Category Archives: Competition law

Rules relating to Services of General Economic Interest in the postal sector : high evidential threshold for contesting actions of Member States and lack of objections raised by European Commission

Joined cases T-282/16 and T-283/16
PartiesCourtChamberJudge-RapporteurAdvocate GeneralSubject-matter
AppealInpost Paczkomaty and Inpost v European CommissionGeneral Court3rd Chamber (Extended Composition) E. Perillo/State aid – SGEI – Compensation – Universal postal service
KeywordsState aid – Postal sector – Compensation for the net cost resulting from universal service obligations – Decision declaring the aid compatible with the internal market – Action for annulment – Interest in bringing proceedings – Obligation to state reasons – Equal treatment – Proportionality – Right to property – Freedom of enterprise
SummaryThe source of this dispute is a Commission decision not to raise objections to the measure notified by Polish authorities concerning the aid granted to Poczta Polska ("PP") in the form of compensation for the net cost of fulfilling its universal postal service obligations for the period between 1 January 2013 and 31 December 2015 (Case SA.38869). The aid scheme concerned, on the one hand, a mechanism for sharing the net cost of universal service obligations and, on the other hand, a compensation fund which was partly financed by contributions from certain postal operators, a system open to Member States under the Postal Directive 97/67/EC, amended in particular by Directive 2008/6/EC (the "Postal Directive"). The Commission considered that the operating principles of the compensation fund were compatible with Article 106(2) TFEU. Two competing companies on the postal services market, which were subject to this contribution to the compensation fund, challenged the decision.

In the case at hand, the applicants' request was only considered admissible for the year in which the device was activated.

On the merits, the applicants raised seven pleas in law alleging, in substance, an infringement of Article 106(2) TFEU, in particular the Commission communication on Services of General Economic Interest (the “SGEI Framework”) and Articles 7 and 8 Postal Directive, as well as an infringement of Articles 16 and 17 Charter of Fundamental Rights and the obligation to state reasons.

By their first plea, the applicants contested the method of allocating the universal postal services to PP. The General Court (the “GC”) pointed out in this regard that neither Article 7(2) of the Postal Directive nor the SGEI Framework obliges the Member State concerned to use a public procurement procedure. This is only one option among others (para. 35). The essential condition remains that the principles of transparency, equal treatment and non-discrimination are duly respected (para. 35). In this respect, a direct and exclusive designation by legislative means is, of itself, compatible with these principles (para. 39). This is an important point for Member States contemplating an appropriate method for choosing their future universal service obligation providers. However, the GC was careful to point out in advance that the designation of PP as a universal postal service provider for a given period had already been considered during the public consultation in September 2010, which took place precisely in the context of the relevant national legislative process (para. 38). Moreover, the GC highlighted that the fact that PP's capital is wholly owned by the State has no impact on the outcome (para. 40).

The second plea concerned the transparency required by the SGEI Framework. According to the applicants, the public consultation requirements in the allocation of the universal postal service were not respected. The GC stated that the organisation of a public consultation is not compulsory if other appropriate means exist for users and service providers to make their comments known, as mentioned in point 14 of the SGEI Framework (para. 45-46). In this case, a public consultation was held (para. 47).

The central element in the Court's reasoning on this plea concerned the terms of the fund for compensating the net cost of the universal postal service obligations entrusted to PP (para. 56 et seq.). The legal problem at stake was mainly based on the contribution of the taxable companies. First, regarding the principle of non-discrimination, the GC note that the Commission could consider that Polish authorities had correctly defined taxable persons in an equivalent situation, as well as interchangeable services, and considered that express courier services did not fulfil these conditions (para. 90-97). Secondly, the applicants contested the maximum amount of the contribution and the income threshold set for being subject to it. As a preliminary remark, the Court referred to the limit of its review of proportionality in the context of SGEIs (para. 115), which is limited to verifying the necessity of the measure for the performance of the task or, conversely, its manifestly inappropriate nature. Thus, in its analysis of the proportionality of the percentage determining the maximum amount of the contribution, the GC allows some leeway. It referred in this instance to a number of economic findings made by the Commission and considered that the arguments put forward by the applicants were not sufficient to call into question the plausibility of those findings.

Thirdly, the GC discussed at length the definition of unfair financial burden (para. 147 et seq.) and the case law of the Court of Justice of the European Union (the “CJEU”) on the matter. According to the latter, "the assessment of the unfair nature of the burden associated with the provision of universal service requires a specific examination both of the net cost which provision of that service represents for each operator concerned and of all the characteristics particular to each operator, such as the quality of its equipment, its economic and financial situation and its market share" (CJEU, 6 October 2010, Commission v Belgium, C-222/08, EU:C:2010:583, para. 59 and Base and others, C-389/08, EU:C:2010:584, para. 51).

The GC concluded that, under the Postal Directive, it is for the Member States to determine what constitutes an unfair burden in the context of universal service obligations (para. 153) but that any net cost does not automatically fulfil that criterion (para. 155). In the present case, PP will only receive compensation for ex-post accounting losses resulting from the provision of universal service. This method is stricter than the SGEI Framework, which allows full compensation for the total net cost of these obligations (para. 156). Efficiency adjustments are also provided for and PP is required to submit a corrective action plan each year to eliminate or, at least, limit losses resulting from its universal service provision in order to avoid encouraging any mismanagement of the service.

The GC deals much more rapidly with the fourth, fifth and sixth pleas because of the lack of precision in the applicants' arguments.

Finally, the GC did not deal with the obligation to state reasons. It justified this approach by the fact that the applicants confused, in substance, the validity of the factual and legal elements underlying the contested decision with the insufficient statement of reasons for that act. It therefore found that the applicants’ complaints must be considered ineffective (para. 184).

At the end of its analysis, the GC dismissed the actions and confirmed the Commission's decision not to raise objections, considering that the system, although constituting State aid, is compatible with the internal market under Article 106(2) TFEU.
CommentaryFirst, this judgment is of interest in that it deals with the combined application of the horizontal SGEI Framework, which concerns the SGEIs that are potentially most restrictive of competition, particularly in view of the amount of compensation paid to their operator (more than €15 million per year), and the postal sector rules. Admittedly, its interpretation is difficult, considering the specificities of the case and what seems to constitute certain limits of the appeal lodged. However, it provides guidance on the conditions to be met by the award procedure (public consultation or other appropriate means of taking into account the interests of users and service providers, transparency, equal treatment and non-discrimination). This judgment also provides insights into the extent to which competitors of the SGEI operator can be called upon to finance the fulfilment of its public service obligations, as well as into the significant room for manoeuvre that States retain on these points.

In this respect, the judgment also reveals that competitors must, in addition to solid legal arguments, develop a well-founded and convincing economic analysis in order to have a chance of their arguments being accepted by the Commission or by the EU courts. Admittedly, the concept of unfair financial burden for a universal service operator and the criteria for determining which competitors are subject to the financing of the compensation fund require an approach that closely combines law, economics and finance. In the case at hand, it seems that the applicants were not in a position to produce arguments capable of undermining the presumption of legality enjoyed by the Commission's decision not to raise objections to the aid scheme in question.

Règles sectorielles relatives aux SIEG postaux : une solide analyse économico-juridique est nécessaire pour contester les choix de l’Etat et une décision de non objection de la Commission européenne

Affaires jointes T-282/16 et T-283/16
PartiesJuridictionFormationJuge RapporteurAvocat GénéralSujet
AppelInpost Paczkomaty et Inpost contre Commission européenneTribunal de l’Union européenne3ème Chambre élargie E. Perillo/Aides d’Etat – SIEG – Compensation Service postal universel
Mots-clésAides d’État – Secteur postal – Compensation du coût net résultant des obligations de service universel – Décision déclarant l’aide compatible avec le marché intérieur – Recours en annulation – Intérêt à agir – Obligation de motivation – Égalité de traitement – Proportionnalité – Droit de propriété – Liberté d’entreprise
RésuméÀ l’origine de ce litige se trouve une décision de la Commission de ne pas soulever d’objections à l’égard de la mesure notifiée par les autorités polonaises relative à l’aide octroyée à Poczta Polska (« PP ») sous la forme d’une compensation du coût net résultant de l’accomplissement de ses obligations de service postal universel pour la période comprise entre le 1er janvier 2013 et le 31 décembre 2015 (Affaire SA.38869). Le régime d’aide concernait, d’une part, un mécanisme de répartition du coût net des obligations de service universel, et, d’autre part, un fonds de compensation qui était financé pour partie par les contributions de certains opérateurs postaux, une formule ouverte aux États membres par la directive postale 97/67/CE, modifiée notamment par la directive 2008/6/CE (la « directive postale »). La Commission avait considéré que les principes de fonctionnement du fonds de compensation étaient compatibles avec l’article 106§2 TFUE. Deux sociétés concurrentes sur marché des services postaux, qui étaient soumises à cette contribution au fonds de compensation, ont attaqué cette décision.

En l’espèce, la demande des requérantes n’a été jugée recevable que pour l’année où le dispositif a été actionné (pt 27).

Sur le fond, les requérantes soulèvent sept moyens tirés en substance de la violation de l’article 106§2 TFUE, notamment en ce que l’encadrement SIEG et les articles 7 et 8 de la directive postale n’ont pas été respectés, ainsi que d’une violation des articles 16 et 17 de la Charte des droits fondamentaux et de l’obligation de motivation.

Aux termes de leur premier moyen, les requérantes contestent la méthode d’attribution à PP des services postaux universels. Le Tribunal rappelle que ni l’article 7§2 de la directive postale, ni l’encadrement SIEG n’obligent l’Etat membre à recourir à une procédure de passation de marché public. Il ne s’agit que d’une option parmi d’autres (pt 35). La condition essentielle demeure que « les principes de transparence, d’égalité de traitement et de non-discrimination soient dûment respectés » (pt 35). À cet égard, la désignation directe et exclusive par la voie législative est jugée ne pas violer par elle-même ces principes (pt 39). Cette mention est importante à relever pour les Etats membres qui s’interrogent sur la méthode appropriée pour choisir leurs futurs prestataires d’obligations de service universel. Toutefois, le Tribunal avait pris le soin de souligner au préalable que la désignation de PP en tant que prestataire des services postaux universels pendant une période déterminée avait déjà été envisagée lors de la consultation publique de septembre 2010, intervenue précisément dans le cadre du processus législatif national pertinent (point 38). Par ailleurs, le fait que le capital de PP soit intégralement détenu par l’Etat n’a pas d’incidence sur la solution (pt 40).

Le deuxième moyen porte sur la transparence exigée par l’encadrement SIEG. Selon les requérants, les exigences de consultation publique dans l’attribution du service postal universel n’ont pas été respectées. Le Tribunal leur oppose que l’organisation d’une consultation publique n’est pas obligatoire si d’autres moyens appropriés existent pour que les utilisateurs et prestataires de services puissent faire valoir utilement leurs observations, ainsi que cela ressort du point 14 de l’encadrement SIEG (pts 45-46). En tout état de cause, une consultation publique a eu lieu en l’espèce (point 47).

L’élément central du raisonnement du Tribunal porte sur les modalités du fonds de compensation du coût net des obligations de service postal universel confiées à PP (pts 56 et s.). Le problème juridique repose principalement sur la contribution des entreprises assujetties. Premièrement, sur le principe de non-discrimination, les juges de l’UE constatent que la Commission a pu considérer que les autorités polonaises avaient correctement défini les assujettis se trouvant dans une situation équivalente, ainsi que les services interchangeables, et estimé que les services de courrier express ne remplissent pas ces conditions (pts 90-97). Deuxièmement, les requérants contestaient le montant maximal de la contribution et le seuil de revenus fixé pour y être assujetti. A titre liminaire, le Tribunal évoque la limite de son contrôle de proportionnalité dans le contexte des SIEG (pt 115) qui se limite à vérifier la nécessité de la mesure pour l’accomplissement de la mission ou inversement son caractère manifestement inapproprié. Ainsi, dans son analyse de la proportionnalité du pourcentage déterminant le montant maximal de la contribution, le Tribunal conserve une certaine réserve. Il relève un certain nombre de constatations économiques effectuées par la Commission dont il considère que les éléments avancés par les requérants n’ont pas suffi à remettre en cause la plausibilité, après un examen circonstancié de ceux-ci. Troisièmement, le Tribunal revient longuement sur la définition à donner à la charge financière inéquitable (pts 147 et s.) et à la jurisprudence de la CJUE en la matière. Cette dernière exige de chaque Etat membre que « l’appréciation de ce caractère excessif de la charge liée à la fourniture du service universel suppose un examen particulier à la fois du coût net que représente cette fourniture pour chaque opérateur concerné et de l’ensemble des caractéristiques propres à cet opérateur, telles que le niveau de ses équipements, sa situation économique et financière ainsi que sa part de marché » (CJUE, 6 octobre 2010, Commission contre Belgique, C-222/08, EU:C:2010:583, pt 59 et Base e.a., C-389/08, EU:C:2010:584, pt 51).

Il en conclut qu’aux termes de la directive postale, il revient aux Etats membres de déterminer ce qui constitue une charge inéquitable dans le contexte des obligations de service universel (pt 153) mais que tout coût net ne remplit pas automatiquement ce critère (pt 155). En l’espèce, PP ne recevra une compensation que pour les pertes comptables ex post, résultant des prestations de service universel. Cette méthode est plus stricte que l’encadrement SIEG, qui permet une compensation totale pour l’ensemble du coût net de ces obligations (pt 156). Des ajustements d’efficacité sont également prévus et PP est tenue de présenter chaque année un plan d’actions correctives visant à éliminer ou, du moins, à limiter les pertes résultant de ses prestations de service universel afin d’éviter d’encourager toute mauvaise gestion de celui-ci.

Le Tribunal traite beaucoup plus rapidement les quatrième, cinquième et sixième moyens à cause de leur manque de précision dans l’argumentation des requérants.

Enfin, de manière originale, le Tribunal finit par la méconnaissance de l’obligation de motivation. Il justifie cette approche par le fait les « requérants confondent, en substance, le bien-fondé des éléments de fait et de droit se trouvant à la base de la décision attaquée avec le défaut ou l’insuffisance de motivation de cet acte, de sorte que de tels griefs doivent être considérés comme étant inopérants » (pt 184).

Au terme de son analyse, le Tribunal rejette les recours et confirme l’analyse de la Commission dans sa décision de ne pas soulever d’objection, considérant que le système bien qu’il constitue une aide d’Etat, est compatible sur la base de l’article 106§2 TFUE.
CommentaireL’arrêt présente en premier lieu l’intérêt de porter sur l’application combinée de l’encadrement SIEG de portée horizontale, qui concerne les SIEG potentiellement les plus attentatoires à la concurrence, compte tenu notamment du montant de la compensation versée à leur exploitant (plus de 15 millions d’euros par an) avec les règles sectorielles dans le domaine postal. Certes, son interprétation est délicate, compte tenu des spécificités de l’espèce et de ce qui semble constituer certaines limites du recours introduit. Toutefois, il fournit des enseignements sur les conditions auxquelles doit répondre la procédure d’attribution (consultation publique ou autres moyens appropriés de tenir compte des intérêts des utilisateurs et des prestataires de services, transparence, égalité de traitement et non discrimination) et sur la mesure dans laquelle des concurrents de l’exploitant du SIEG peuvent être mis à contribution pour financer l’accomplissement de ses obligations de service public, de même que sur les marges de manœuvre non négligeables que les Etats conservent sur ces points.

À cet égard, l’arrêt révèle en second lieu que les concurrents doivent, en plus d’une argumentation juridique solide, développer une analyse économique très étayée et convaincante pour avoir une chance de se voir donner raison par la Commission européenne ou par le Tribunal de l’Union européenne. Il est vrai que les concepts de charge financière inéquitable pour l’exploitant du service universel et les critères de détermination des concurrents assujettis au financement du fonds de compensation en fonction des activités poursuivies de même que du niveau des contributions nécessitent une approche combinant étroitement droit, économie et finance. Il semble que les requérants n’aient pas été en mesure de produire un tel argumentaire de nature à ébranler la présomption de légalité dont bénéficiait la décision de la Commission de ne pas soulever d’objection à l’égard du régime d’aides en cause.

Une remise en question par le Tribunal de l’Union européenne de la méthodologie de la Commission européenne et, implicitement, de son impartialité et du respect du principe de bonne administration, dans les affaires d’aides d’Etat de nature fiscale à l’occasion du dossier concernant le régime belge relatif aux bénéfices excédentaires

Affaires T-131/16 et T-263/16
PartiesJuridictionFormationJuge RapporteurAvocat GénéralSujet
AppelRoyaume de Belgique et Magnetrol International contre Commission européenneTribunal de l’Union européenne7ème Chambre élargie V. Tomljenović/Aides d’Etat – Décision fiscale anticipée – Régime d’aides
Mots-clésAides d’État – Régime d’aide mise en exécution par la Belgique – Décision déclarant le régime d’aides incompatible avec le marché intérieur et illégal et ordonnant la récupération de l’aide versée – Décision fiscale anticipée (tax ruling) – Exonération des bénéfices excédentaires – Autonomie fiscale des États membres – Notion de régime d’aides – Mesures d’application supplémentaires
RésuméSi l’on excepte le premier moyen tiré de l’atteinte à la compétence exclusive des Etats membres en matière de fiscalité directe, qui s’opposerait à un contrôle de la Commission au titre des aides d’Etat, qui a été balayé car outrancier, l’objet immédiat de l’arrêt rendu par le Tribunal de l’Union européenne (le « TUE ») le 14 février 2019 dans les affaires T-131/16 et T-263/16, opposant respectivement la Belgique et Magnetrol International à la Commission européenne était technique puisqu’il portait sur la notion de régime d’aides par opposition à celle d’aide individuelle. Toutefois, l’arrêt rendu revêt de l’importance dans l’offensive lancée par la Commission européenne contre ce qu’elle considère comme des cadeaux fiscaux de certains Etats membres aux groupes multinationaux. Au travers d’une motivation précise et fouillée, il met, en effet, en évidence certains manquements méthodologiques des services de la Commission dans ses investigations au titre des règles relatives aux aides d’Etat dans le domaine fiscal.

La Commission avait-elle pu à bon droit identifier un régime d’aides concernant l’exonération des bénéfices excédentaires en droit belge ? Les enjeux pratiques étaient substantiels. L’existence d’un régime d’aides dispensait la Commission européenne d’examiner toutes les mesures individuelles octroyées à des entreprises sur la base du supposé régime. Elle pouvait prendre une seule décision sur le régime, interdisant son maintien. C’était donc toute une législation qui cessait de s’appliquer. L’atteinte à la politique fiscale de la Belgique était bien plus substantielle et rapide, au prix d’un investissement en travail nettement moindre de la Commission européenne. Pour la même raison, une telle qualification était de nature à permettre à la Commission européenne, dès l’ouverture de la procédure formelle d’investigation, d’ordonner la suspension de l’application de la législation concernée dans son ensemble. Si, en l’espèce, elle ne l’avait pas requis expressément, la Belgique, consciente des risques encourus, avait opté pour une telle suspension.

Pour retenir la présence d’un régime d’aides et non d’un faisceau d’aides individuelles disparates, la Commission européenne n’avait pas pu identifier un acte juridique instituant un tel régime d’aides. Comme la jurisprudence l’y autorise dans un tel cas, elle avait cherché à se fonder sur un ensemble de circonstances de nature à déceler l’existence en fait d’un tel régime. A cet effet, elle avait retenu pas moins de quatre éléments juridiques, de nature différente et s’échelonnant dans le temps : une disposition légale (l’article 185, paragraphe 2, sous b), un extrait de ses travaux préparatoires (l’exposé des motifs de la loi du 21 juin 2004), une circulaire administrative (du 4 juillet 2006) et, enfin, les réponses du ministre des Finances aux questions parlementaires sur l’application de ladite disposition légale. Selon elle, ceux-ci constituaient les actes sur la base desquels l’exonération des bénéfices excédentaires est accordée.

Le TUE a toutefois mis en lumière que plusieurs des éléments essentiels du prétendu régime d’aides, dégagés par la Commission européenne, ne découlaient pas des bases du régime retenues par la Commission mais provenaient de l’examen d’un échantillon des mesures individuelles. Autrement dit, elle avait bâti, à partir de certaines mesures individuelles, un prétendu régime général qu’elle avait cherché à rattacher à des fragments juridiques de portée générale du droit fiscal belge. Bref, elle avait construit un dossier à charge de l’Etat belge. Les éléments essentiels en question prêtés au régime postulé étaient la méthode de calcul en deux étapes des bénéfices excédentaires et certaines formes d’intensification de la présence en Belgique.

Dans un ordre d’idées proche, le TUE a également relevé que la catégorie des bénéficiaires identifiée par la Commission ne correspondait pas à celle figurant dans la disposition légale retenue par elle comme l’une des bases du régime. Ceci constituait une nouvelle distorsion du cadre juridique belge par la Commission et confirmait que le rattachement du régime qu’elle prétendait avoir identifié aux bases qu’elle avait retenues était forcé.

Le TUE a également relevé que l’un des éléments présentés comme essentiels par la Commission, la méthode de calcul en deux étapes des bénéfices excédentaires, n’avait pas été systématiquement adoptée.

Par ailleurs, le TUE a souligné que l’administration fiscale belge disposait d’une marge d’appréciation substantielle pour déterminer s’il y avait lieu à ajustement des bénéfices, ce qui contredisait la thèse d’un régime général donnant lieu à de simples mesures individuelles d’application.

D’une part, l’approche était au cas par cas. L’ajustement ne nécessitait pas l’attribution des bénéfices concernés à une autre société. Contrairement au prescrit de la disposition légale, le montant à exonérer et les bénéfices excédentaires ne faisaient pas l’objet d’une définition dans les actes de base. Seuls 50 % des dossiers soumis à l’administration donnaient lieu à une décision anticipée.

D’autre part, il n’y avait pas une ligne systématique de conduite de l’administration fiscale, que la Commission, dans une attitude de repli devant le TUE, avait cherché à présenter, à titre subsidiaire, comme la base du prétendu régime général. Non seulement le Tribunal a fort logiquement écarté cette prétention, en indiquant qu’il ne pouvait pas accepter une motivation postérieure à l’adoption de la décision de la Commission, mais il s’est employé à démonter les faiblesses méthodologiques de la Commission dans l’invocation de cette prétendue ligne systématique. C’est ainsi que celle-ci avait cherché à asseoir l’existence d’une telle constance sur un échantillon couvrant un tiers des décisions individuelles de l’administration fiscale belge sans préciser le choix de cet échantillon ni les raisons pour lesquelles il avait été considéré comme représentatif de l’ensemble des décisions individuelles. Pour un autre point, elle s’était contentée de se référer à un onzième des décisions sans aucune précision sur le caractère suffisamment représentatif de l’échantillon.
A retenirLe Tribunal a mis en lumière de substantielles erreurs de la Commission européenne dans l’analyse des éléments de droit fiscal belge concernés comme constituant un régime général d’aides. Celles-ci confinent à l’arbitraire. Leur multiplication donne l’impression que, pour certains services de la Commission, la fin (lutte contre l’optimalisation fiscale et la concurrence fiscale dommageable, recherche d’une imposition plus substantielle des multinationales, harmonisation fiscale au sein de l’Union européenne) justifie les moyens (la dénaturation des faits, distorsion de la réalité). Au terme d’une démonstration magistrale, le Tribunal rappelle que l’instrumentalisation politique a des limites dans une Union de droit. Que la Commission développe des thèses juridiques nouvelles et ambitieuses est une chose, qu’elle prenne des distances avec la réalité et plie les faits à ses désirs en est une autre.

Ce manque de rigueur, cette dérive de la Commission amènent à s’interroger sur la fiabilité du traitement qu’elle a réservé à d’autres affaires d’aides d’Etat de nature fiscale, encore en cours. Les prochains arrêts du Tribunal en la matière sont attendus avec impatience.

Par ailleurs, le souci du Tribunal et le contrôle par celui-ci d’une analyse rigoureuse du droit fiscal national en cause par la Commission européenne procède peut-être d’une volonté de trouver un point d’équilibre entre, d’une part, la compétence exclusive des Etats membres en matière de fiscalité directe et, d’autre part, le fait que le droit des aides d’Etat a néanmoins vocation à s’appliquer à cette matière. Le respect par la Commission de la compétence des Etats membres à continuer à développer une politique fiscale vis-à-vis des entreprises requiert, à tout le moins, qu’elle traite le droit fiscal d’un Etat membre comme il l’est, sans a priori défavorable.

The methodology of the EU Commission and implicitly its impartiality and its good admnistration questioned by the General Court in the tax state aid investigations in the Belgian excess profit regime case

Cases T-131/16 and T-263/16
PartiesCourtChamberJudge-RapporteurAdvocate GeneralSubject-matter
AppealKingdom of Belgium and Magnetrol International v European CommissionGeneral Court7th Chamber (Extended Composition) V. Tomljenović/State aid – Aid scheme – Tax ruling
KeywordsState aid – Aid scheme implemented by Belgium – Decision declaring the aid scheme incompatible with the internal market and unlawful and ordering recovery of the aid granted – Tax ruling – Excess profit exemption – Fiscal autonomy of the Member States – Concept of an aid scheme – Further implementing measures
Significant pointsApart from the first plea alleging infringement of the exclusive competence of Member States in the field of direct taxation, which would preclude the Commission’s control of State aid, and which was dismissed by the General Court (“GC”), the object of the judgment delivered was technical. It concerned the concept of an aid scheme as opposed to that of an individual aid measure. The judgment is important with regard to the Commission's offensive against what it considers to be tax gifts granted by certain Member States to multinational groups. By means of a precise and detailed statement of reasons, it highlights certain methodological shortcomings by the Commission in its investigations in this area.

Could the Commission have correctly identified an aid scheme concerning the exemption of excess profits under Belgian law? The practical stakes were substantial because the existence of an aid scheme exempted the European Commission from examining all individual measures granted to companies on the basis of the alleged scheme. The Commission could instead take a single decision on the regime as a whole, prohibiting its continuation. Such a classification enabled the Commission to order the suspension of the legislation concerned as a whole as soon as the formal investigation procedure was opened. While the Commission had not expressly requested it in the case at hand, Belgium opted for such a suspension considering the risks at stake.

In determining the existence of an aid scheme and not of a range of individual aid measures, the Commission had been unable to identify a legal act establishing such a scheme. On the basis of established case law allowing it to do so, the Commission had sought to rely on a set of circumstances likely to demonstrate the existence of the scheme. To this end, the Commission had identified four legal elements: a legal provision (Article 185(2)(b)); an extract from its preparatory work (the explanatory memorandum of the law of 21 June 2004); an administrative circular (of 4 July 2006); and, finally, the responses of the Minister of Finance to parliamentary questions on the application of the abovementioned legal provision. In the Commission’s view, these were the acts on the basis of which the excess profits exemption was granted.

However, the GC pointed out, first, that several essential elements of the alleged aid scheme did not actually stem from the bases of the scheme identified by the Commission but were derived from the examination of a sample of individual measures. In other words, the Commission had built, on the basis of certain individual measures, an alleged general scheme which it sought to link to elements of Belgian tax law. The essential elements of the alleged scheme in question that were not present in the bases of the scheme were the two-step method of calculating excess profits and certain forms of intensification of the companies’ presence in Belgium].

In a similar vein, the GC noted that the category of beneficiaries identified by the Commission did not correspond to that contained in the legal provision considered by the Commission as one of the bases of the alleged scheme. This constituted a further distortion of the Belgian legal framework by the Commission and confirmed that the connection established between the alleged aid scheme and the elements identified as the bases of that scheme was erroneous.

The GC also considered that one of the elements presented as essential by the Commission, the two-step method of calculating surplus profits, had not been systematically adopted in the decision.

In addition, the GC pointed out that the Belgian tax administration enjoyed a substantial margin of discretion in determining whether or not a profit adjustment was necessary, which contradicted the hypothesis of a general scheme giving rise to simple individual implementing measures.

On the one hand, the approach by the Belgian taw authorities was done on a case-by-case basis. The profit adjustment did not require the allocation of the profits concerned to another company. Contrary to the requirement of the legal provision, the amount to be exempted and the excess profits were not defined in the basic acts. Only 50% of the files submitted to the administration gave rise to an advance ruling.

On the other hand, the tax administration’s systematic approach presented in the alternative by the Commission as the basis of the alleged general scheme was inexistent. Not only did the GC logically dismiss this claim, indicating that it could not accept a justification provided only subsequent to the adoption of the Commission's decision, but it also sought to demonstrate the Commission's methodological weaknesses in invoking this alleged systematic approach. In particular, the Commission had sought to establish the existence of such a consistent treatment on a sample covering one third of the Belgian tax administration's individual decisions without justifying the choice of this sample or specifying the reasons why it had been considered representative of all individual decisions. On another point, it had simply referred to one eleventh of the decisions without any clarification on the sufficiently representative nature of the sample.
NoteworthyThe GC found substantial errors on the part of the Commission in its analysis of the elements of Belgian tax law considered as constituting a general aid scheme. The assessment was regarded as arbitrary by the GC. Their multiplication gives the impression that, for the Commission, the ends (i.e. the fight against tax optimisation and harmful tax competition, increased taxation of multinationals, tax harmonisation within the European Union) justified the means (i.e. distortion of facts, distortion of reality). At the end of a masterful demonstration, the GC recalls that political instrumentalisation has limits in a Union governed by the rule of law. It is one thing for the Commission to develop new and ambitious legal arguments, but it is another to distance itself from reality and manipulate the facts in order to serve specific purposes.

The Commission's drift and lack of rigour in the decision raise questions as to the reliability of the way in which it has handled other tax-related state aid cases, which are still ongoing. The GC's forthcoming judgments in this area are eagerly awaited, therefore.

Moreover, the GC's concern and its scrutiny of the Commission’s analysis of the national tax law at stake may stem from a desire to strike a balance between, on the one hand, the exclusive competence of the Member States in the field of direct taxation and, on the other hand, the fact that State aid law is nevertheless intended to apply in this area. The Commission's respect for the competence of the Member States to continue to develop tax policies towards companies requires, at the very least, that it treats the tax law of a Member State as it is, without pre-conceived bias.

Le périmètre des services d’intérêt économique général doit être défini clairement et précisément

Affaires jointes T-202/10 RENV II and T-203/10 RENV II
PartiesJuridictionFormationJuge RapporteurAvocat GénéralSujet
AppelStichting Woonlinie e.a. contre Commission européenneTribunal de l’Union européenneHuitième Chambre élargieG. De Baere /Aides d’Etat – SIEG
Mots-clésAides d’État – Logement social – Régime d’aides en faveur des sociétés de logement social – Aides existantes – Engagements de l’État membre – Décision déclarant l’aide compatible avec le marché intérieur – Service d’intérêt économique général (SIEG) – Définition de la mission de service public
RésuméLe 15 novembre 2018, le Tribunal de l’Union européenne (le "Tribunal") a rejeté au fond un recours contre une décision de la Commission européenne relative à un régime d'aides au profit de sociétés de logement social néerlandaises.

Par une décision de 2009, partiellement modifiée en août 2010, la Commission avait accepté les propositions de mesures utiles présentées par les Pays-Bas concernant un régime de financement de logements sociaux considéré comme un régime d'aides existant. Le constat essentiel de la Commission était que la mission de service public pour laquelle une compensation avait été accordée aux sociétés de logement social n'était pas définie avec suffisamment de précision.

Stichting Woonlinie et d'autres sociétés de logement social à but non lucratif néerlandaises ont demandé l'annulation de la décision de la Commission devant le Tribunal. Ce dernier a déclaré leurs recours irrecevables (affaires T-202/10 et 203/10). Sur pourvoi, la Cour de justice de l’Union européenne (CJUE) a jugé les recours recevables et renvoyé les affaires devant le Tribunal pour qu'il statue sur le fond (affaires C-132/12 P et C-133/12 P). Le Tribunal a rejeté les recours comme étant manifestement non fondés (affaires T-202/10 RENV et T-203/10 RENV). Toutefois, ses ordonnances ont été annulées par la CJUE par arrêts du 15 mars 2017 (C-414/15 P et C-415/15 P). Celle-ci a, à nouveau, renvoyé les affaires devant le Tribunal, donnant lieu à l'arrêt en cause.

Dans son arrêt du 15 novembre 2018, le Tribunal a rejeté les recours au motif principal que la définition du service d'intérêt économique général ("SIEG") de logement social en cause était insuffisamment précise (deuxième au sixième et huitième moyens). Il a également écarté les moyens relatifs aux notions de régime d’aides et d’aides existantes (premier et septième moyens).

Le Tribunal a rappelé qu’une compensation publique pour un SIEG échappe à la qualification d’aide d’Etat moyennant le respect des quatre conditions développées par la CJUE dans son célèbre arrêt Altmark. La première de ces conditions est que l'entreprise bénéficiaire soit chargée de l’exécution d’obligations de service public clairement définies. Celle-ci doit également être remplie dans les cas où la dérogation prévue à l'article 106, paragraphe 2, du TFUE a vocation à être appliquée, c'est-à-dire lorsque certaines des autres conditions de la jurisprudence Altmark ne sont pas remplies.

Le Tribunal a également rappelé que les Etats membres jouissent d'un large pouvoir d’appréciation dans la définition de ce qu'ils considèrent comme un SIEG. Par conséquent, cette définition ne peut être remise en question par la Commission qu’en cas d’erreur manifeste. A cet égard, les Etats membres doivent démontrer que le périmètre du SIEG est nécessaire et proportionné par rapport à un besoin réel et actuel de service public. L'absence de la preuve que ces critères sont satisfaits peut constituer une erreur manifeste d'appréciation.

En l'espèce, le Tribunal a considéré que la définition du SIEG en cause était entachée d’une erreur manifeste car elle prévoyait de donner la priorité aux personnes "qui avaient des difficultés à trouver un logement adéquat", sans définir précisément le groupe cible de personnes défavorisées (deuxième moyen).

Le Tribunal a estimé que, contrairement à ce qu’alléguaient les requérants, la Commission n'avait pas exigé une définition du SIEG fondée sur une limite de revenus et qu'elle était tout à fait en droit d'exiger une définition plus précise du groupe cible que celle retenue par le droit néerlandais (quatrième et sixième moyens).

Le Tribunal a également estimé que la Commission n'avait pas commis d'erreur dans son appréciation du périmètre du SIEG. Au contraire, cette dernière avait correctement constaté que l'absence d'une délimitation précise du SIEG comportait le risque que les compensations octroyées aux sociétés de logement bénéficient également à leurs activités accessoires (rentables), qui ne seraient donc pas exercées aux conditions du marché (troisième moyen).

Le Tribunal a par ailleurs jugé que, si la Commission ne peut faire dépendre la définition du SIEG de son mode de financement, une définition claire du SIEG est néanmoins nécessaire pour garantir le respect de la condition de proportionnalité de la compensation à la mission de service public et pour éviter que les activités exercées par les sociétés de logement en dehors du SIEG ne bénéficient d’aides d'État ( risque de subventions croisées) (cinquième moyen).

En outre, le Tribunal a estimé que, contrairement aux allégations des requérants, la Commission n'avait pas imposé une liste exhaustive de bâtiments pouvant être qualifiés de "sociaux". La liste avait en réalité été établie par les autorités néerlandaises afin de répondre à la crainte, exprimée par la Commission au cours de la procédure, que l'aide accordée pour le financement du SIEG ne bénéficie à des bâtiments dans lesquels sont exercées des activités commerciales (huitième moyen).

S’agissant du premier moyen selon lequel certaines mesures auraient été considérées à tort comme faisant partie d'un régime d'aides par la Commission, alors qu'il s'agissait d'aides individuelles non prévues par un texte législatif, le Tribunal a opposé que le fait que des aides individuelles soient accordées n'exclut pas l'existence d'un régime sur le fondement duquel ces aides sont octroyées. En outre, le règlement 659/1999 ne requiert pas qu’un régime d’aides soit fondé sur une disposition législative.

Les requérantes avançaient également que la Commission avait commis une erreur en n'examinant pas l'existence d'une surcompensation dans le système initial de financement du logement social. Cependant, le Tribunal a opposé qu’un tel examen n’est pas, en soi, indispensable pour évaluer correctement, à propos d’un régime d’aides existant, des mesures utiles pour l’avenir (septième moyen).

Par conséquent, le Tribunal a rejeté les recours interjetés par les requérants.
A retenirCet arrêt est le dernier épisode en date du dossier des organismes de logement social à but non lucratif néerlandais.

Il en ressort que si la Commission n'est pas compétente pour définir un SIEG, cette tâche revenant à l'Etat membre concerné, les SIEG doivent toutefois être clairement et précisément définis.

A cet égard, il appartient à l'Etat membre de démontrer que le périmètre du SIEG est nécessaire et proportionné par rapport à un besoin de service public qui doit être réel et actuel. Une définition précise garantit l’absence de surcompensation et permet d’assurer que les activités accessoires des entreprises responsables du SIEG ne bénéficient pas d'aides d'Etat. En outre, les autorités nationales doivent fournir des informations sur tous les coûts supplémentaires encourus par les entreprises chargées des missions de service public.

Il n'appartient pas à la Commission de donner des indications ou de proposer des éléments de définition concernant le mode de détermination des bénéficiaires du SIEG. Le Tribunal a également jugé qu'il n'était pas nécessaire d'indiquer une limite de revenus. En sens inverse, réduire la valeur maximale des logements pouvant être considérés comme des logements sociaux ne permet pas d’identifier les personnes auxquelles les SIEG sont ouverts.

Enfin, une activité accessoire rentable est autorisée aux opérateurs en charge des SIEG. Dans le cas présent, par exemple, en cas de surcapacité des logements sociaux, l'opérateur pourrait louer un logement à des personnes qui n'entrent pas dans la catégorie des personnes cibles. Toutefois, l'activité accessoire ne sera pas considérée comme un service public et ne pourra pas être incluse dans le périmètre du SIEG. En outre, elle doit faire l'objet d'une comptabilité séparée afin de refléter l'absence de subventions croisées. Enfin, elle doit être menée aux conditions du marché.

The GC underlines the importance of clearly and precisely defining the scope of Services of General Economic Interest

Joined cases T-202/10 RENV II and T-203/10 RENV II
PartiesCourtChamberJudge-RapporteurAdvocate GeneralSubject-matter
AppealStichting Woonlinie and Others
European Commission
General Court8th Chamber, Extended
G. de Baere/State aid - SGEI
KeywordsState aid — Social housing – Aid schemes in favour of social housing corporations – Existing aids – Member State commitments — Decision declaring the aid compatible with the internal market — Article 17 of Regulation (EC) No 659/1999 — Service of general economic interest — Article 106(2) TFEU — Definition of the public service mission
Significant pointsOn 15 November 2018, the General Court (the “GC”) dismissed an appeal against a Commission decision on a Dutch aid scheme in favour of social housing corporations.

By decision in 2009, partially modified in August 2010, the Commission had accepted proposals for appropriate measures submitted by the Netherlands concerning a social housing financing scheme considered as an existing aid scheme by the Commission. The Commission’s main finding was that the public service mission for which compensation was granted was not defined precisely enough.

Stichting Woonlinie and other Dutch non-profit social housing corporations sought the annulment of the Commission’s decision before the GC. The latter declared their actions inadmissible (cases T-202/10 and 203/10). On appeal, the Court of Justice of the European Union (the “CJEU”) found the actions admissible and referred the cases back to the GC for judgment on the merits (cases C-132/12 P and C-133/12 P). The GC dismissed the actions as manifestly unfounded (cases T-202/10 RENV and T-203/10 RENV). However, its judgments were annulled by the CJEU by judgments of 15 March 2017 (C-414/15 P and C-415/15 P). The Court again referred the case back to the GC, giving rise to the judgment at hand.

In the judgment on 15 November 2018, the GC dismissed the actions on the main ground that the definition of the service of general economic interest (“SGEI”) for social housing at stake, which was provided for in Dutch legislation, was insufficiently precise (2nd to 6th and 8th pleas). It also dismissed the pleas relating to the concepts of aid scheme and existing aid (1st and 7th pleas).

The GC recalled that public compensation for SGEIs does not constitute State aid if four conditions are fulfilled as per the Altmark judgment. The first of these conditions is that the recipient undertaking must have public service obligations which are clearly defined. This condition must also be fulfilled in cases where the derogation provided for in Article 106(2) TFEU is to be applied, i.e. when some of the other Altmark conditions are not met.

The GC also recalled that Member States enjoy broad discretion as to the definition of what they consider to be an SGEI. Therefore, this definition can only be called into question by the Commission in the event of a manifest error. In this regard, Member States must demonstrate that the scope of the SGEI is necessary and proportionate in relation to the public service need, which must be genuine. The absence of evidence that these criteria are met may constitute a manifest error of assessment.

In the case at hand, the GC considered that the definition of the SGEI was vitiated by a manifest error because it provided that housing for rental be prioritised for people “who had difficulty in finding suitable housing", without precisely defining the target group of disadvantaged people (2nd plea).

The GC found that the Commission did not require a definition of the SGEI based on an income ceiling as alleged by the applicants and was entitled to demand a more precise definition of the target group than that provided for by Dutch legislation (4th and 6th pleas).

The Court also found that the Commission had not erred in its assessment of the scope of the SGEI. On the contrary, it had correctly noted that the absence of a precise delimitation of the SGEI entailed the risk that the compensation granted to housing companies would also benefit their ancillary (profitable) activities, which would therefore not be exercised under market conditions (3rd plea).

The Court also held that, although the Commission cannot make the definition of the SGEI dependent on its method of financing, a clear definition of the SGEI is nevertheless necessary to ensure compliance with the condition of proportionality of the compensation to the public service mission and to avoid that the activities carried out by housing companies outside the SGEI do not benefit from State aid (risk of cross-subsidies) (5th plea).

In addition, the GC found that, contrary to the applicants’ allegations, the Commission had not imposed an exhaustive list of buildings that could be qualified as "social”. The list was established by the Dutch authorities in order to address the concern, expressed during the procedure, that the aid granted for the financing of the SGEI could benefit buildings in which commercial activities were carried out (8th plea).

As regards the first plea according to which certain measures were wrongly considered by the Commission to form part of an aid scheme even though they were individual aid measures not provided for in a legislative text, the GC objected that the fact that individual aid was granted did not exclude the existence of a scheme on the basis of which such aid was granted. Moreover, Regulation 659/1999 does not require that an aid scheme be based on a legislative provision.

The applicants also argued that the Commission erred in failing to examine the existence of overcompensation in the original system of financing social housing. However, the GC objected that such an examination is not, in itself, necessary to properly assess appropriate measures for the future in respect of an existing aid scheme (7th plea).

As a result, the GC dismissed the appeals brought by the applicants.
NoteworthyThis judgment is the latest episode in the case of Dutch non-profit social housing organisations.

It follows that, although the Commission is not competent to define a SGEI since this task falls to the Member State concerned, SGEIs must nevertheless be clearly and precisely defined.

In this regard, it is for the Member State to demonstrate that the scope of the SGEI is necessary and proportionate in relation to the public service need, which has to be genuine. A precise definition ensures that there is no overcompensation and that the ancillary activities carried out by the companies in charge of the SGEI do not benefit from State aid. Moreover, national authorities must provide information on any additional costs incurred by the companies in charge of the public service missions.

It is not for the Commission to give indications or propose elements of definition concerning the method of determining the beneficiaries of the SGEI. The GC also judged that there is no requirement to indicate an income ceiling. Conversely, reducing the maximum value of housing that can be considered as social housing does not make it possible to identify the persons to whom the SGEI is opened.

Finally, an ancillary profitable activity is permitted for operators in charge of SGEIs. In this case, for example, in the event of overcapacity in housing, the operator could rent housing to people who did not fall within the category of target persons. However, the ancillary activity will not be regarded as a public service and may not be included in the scope of the SGEI. In addition, it must be subject to separate accounting to reflect the absence of cross-subsidy. Finally, the activity must be carried out under market conditions.

Services of general economic interest are only valid where the public service obligation is clearly and precisely defined

C-91/17P & C-92/17P
PartiesJurisdictionFormationJudge RapporteurAdvocate GeneralSubject-matter
Preliminary rulingCellnex Telecom SA and Telecom Castilla-La Mancha, SA v European CommissionCourt of Justice ninth ChamberK. JürimäeE. SharpstonState aid – Service of general economic interest
KeywordsAppeal - State aid - Digital Television - Deployment of terrestrial digital television in areas remote and less urbanized of the Comunidad autonomous of Castilla - La Mancha (autonomous community of Castile - La Mancha, Spain) - Grant for the digital terrestrial television platform operators - Decision declaring partially measures of aid incompatible with the internal market - Service of general economic interest (“SGEI”) - Definition
Significant pointsBy judgment on April 26, 2018, the Court of Justice put an end to proceedings concerning a measure of support for the development of digital television in a region of Spain with a low level of urbanisation. As a result of this judgment, the development in question cannot be considered a SGEI due to the lack of a clear definition of the service provided.

In 2016, the Commission considered that this measure amounted to State aid and that it was incompatible with the internal market. In its reasoning, the European Commission considered that, in the absence of both a sufficiently precise definition of terrestrial platform exploitation as a public service and an act awarding the public service contract to an operator of a designated platform, the measure in question could not fall into the ambit of Article 106(2) TFEU relating to SGEI.

An appeal against this decision was lodged before the General Court notably on the ground that the European Commission had infringed Articles 106 and 107 TFEU as well as its obligation to state reasons. The General Court rejected the appeal and upheld the European Commission’s decision, confirming notably that the definition of the SGEI by the Spanish authorities was not clear.

Considering that the Commission’s control over the way in which a Member State defines a SGEI should be limited to verifying that the Member State had not committed a manifest error, the applicants argued that the General Court was mistaken when it found that the definition of the SGEI by the Spanish authorities was not clear enough.

In its judgement, the Court of Justice first recalled that, in principle and subject to the manifest error of assessment, Member States have a wide discretion regarding the scope and organisation of the SGEI that they decide to implement in their territory.

However, the Court also recalled that the power of the Member States concerning the definition of the SGEI was not unlimited. This is because the first Altmark condition is essentially intended to determine whether, first, the recipient undertaking actually has public service obligations to discharge and, second, whether those obligations are clearly defined in national law.

The Court considered in this respect that this requirement implies meeting at least the following minimum criteria: the existence of one or more acts of public authority specifying the nature, duration and scope of the public service obligations.

In that respect, a law identifying telecommunications services, including broadcast radio and television networks as a service of general interest is too general to conclude that terrestrial network operators are responsible for the execution of public service obligations clearly defined within the meaning of the first criterion of Altmark.

In addition the Court of Justice pointed out that the definition of a SGEI can be included in an act that is different from the act by which the public authority entrusts a company with the execution of the SGEI.
NoteworthySince the definition of SGEIs forms part of the first Altmark condition, it leaves real room for its judicial review.

The judgement of the Court of Justice is in line with the judgement dated 20 December 2017 (C-66/16P to C-69/16 P) according to which the definition of a SGEI has to be clear and meet minimum listed requirements.

This judgement confirms that the review of the EU Commission and of the EU courts in SGEI matters is not limited to the mere control of manifest errors in any respect. Distinctions have to be made. For example, the requirements for a clear definition of the SGEI and its public service obligations (“PSO”) and for an entrusting act (which form parts of the first criterion) as well as the need for objective, transparent and prior parameters of the compensation (second criterion) are subject to an in-depth scrutiny. More generally, it seems to us that the trend is towards a more stringent control by the EU Commission and the EU courts of the actions of the Member States in SGEI matters, however without jeopardizing their room for political manoeuvre. The EU Commission and the EU Courts only ensure that once a Member State decides to pursue an objective of general interest by setting up a SGEI, it designs it in a correct, serious and – to some extent - efficient way. This entails that the objective of general interest as well as PSOs must be genuine, the undertaking in charge of the SGEI must be capable of fulfilling the objectives assigned (that is to say being economically viable), there must be a genuine market failure, no overcompensation can be paid, the proceeding by which the undertaking is chosen has to be carefully designed and implemented, …. Fulfilling all of these conditions requires serious and significant preparation by the Member States and the other national and subnational public bodies involved. Public authorities should therefore adequately design their SGEI taking into consideration the decision-making practice of the EU Commission and the principles laid down in the EU courts’ case law. Otherwise, the SGEI may be subject to criticism and annulment by the EU Commission and the undertakings allegedly assigned with a SGEI risk being forced to pay back all the money granted by public bodies.

Public procurement & competition law : to what extent related undertakings may submit separate bids in the same tendering procedure?

PartiesJurisdictionFormationJudge RapporteurAdvocate GeneralSubject-matter
Preliminary rulingLloyd’s of London
Agenzia Regionale per la Protezione dell’Ambiente della Calabria (Aparcal)
Court of Justice Sixth ChamberE. ReganE. TanchevPublic procurement - Competition law
KeywordsReference for a preliminary ruling — Public procurement — Articles 49 and 56 TFEU — Directive 2004/18/EC — Reasons for exclusion from a tendering procedure — Insurance services — Participation of several Lloyd’s of London syndicates in the same tendering procedure — Signature of tenders by the Lloyd’s of London General Representative for the country concerned — Principles of transparency, equal treatment and non-discrimination — Proportionality
Significant pointsThe General Court has annulled a decision of This preliminary ruling relates to a tendering procedure launched by an Italian public authority, Arpacal, for the award for a public service contract for insurance. Two Lloyd’s ‘syndicates’ participated in the call for tenders. Their tenders were both signed by an agent of Lloyd’s General Representative for Italy. However, Arpacal excluded those syndicates from the tender due to the fact that the submissions were signed by the same person. It considered that a single decision-making centre would infringe the principles of confidentiality of tenders, equal treatment, fair and free competition, protected under EU law by Articles 49 and 56 TFEU and Directive 2004/18 on public procurement procedures (“Public Procurement Directive”). Lloyd’s argued, however, that those principles were respected because the members, whilst acting through syndicates, did operate independently and in competition with one another despite coming under the same legal entity. Moreover, the syndicates may only act through their sole representative for each Member State. The referring court observed that the Italian legislation did not forbid submissions to public tenders being signed by the same person and asked the Court clarification in this regard.
In its judgment, the Court recalled, first, that the legal grounds for exclusion from public procurement procedures set out in Article 45 of the Public Procurement Directive relate to the professional qualities of the persons concerned only. Nonetheless, the case-law of the Court has interpreted that this does not preclude the option for Member States to maintain or establish, in addition to those grounds for exclusion, substantive rules intended, in particular, to ensure, with regard to public procurement, observance of the principles of equal treatment of all tenderers and of transparency, which constitute the basis of the EU directives on public procurement procedures, provided that the principle of proportionality is observed.
In that regard, the Court firstly observed that the national legislation at issue, which is intended to prevent any potential collusion between participants in the same procedure for the award of a public contract, seeks to safeguard the equal treatment of candidates and the transparency of the procedure. Consequently, the principle of proportionality requires that such legislation must not go beyond what is necessary to achieve the intended objective.
The Court judged that the automatic exclusion of tenderers that are in a relationship of association with other competitors would go beyond what is necessary to prevent collusive behavior and to ensure equal treatment and transparency. An automatic exclusion would constitute an irrebuttable presumption of mutual interference and would not grant the possibility for the candidates of demonstrating that their tenders are in fact independent. Yet previous case law provides that it is of EU interest to ensure the widest possible participation by tenderers in a call for tenders.
The Court concluded, therefore, that the principle of proportionality requires that the contracting authority be required to examine and assess the facts, in order to determine whether the relationship between two entities has actually influenced the respective content of the tenders submitted in the same tendering procedure, a finding of such influence, in any form, being sufficient for those undertakings to be excluded from the procedure.
The Court also stated that the mere fact that tenders have been signed by the same person cannot justify their automatic exclusion from the tendering procedure. It noted in this regard that EU law applicable to insurance activities expressly allows Lloyd’s to be represented vis-à-vis third parties by a single representative for each Member State. Even though Lloyd’s may exercise its insurance activities in Member States only through the competent General Representative, the referring court must still verify that the tenders were determined and submitted independently by each syndicate.
Therefore, the Court considered that the national legislation at issue was compatible with EU law, given that it does not allow an automatic exclusion, but nonetheless allows the contracting authority to exclude tenderers where it finds, on the basis of unambiguous evidence, that their tenders were not drawn up independently.
NoteworthyIn this judgment completely in line with the Assitur case-law (C-538/07), the Court of Justice has again emphasized the importance of the principles of transparency, equal treatment and free competition in public procurement. Any exclusions from the right to tender are to be interpreted strictly and based on a factual analysis, this respecting the principle of proportionality.
The Court explained in length that the bids of undertakings linked by a relationship of control or of association should not be presumed collusive for the purpose of the EU public procurement rules and that a contracting authority had to look at the facts before any exclusion of linked undertakings. It is up to the contracting authority to establish whether the relationship between two entities has actually influenced the respective content of the tenders submitted in the same tendering procedure.
It can be interesting to compare this judgement with competition law, where it is possible for undertakings which share legal or financial links but enjoy commercial autonomy to submit separate and competing tenders, provided that they do not consult each other prior to the submission of their bids, such exchanges having an anti-competitive object (see for example, in French competition law, Paris Court of Appeal, 28 October 2010, n° 2010/03405). Accordingly, it seems that a violation of competition law would be more easily retained, given that the mere false autonomy of the bids allows the existence of a cartel to be presumed, whereas an actual influence on the content of the bids seems to be required to find a violation of EU public procurement law. Further explanations regarding this apparently partial convergence might be provided by the Specializuotas transportas upcoming judgment (C‑531/16).
Of particular importance to the insurance sector is the fact that Lloyd’s syndicates have the right to participate in the same tendering procedure and be represented by the same representative. This does not constitute a ground for exclusion from the procedure so long as the syndicates act independently in the formulation of their submissions.

Application of State aid and Competition law rules to the health sector reinforced by the General Court

PartiesJurisdictionFormationJudge RapporteurAdvocate GeneralSubject-matter
Non-contractual liabilityDôvera zdravotná poist'ovňa et al v European CommissionGeneral CourtSecond ChamberM. J. Costeira-State aid - Health insurance bodies
KeywordsState aid — Health insurance bodies — Capital increase, debt repayment, subsidies and Risk Equalisation Scheme — Decision finding no State aid — Concept of State aid — Concept of undertaking and economic activity — Principle of solidarity — State supervision — Activity that is economic in nature — Competition on quality — Presence of operators seeking to make a profit — Pursuit, use and distribution of profits — Error of law — Error of assessment
Significant pointsThe General Court has annulled a decision of the EU Commission finding no state aid because the recipient of the measure, a Slovak health insurance body was not an undertaking.

According to the General Court, a health insurance body has to be considered an undertaking and is therefore susceptible to benefit from State aid where it offers goods and services on a market and is in competition as regards the quality and scope of services with operators seeking to make a profit and has the ability to make, use and distribute part of its profits, notwithstanding the social and solidarity nature of certain other features of the health system.

The General Court reached this conclusion after having checked first whether the health insurance body met three cumulative criteria which would mean that it should be considered as not pursuing an economic activity. These are: the existence of a social aim of a health insurance scheme, the implementation of the principle of solidarity by this scheme and the supervision by the State.

In this respect, the General Court found that the second criterion was not fulfilled.

On the one hand, health insurance companies’ ability to seek and make a profit showed that, regardless of the performance of their public health insurance task and of State supervision, they were pursuing financial gains and, consequently, their activities in the sector fell within the economic sphere. Therefore, the strict conditions framing the subsequent use and distribution of profit which may result from those activities does not call into question the economic nature of such activities.

On the other hand, the existence of a certain amount of competition as to the quality and scope of services provided by the various bodies within the Slovak compulsory health insurance scheme also had a bearing on the economic nature of the activity. Indeed, the companies could freely supplement the compulsory statutory services with related free services, such as better coverage for certain complementary and preventive treatment in the context of the basic compulsory services or an enhanced assistance service for insured persons. They compete through the ‘value for money’ over the cover they offer and, therefore, on the quality and efficiency of the purchasing process.

Second, the General Court added that, assuming that some health insurance bodies were not seeking to make a profit, they amount to undertakings all the same, provided that the offer exists in competition with that of other operators that are seeking to make a profit. Where other operators on the market in question are seeking to make a profit, the entities involved would have to be considered undertakings too ‘by contagion’.
NoteworthyThere are several lessons to be learnt from this judgment.

First, the delineation between the economic activity of health insurance and social security schemes, which do not fall within the ambit of Competition and State aid law, was once again at stake, (cf caselaw Poucet & Pistre, C-159/91; FFSA and Others, C-244/94; Cisal, C-218/00; AOK Bundesverband and Others, C-355/00; AG2R Prévoyance, C-437/09) and, In accordance with well-established case law, the General Court resorted to the body of evidence technique to reach the overall conclusion that the entity at stake behaves like an undertaking. Key elements were the commercial and managerial autonomy and the competition with profit-driven operators.

Second, after the Iris Judgment (T-137/10), where a decision of the EU Commission not to open formal proceeding against public hospitals in relation with alleged over-compensation was annulled, the General Court confirms its principle in favour of the full application of State aid law to the health sector. By contrast, the EU Commission appears more inclined to close its eyes to and ignore distortions of competition in light of social considerations.

The stringent approach of the General Court is likely based on the belief that social and solidarity goals should not be a blanket for excessive expenses, unnecessary competition distortions, hurdles to innovation and, at the end of the day, stagnation and a poor level of services. In its view, the Commission may not purely abstain from monitoring whether a public financing to health insurance bodies (like in the case at hand) or hospitals is indeed necessary, justified and limited to what is required.

Third, the judgment appears unclear, however, as to the respective weight of the criteria taken into account by the General Court and notably the one on the existence on the concerned market of profit-driven operators. It cannot be ruled out that the General Court paid attention to strengthen the reasoning of its decision to minimize the risk of appeal and possible annulment by the ECJ.

Certainly, the existence on the concerned market of profit-driven operators makes the other operators become undertakings ‘by contagion’. However, the judgment leaves unanswered the question to what extent operators are undertakings when on the contrary none of them is profit driven.

In this respect, the General Court seems to have been ill at ease in rebutting arguments raised from the AOK judgment (referred to above). According to this judgment, where bodies have a degree of freedom to compete to a certain extent in order to attract persons seeking insurance, that competition does not automatically call into question the non-economic nature of their activity, particularly where that element of competition was introduced in order to encourage the sickness funds to operate in accordance with principles of sound management. In our opinion, the General Court has overstated the scope of the AOK case law. In this case, the sickness funds were not competitors since there was a system of neutralisation and compensation between each other so that they were deprived of financial autonomy and were part of a single system of sickness funds. In addition, they did not compete with private operators. Put in other words, they did not form separate entities enjoying autonomy (one of the constituting elements of an undertaking) and there was not a market for services open to competition.

In our opinion, it would have been more appropriate for the General Court to rule that, as soon as bodies have some freedom to compete, for example as regards the offer of services and/or the quality of the services and form distinct and autonomous entities, they amount to undertakings, no matter whether they or some of them are able to seek and make a profit or not.

Fourth, on a more general level, it cannot ruled out that the rather extensive notion of undertaking developed by the General Court in this judgment could have an impact going beyond the matter of health insurance. It might be of relevance to other areas where entities are assigned tasks of general interest while providing goods or services on a market where other independent operators are active.

After ICAP, new developments from the Court of Justice to the notion of a ‘by object’ restriction

PartiesJurisdictionFormationJudge RapporteurAdvocate GeneralSubject-matter
Non-contractual liabilityHoffmann-La Roche e.a. v Autorità Garante della Concorrenza e del MercatoCourt of JusticeGrand ChamberC.G. FernlundH. Saugmandsgaard ØeAntitrust – Concerted practice
KeywordsReference for a preliminary ruling — Competition — Article 101 TFEU — Agreements, decisions and concerted practices — Medicinal products — Directive 2001/83/EC — Regulation (EC) No 726/2004 — Allegations of risks associated with the use of a medicinal product for a treatment not covered by its marketing authorisation (off-label) — Definition of relevant market — Ancillary restriction — Restriction of competition by object — Exemption
Significant pointsThis case concerned the marketing of two medicinal products developed by Genetech, a Roche subsidiary, one for the treatment of cancer (Avastin) and the other for the treatment of ophthalmological conditions (Lucentis). Avastin was marketed by Roche itself through a licensing agreement with its subsidiary, while Novartis commercialised Lucentis through another licensing agreement with Genentech.

Following an investigation into these arrangements, the Italian Competition Authority imposed a fine on both pharmaceutical companies on the grounds of anticompetitive behavior pursuant to Article 101 TFEU. The authority found that they had concluded an agreement which was designed to disseminate misleading information relating to the adverse reactions resulting from the off-label use of Avastin in the field of ophthalmology.

The Italian Supreme Court, before which the case was brought on appeal, referred five questions to the Court of Justice for a preliminary ruling.

Firstly, the Court addressed a question related to the definition of the relevant market when a medicine is used outside of its marketing authorisation (MA). The Court considered that the fact that pharmaceutical products are manufactured or sold illegally prevents them, in principle, from being regarded as substitutable or interchangeable products. However, the Court noted that the EU rules on pharmaceutical products do not prohibit as such the off-label prescription of a medicine, provided that it complies with the conditions laid down in those rules. Given that compliance verification is not the responsibility of competition authorities but of pharmaceutical regulatory authorities (or national courts), the Court judged that, in order to assess whether a medicine whose MA does not cover the treatment of certain diseases falls within the same relevant market as a medicine with a MA, a competition authority must take into account the outcome of the compliance examination that may be carried out by the competent authorities (paras 48 to 61).

Secondly, the Court discussed whether the restriction described above may fall outside the scope of Article 101 TFEU as an ancillary restraint to a licensing agreement. This argument was promptly dismissed by the Court on two grounds: first, the disputed conduct was not designed to restrict the commercial autonomy of the parties to the licensing agreement but rather the conduct of third parties (in particular healthcare professionals); second, the restraint was agreed upon several years after the licensing agreement was concluded and so was not necessary for the latter (paras 68 to 75).

Finally, the Court addressed the main issue at stake, namely whether such an arrangement constituted a restriction of competition ‘by object’. The arrangement between the two undertakings marketing the two competing products concerned the dissemination of information, in a context of scientific uncertainty on the matter, relating to adverse reactions resulting from the use of one of those medicinal products for illnesses not covered by its MA, with a view to reducing the competitive pressure resulting from that use on another medicinal product covered by an MA covering those illnesses. On this point, the Court first noted that the fact that two undertakings colluded with each other with a view to disseminating information specifically relating to the product marketed by only one of them might constitute evidence that the dissemination of information pursues objectives unrelated to pharmacovigilance, given that pharmacovigilance obligations apply only to the company which markets the product. Then the Court observed that the dissemination of misleading information (i) encouraged doctors to refrain from prescribing the product, thus resulting in a reduction in demand, and (ii) constituted an infringement of the EU pharmaceutical regulations. Given these circumstances, the Court stated that the restriction at issue constitutes a restriction of competition ‘by object’ to the extent that the referring court confirms the misleading nature of the information communicated to the pharmaceutical regulatory authorities and the general public (paras 91 to 95).
NoteworthyA couple of weeks after the Icap judgment of the General Court (T-180/15), the Court of Justice brings further developments to the notion of a ‘by object’ restriction of competition. The Court confirms its use of this notion, which exempts competition authorities from the burden of proving the anticompetitive effect of a particular practice. The notion proves to be quite open and not limited to an exhaustive list. The Court indeed recalls that the ‘by object’ category is not limited solely to prima facie restrictions of competition. In this case, the infringement is an atypical consisting in the artificial creation of two different markets (even though it that has some analogies with a market-sharing cartel). The notion of ‘by object restriction’ also allows for consideration of the economic and legal context of the practice. In addition, in the case at hand, the Court relies on an in-depth analysis of the regulatory context of the agreement. On this basis, the Court uses the body of evidence method. On the one hand, it takes into account the incongruous nature of the agreement. On the other hand, the Court takes into consideration the "expected" result of the collusion. While it is well established case-law that anti-competitive intent is not a necessary element in determining whether an agreement has the object of restricting competition, the Court seems to recall here that this intention can nevertheless be taken into account.