internal market’.⁶² As far as we can see, the Court has now for the first time stated that account must be taken of the fact that ‘the existence of a dominant position has its origins in a former legal monopoly’.⁶³ The apparent meaning of this statement becomes clear only when we delve deeper into the cost allocation problems that arise. The Danish authority had used the incremental cost standard. This standard relates to those costs that would disappear in the next 3–5 years were Post Danmark to cease distributing unaddressed mail.⁶⁴ However, much like the situation in Chronopost, Post Danmark provided services in the non-reserved sector with the infrastructure and staff that it used in the reserved sector to meet the universal service obligation. This means that the costs of its universal service obligation activities would be reduced over a period of 3–5 years if Post Danmark would no longer distribute unaddressed mail. As a result, a portion of the common costs that related to both the reserved sector and commercial activities was included in the incremental costs.⁶⁵ This in turn is connected to the degree of efficiency with which a reserved activity is undertaken by the undertaking charged with the service of general economic interest. In relation to the fourth Altmark criterion, the effect is that a lack of efficiency in the reserved sector will translate into higher incremental costs because of the higher common costs. If, for example, Post Danmark would use the same postman to deliver both mails within the universal service remit and unaddressed mail, the lack of efficiency in delivering reserved mail would increase the costs that would need to be attributed to the delivery of non-addressed mail. As a rule, higher costs for a certain activity also mean that the price charged for that service needs to be higher. Given that whether or not a price is abusive depends on it exceeding, inter alia, average incremental costs, this would mean that Post Danmark would have to charge a relatively higher price if it wanted to avoid accusations of abusive conduct.

Efficiency again appears where the Court stated that only the exclusion of an ‘as efficient competitor’ would be abusive⁶⁶ and that exclusionary effects could be objectively justified on the basis of efficiencies that benefit consumers.⁶⁷ This again provides the public undertaking with an incentive to be efficient, as inefficiency on its part will make it easier for competitors to claim that they are ‘as efficient’ thus contributing to a finding of exclusionary abuse. On a similar note, it will be more difficult to argue that seemingly abusive behaviour is in fact objectively justified. The bottom line of Post Danmark is that an undertaking discharging a service of general economic interest whilst also providing commercial services is well-advised to be as efficient as possible.

5 Conclusions

Life after Altmark has not become any easier as far as the Courts’ jurisprudence is concerned. There is no clear standard and apparently complex services of general economic interest warrant a more flexible approach to the Altmark criteria, and in particular the fourth criterion designed to ensure efficiency. Probably as a result of the constitutional perspective, the Member States were left relatively free under the Altmark criteria. The option of using a benchmark as an alternative for a tendering procedure clearly follows from the constitutional framework whereby the EU leaves the Member States free in their decisions to organise markets, and thus also services of general economic interest, themselves. This sovereignty, however, impacts the judicial protection perspective as it translates into significant leeway for the Member States and a limited review by the Courts.

Nonetheless, Member States treading the fine line between markets and public intervention are well-advised to ensure the efficiency of the provision of the service of general economic interest as competitors may not only have recourse to the protection offered to them by the State aid rules, but also the antitrust rules enshrined in the Treaty. It is in this regard that we come to our main conclusion that legal certainty and judicial protection do not appear to have been the prime purposes of Altmark. However, being firmly set in the competition rules, efficiency and its close corollary consumer welfare do appear to underlie Altmark and the other competition rules applied to services of general economic interest. Such efficiency reviews may well be triggered by competitors in judicial proceedings, and thus fit in the judicial protection perspective. As to the constitutional perspective, the deference in BUPA is clearly set in the Member States’ wish to define and execute services of general economic interest in a sovereign manner. We see that much of the deference disappears where activities within the public service remit are undertaken together with commercial activities whilst entering the realm of the antitrust provisions. It is in relation to antitrust that the Treaty’s efficiency paradigm was clear from 1958 onwards. Another way of looking at this would be to state that the Courts are deferent as regards the ex ante (creation) part of a service of general economic interest. Concerning the ex post (operation) of the service of general economic interest, the Courts are stricter. We find, to answer the question in the title, evolution, but not so much in relation to the Altmark exception itself.