OPINION OF ADVOCATE GENERAL WAHL delivered on 23 April 2015 ( [1]1 ) Case C-172/14 ING Pensii -- Societate de Administrare a unui Fond de Pensii Administrat Privat SA v Consiliul Concurentei (Request for a preliminary ruling from the Înalta Curte de Casatie si Justitie (Romania)) 'Reference for a preliminary ruling -- Competition -- Agreements, decisions and concerted practices -- Arrangements for allocating clients on the Romanian market for the private management of compulsory pension funds -- Existence of a restriction of competition `by object' for the purposes of Article 101(1) TFEU -- Effect on trade between Member States' 1. The present case originates from legal proceedings between ING Pensii -- Societate de Administrare a unui Fond de Pensii Administrat Privat SA (`ING Pensii'), a company managing a private pension fund, and the Consiliul Concurentei (Romanian competition authority), for the annulment of a decision ordering the company to pay a fine for participation in an agreement the aim of which was to restrict competition on the private pension fund market. It relates more specifically to agreements, established by the managers of mandatory private pension funds, to allocate `duplications', that is individuals who, either through ignorance of the relevant rules or through the negligence of certain commercial agents, subscribed to two or more funds during the initial affiliation procedure provided for by statute in 2007 (`duplications'). 2. This reference for a preliminary ruling, the reason for which is, to a large extent, due to differences in interpretation by the Romanian courts, requires the Court to provide clarification regarding the concepts of restriction `by object' and `appreciable distortion of competition'. It provides, in particular, an opportunity to recall the factors which must be taken into account and the method of analysis which must be used in determining whether an agreement has an anti-competitive object for the purposes of Article 101(1) TFEU. I - National legal framework 3. The system for allocating duplications at issue in this case forms part of Romanian national legislation relating to the establishment, organisation and operation of the compulsory private pension fund market. 4. The relevant national law is undeniably characterised by a certain degree of complexity. While it cannot be claimed that they are exhaustive, for the purposes of the analysis below I would like to draw attention to the considerations which follow. A - General rules relating to affiliation to private pension funds 5. The pension system in Romania is composed of the following: -- first pillar: mandatory component, based on redistribution, publicly managed; -- second pillar: mandatory component, based on capitalisation, privately managed by public limited pension fund management companies which are governed by Law No 411/2004 on private pension funds (legea privind fondurile e pensii administrate privat); ( [2]2 ) -- third pillar: voluntary component, based on capitalisation, also privately managed. 6. Pursuant to Law No 411/2004, 18 commercial companies with the management of pension funds as their sole object were approved by the Comisia de Supraveghere a Sistemului de Pensii Private (Private Pension Schemes Board) (`CSSPP') during the period from 25 July 2007 to 9 October 2007, each company being entitled to manage only one pension fund in Romania. 7. Under Article 30(1) of Law No 411/2004, persons aged 35 years or under who are members of and pay contributions to the public pension scheme (first pillar) must also belong to a private pension fund (second pillar). 8. Persons aged over 45 and those who do not contribute to the public pension scheme (first pillar) may not belong to the second pillar. Persons aged between 35 and 45 who contribute under the first pillar may choose whether or not to participate under the second pillar (Article 30(2) of Law No 411/2004). Regardless of whether they make contributions under the first and second pillars, any natural person may contribute under the third pillar, although that cannot have the effect of replacing contributions under the first and second pillars in the case of persons obliged to contribute to those pension schemes. 9. In relation to participants under the second pillar, who are the only ones affected by the case in the main proceedings, Article 31 of Law No 411/2004 provides that a person may not belong at the same time to more than one pension fund and may have only one account with the pension fund to which he belongs. 10. Under Article 32(1) of Law No 411/2004, a person may register with a pension fund only by signing an individual affiliation application, on his own initiative, or following his allocation to the fund by the census authority. 11. Under Article 33 of Law No 411/2004, any person who has not already become affiliated to one of the private pension funds within a certain time-limit is to be allocated on a random basis to a pension fund by the census authority. This allocation is carried out in proportion to the market share obtained by each of the fund managers (Article 33(2)). B - Specific rules relating to the initial affiliation to a private pension fund, the procedure for validating affiliations and the random allocation of non-affiliated participants 12. The procedure for initial affiliation to a private pension fund and the procedure for validating and allocating participants were governed by Order No 18/2007 concerning initial affiliation and registration of participants in private pension funds, ( [3]3 ) as amended and supplemented by Order No 31/2007 (`Order No 18/2007'). ( [4]4 ) 13. Pursuant to Article 17(1) of Order No 18/2007, the initial procedure for affiliation to a private pension fund was to extend over a period of four months, which commenced on 17 September 2007 and ended on 17 January 2008 (Article 5(6) of Order No 18/2007). During that period, any persons who had not reached the age of 35 by 31 December 2007, or who reached that age on that date, and who met certain other requirements, were obliged to join a private pension fund (Article 4(1) and Article 5(3) of Order No 18/2007). 14. In relation to the validation procedure and the random allocation of participants (Articles 19 to 31), Order No 18/2007 required managers to report twice per month (on the 1st and 15th of every month) to the National Office for Pensions and other Welfare Benefits (Casa Nationala de Pensii si alte Drepturi de Asigurari Sociale, `the CNPAS') with the details of persons who had signed an individual affiliation application during the two-week period preceding the filing of the report. If, in one of those bi-monthly reports, a person was named by one or more managers as having signed more than one individual affiliation application, or if it was ascertained that that person's application had been validated on a temporary basis in an earlier report, the CNPAS was to enter the name of that person in the electronic table of duplications (Article 21(1) of Order No 18/2007). In that event, the fund managers were required by law to verify the authenticity of the individual affiliation application and the copy of the identity card each bearing the signature of the person who appeared to have signed more than one affiliation application and, if the application was found to be genuine, the managers could re-submit the details of the affiliation of the person concerned to the CNPAS in their next bi-monthly report. 15. At the end of the initial affiliation procedure, those persons whose affiliation had appeared to be validated on a temporary basis were registered as validated, and those who appeared to have signed more than one individual affiliation application were entered on the register of participants as not validated and were randomly allocated. ( [5]5 ) Consequently, random allocation was to apply to those individuals who, although obliged to join a private pension fund during the initial affiliation stage, had failed to do so, and to those registered as persons whose affiliation had not been validated. Allocation to one of the private funds was carried out by the CNPAS, on a basis directly proportional to the number of persons whose affiliation had been validated for each private pension fund by reference to the total number of persons whose affiliation had been validated across all the private pension funds. 16. Order No 31/2007, which amended and supplemented Order No 18/2007, also required managers to notify participants, within 15 days of the date of validation, of the pension fund with which their affiliation had been validated. If the participants so notified had not signed an individual affiliation application for the fund in question, they had the opportunity to submit a written request to the CSSPP and, once a decision had been taken on the request, the CSSPP would inform the manager whether or not the participant was affiliated to the privately-managed fund. II - Facts, procedure and the question referred 17. ING Pensii is a private company that manages pension funds. It operates, inter alia, in the compulsory private retirement pension market in Romania. 18. Underlying the present case is an investigation carried out by the Consiliul Concurentei in relation to a potential infringement of Article 5(1) of Law No 21/1996 on competition (legea concurentei) ( [6]6 ) and of Article 81 EC (now Article 101 TFEU) on the Romanian compulsory private pension fund market (second pillar). 19. At the end of that investigation, the Consiliul Concurentei concluded, in particular, that anti-competitive agreements existed on that market to share clients between the commercial companies managing those funds. 20. By Decision No 39 of 7 September 2010, the Consiliul Concurentei fined 14 of those companies, including the applicant. The agreements in question, which were for the most part entered into on a bilateral basis and took various forms, concerned duplications. By entering into the agreements, the private pension fund administrators shared out the duplications equally between them (on a 50/50 basis) and thus sought to avoid any allocation of the duplications by the CNPAS. 21. On 4 October 2010, ING Pensii claimed that the Curtea de Apel Bucuresti (Court of Appeal, Bucharest) should annul Decision No 39/2010 and, in the alternative, annul in part that decision with a view to reducing the amount of the fine imposed. 22. The applicant claimed that the agreement in question did not infringe Article 5(1) of the Law on competition and that the conditions necessary for Article 101(1) TFEU to apply were not met. In particular, the applicant submitted that the allocation of participants registered as duplications had not had the effect of restricting, preventing or distorting competition on the Romanian compulsory private pension fund market or any significant part of that market. ING Pensii also argued that competition between managers of private pension funds had not been eliminated since they had been in competition with each other during the initial affiliation period. 23. The Consiliul Concurentei argued that, in order to establish the anti-competitive nature of the arrangements agreed between the private pension fund managers, including ING Pensii, it was necessary to take account of the legal framework on which the establishment and operation of the compulsory private pension fund market was based, as well as the specific features of the market on which those arrangements had been made. 24. By judgment No 749 of 6 February 2012, ING Pensii's appeal was dismissed by the Curtea de Apel Bucuresti. 25. The applicant then lodged an appeal in cassation before the referring court. It submitted, inter alia, that choosing an algorithm for calculating duplications other than the statutory algorithm did not constitute an infringement of the Law on competition but, at most, an infringement of the specific legislation applicable to the compulsory private pension sector. In addition, since the arrangement was limited to the allocation of duplications, it could not have affected competition on the market in question, since the duplications, the number of which accounted for less than 1.5% of the market, were not the subject of competition between the managers of the private pension funds. 26. ING Pensii also submitted that it had no practical or economic interest in the allocation of duplications in equal shares, given that it already had the greatest share of the market as at 15 October 2007. Moreover, the agreement had had a positive outcome in that it made the compulsory private pension affiliation procedure more efficient by giving participants a greater chance of being granted their choice than would have been the case with random allocation. 27. Finally, the applicant stated that, in the present case, there was no evidence of any partitioning of the national compulsory private pensions market as a result of the choice of a different algorithm for calculating duplications. It was clear that the actual or potential effects of an agreement affecting a marginal percentage of the Romanian market were negligible and not at all of the kind to have any impact on the European Union market. 28. The Consiliul Concurentei contended that the appeal should be dismissed, submitting in essence that the agreement for allocating duplications was capable of distorting competition on the compulsory private pensions market and, as such, pursued an anti-competitive object. According to the Consiliul Concurentei, the ability of an agreement to produce negative effects and the finding of an infringement consisting in the sharing of markets and sources of supply was not dependent on the number of clients actually shared, which is a factor relevant to the actual effects of an agreement. 29. In those circumstances, the referring court decided to stay proceedings and to refer the following question to the Court for a preliminary ruling: `In relation to a practice by virtue of which clients are shared out, is the specific and definitive number of those clients relevant in deciding whether the condition of a significant distortion of competition for the purposes of Article 101(1)(c) TFEU is fulfilled?' 30. ING Pensii, the Consiliul Concurentei, the Romanian Government and the European Commission submitted written observations to the Court and presented oral argument at the hearing held on 11 February 2015. III - Analysis 31. This request for a preliminary ruling invites the Court to indicate whether and to what extent the number of persons affected by sharing agreements such as those at issue in the main proceedings is relevant for the purposes of determining whether those agreements are caught by the prohibition on such agreements laid down by Article 101(1) TFEU. 32. Bearing in mind the information provided by the referring court, two preliminary observations must be made. 33. First, it is my opinion that, while in the main proceedings it would appear that it is only the application of national provisions of Romanian competition law, in particular Article 5(1) of the Law on competition, that is directly called into question, the admissibility of the question must clearly be conceded. In this instance, ING Pensii called into question the applicability of Article 101 TFEU, but that objection was dismissed by the referring court. Since it is precisely the conditions under which that provision is applicable that constitute the key issue of the case, one can only endorse the approach adopted by that court. 34. In addition, it should be noted that that national provision, in the version currently in force, reproduces almost word for word the prohibition on agreements restricting competition laid down in Article 101(1) TFEU. In those circumstances, I consider that it is necessary, in line with the solutions adopted in numerous previous cases relating to restrictive practices under national competition law, to conclude that the present reference for a preliminary ruling is admissible. ( [7]7 ) It is clearly in the EU interest, in order to avoid future differences of interpretation, that provisions or concepts taken from EU law should be interpreted uniformly, irrespective of the circumstances in which they are to apply. ( [8]8 ) 35. Secondly, it should be pointed out that, while the issue at the heart of the debate in the main proceedings is the existence of a restriction of competition `by object', which will form the main focus of the issues addressed in this Opinion, the general terms used by the referring court in wording the question referred for a preliminary ruling also suggest that a wider consideration of the condition relating to a significant effect on competition is called for. The referring court has also made it clear that, in this case, it is seeking a ruling on both the anti-competitive nature of the agreement in question and on the question whether the agreements to share duplications were capable of altering the structure of the compulsory private pension fund market in that they have a significant effect on that market. A - The relevance of the number of persons actually affected by the agreements at issue in determining whether they seek to restrict competition 36. In this case, in order to answer the question put by the referring court, it seems to me imperative, having reviewed the relevant case-law, to examine the sharing agreements at issue. While, ultimately, it is only the referring court that can decide whether those agreements are such as to be caught by the prohibition of concerted practices laid down by Article 101(1) TFEU, the Court of Justice is none the less invited to provide guidance to the national court in determining whether those agreements have an anti-competitive object or, if they do not, whether they are anti-competitive in their effect. 1. Review of case-law 37. It must be recalled that, to be caught by the prohibition laid down in Article 101(1) TFEU, an agreement, a decision by an association of undertakings or a concerted practice must have `as [its] object or effect' ( [9]9 ) the prevention, restriction or distortion of competition in the internal market. 38. According to settled case-law established by the judgment in LTM, ( [10]10 ) the alternative nature of that requirement indicated by the conjunction `or', leads to the need to consider, in the first place, the precise purpose of the agreement, in the economic context in which it is to be pursued. Accordingly, where the anti-competitive object of an agreement is established, it is not necessary to examine its effects on competition. 39. According to settled case-law which was referred to recently and emphatically in the case giving rise to the judgment in CB v Commission, ( [11]11 ) the distinction between infringements `by object' and infringements `by effect', derives from the fact that certain types of coordination between undertakings cause a sufficient degree of harm to competition so that it may be concluded that there is no need to examine their effects. 40. In this regard, it is now well-recognised that certain types of coordination between undertakings can be regarded, by their very nature, as being injurious to the proper functioning of normal competition. Thus, it must be acknowledged that certain kinds of collusive behaviour, such as that leading to horizontal price-fixing by cartels, may be considered so likely to have negative effects, in particular on the price, quantity or quality of the goods and services in question, that it may be considered redundant, for the purposes of applying Article 101(1) TFEU, to prove that they have actual effects on the market. Experience shows that such behaviour leads to falls in production and price increases, resulting in poor allocation of resources to the detriment, in particular, of consumers. ( [12]12 ) 41. In order to determine whether an agreement between undertakings or a decision by an association of undertakings causes a sufficient degree of harm to competition so that it may be considered a restriction of competition `by object' for the purposes of Article 101(1) TFEU, regard must be had to the content of its provisions, its objectives and the economic and legal context of which it forms part. When determining that context, it is also necessary to take into consideration the nature of the goods or services affected, as well as the actual conditions of the functioning and structure of the market or markets in question. In addition, although the parties' intention is not a necessary factor in determining whether an agreement between undertakings is restrictive, there is nothing to prevent the competition authorities, the national courts or the Courts of the European Union from taking that factor into account. ( [13]13 ) 42. As I observed in my Opinion in CB v Commission, ( [14]14 ) while the more standardised assessment resulting from recourse to the concept of restriction `by object' requires a detailed, individual examination of the agreement in question, that examination must, none the less, be clearly distinguished from the examination of the actual or potential effects of the conduct of the undertakings concerned. 43. In my view, consideration of the economic and legal context in order to identify an anti-competitive object must, at the risk of introducing a shift that is detrimental to a proper reading of Article 101(1) TFEU, be clearly distinguished from the demonstration of anti-competitive effects under the second part of the alternative referred to in that provision. Consideration of the context in identifying the anti-competitive object can only serve to reinforce or counteract the actual examination of the terms of a purported restrictive agreement. It can in no way remedy a failure actually to identify an anti-competitive object by demonstrating the potential effects of the measures in question. In other words, regardless of the conceptual similarities between the two parts of that alternative, recourse to the economic and legal context in identifying a restriction `by object' cannot lead to a classification to the detriment of the undertakings concerned in the case of an agreement whose terms do not appear to be harmful to competition. ( [15]15 ) 44. It is in the light of these considerations that I propose to examine the agreements at issue, after briefly setting out the general context and manner in which they came into existence. 2. Description of the general context and manner in which the agreements at issue came into existence 45. The context of the bilateral agreements to share duplications entered into by the private pension fund managers which are at issue in the main proceedings is the requirement, as of 2007, for persons under 35 years of age (those falling within the second pillar), to join one of the 18 private pension funds approved for that purpose. 46. Bearing in mind that it was compulsory for those persons falling within the second pillar to be affiliated to one (and one only) pension fund, it could not be envisaged that, at the end of the initial affiliation period, which lasted approximately four months, the individuals concerned would not be deemed affiliated to a specific fund. 47. In that context it is important to note that, under the legislation in force, affiliation to a pension fund is recognised as legally valid only when registered with the CNPAS. Prior to that registration, there is no valid affiliation or presumption of affiliation to a particular fund. It should also be noted that it is not possible for an individual who has affiliated to a private pension fund and has been registered as such with the CNPAS to alter his affiliation within a period of 2 years without having to pay significant charges. This, to a degree, fixes affiliations for the future. 48. Therefore, in cases where participants were named in the reports of more than one manager (duplications) and it was not possible, for one reason or another, to verify the affiliation application, the legislature had provided for the participants to be allocated `randomly' on a basis directly proportional to the number of persons whose affiliation had been validated for each private pension fund by reference to the total number of persons whose affiliation had been validated across all the private pension funds. 49. It seems that, in the main, this allocation reflected competition as it stood between the participating pension funds during the affiliation procedure, because it was proportional to the number of persons validated by each fund at the time of allocation. It is apparent from the information provided to the Court by the Consiliul Concurentei that allocation by reference to the number of validated affiliation applications was a measure designed to encourage more active participation among the market players with a view to attracting participants and guaranteeing that the market operated on a competitive basis. 50. The information provided by the Consiliul Concurentei also shows that the conclusion of the sharing agreements at issue, which took place on an informal basis, commenced prior to the initial affiliation period, which began on 14 September 2007. ( [16]16 ) At that time, it was not possible to quantify or even to estimate the final number of participants registered as duplications. In other words, by entering into the agreements at issue, the fund managers decided to depart from the statutory system for allocation of duplications by sharing out their clients themselves without knowing the exact number of persons who would ultimately be affected. 51. Lastly, the allocation of participants registered as duplications was most frequently carried out on a bilateral basis between the majority of pension fund management companies. The equal sharing-out in question carried out among the managers led to the removal of the individuals concerned from the register of duplications and the validation of their affiliation to one of the two funds with the CNPAS. By not conducting the statutory control on authenticity of affiliations and proceeding to a 50/50 split, the managers knowingly avoided rules that reflected the result of the competition that the managers were supposed to deliver. 52. It is in the light of these considerations that it is necessary to examine whether or not the agreements at issue may be considered, in themselves, to involve a sufficient degree of harm to competition and, therefore, to be restrictive `by object'. 3. Examination of whether the duplication-sharing agreements at issue entail a restriction `by object' for the purpose of Article 101(1) TFEU 53. In order to do so, I am going to return to the analytical framework set out above and endeavour to examine in turn the content of the sharing agreements, their objectives and the economic and legal context of which they form part. a) The content of the sharing agreements at issue 54. On the face of it, by the agreements at issue, certain fund managers agreed to share out, on a bilateral basis, those individuals not definitively affiliated and registered as having signed more than one application to affiliate themselves to different pension funds. 55. In this way, the managers of the private pension funds whose participants appeared to have signed more than one individual affiliation application agreed to share out the duplications (who represented potential clients for all the management companies) on a 50/50 basis, whereas they should have first checked the true intention of any participant who appeared to have signed two affiliation applications and, where applicable, reported to the CNPAS any such affiliation application, as those applications had to be registered as invalid and such participants included in the random allocation system. 56. Even if, as ING Pensii submitted in its observations, it could to some degree be assumed that those individuals had indicated a particular preference to join one of the two funds in question, that does not alter the fact that, under the relevant national legislation, a person is recognised as a participant in a pension fund only if he signs an individual affiliation application, which must then be duly registered with the competent institution (Article 32(1) to 32(4) of Law No 411/2004). b) The objective aims of the agreements at issue 57. It seems pretty clear that the objective of entering into the agreements at issue was to derogate from the system for random allocation of duplications provided for in the relevant legislation. 58. Under that legislation, duplications were to be invalidated and allocated randomly across all managers who, during the initial affiliation period, numbered 18. By entering into the agreements at issue, the managers in question were assured of an equal split between them. 59. In view of the content and the objective aims of the agreements at issue, it would appear that they can be likened to client-sharing agreements that, by their very nature, are likely to involve a sufficient degree of harm. It appears that the agreements could distort competition in the compulsory private pensions market at a key stage in the establishment of that market. 60. In that context, it seems to me important to recall that, at the time the sharing agreements were entered into between the managers, which often occurred before the affiliation period had even begun, there was uncertainty as to how many people would be listed on the register of duplications. The ability of those agreements to have an appreciable effect on the market must be examined in the light of the factors known at the time the agreements were entered into and no account may be taken of findings of fact made after they were entered into. 61. In this case, the ability of the agreements at issue to alter the structure of the market, by removing a certain number of clients from competition between the pension funds, cannot be denied. This ability exists regardless of the actual effects which were, or which could have been, observed. 62. At this point in my examination, I am therefore of the opinion that the agreements in question are, prima facie, restrictive of competition by very reason of their object. 63. Taking into account the relevant economic and legal context does not seem to me to alter this initial conclusion, as I will explain below. c) Taking into account the economic and legal context in which the agreements at issue came into existence 64. As I have referred to above, the need to distinguish between the identification of a restriction `by object' and the identification of a restriction `by effect' means that examining the context in which the agreements at issue came into existence is not the same as examining the actual effects of the agreement. While the national court must in all cases carry out an individual assessment of the agreements in the legal and economic context in which they were entered into, that assessment does not amount to a precise examination of the actual and potential effects produced by the agreements or, if applicable, the significance of those effects. 65. Taking the context into consideration when identifying the anti-competitive object can therefore only reinforce or counteract the examination of the terms themselves and the objective aims of the alleged restrictive agreement. It cannot make up for the lack of any actual identification of an anti-competitive object by proving the potential effects of the measures concerned. This conclusion seems to me easily deducible from the wording adopted by the Court, which indicates quite clearly that, when the analysis of the content of the agreement does not reveal a sufficient degree of harm to competition, the effects of the agreement should then be considered and, for it to be caught by the prohibition, it is necessary to find that those factors are present which show that competition has in fact been prevented, restricted or distorted to an appreciable extent. ( [17]17 ) 66. That clarification seems significant to me in this case since, now that the content and the objective aims of the agreements at issue have been examined, it appears that the agreements, by their nature, involve a degree of harm to competition such that they should, a priori, be considered to be restrictive of competition `by object' for the purpose of Article 101(1) TFEU. 67. Therefore, examination of the context in circumstances such as those in the main proceedings must serve to determine whether, notwithstanding the fact that they are, prima facie, anti-competitive, it is possible to exonerate the private pension fund management companies in the light of the economic and legal context in which the sharing agreements at issue were entered into, or whether, on the contrary, examination of that context confirms the conclusion reached at the end of the examination of the terms and the objectives of those agreements, that they have an anti-competitive object. 68. With regard to the economic and legal contexts, which in this case are closely interlinked, it is apparent from the information supplied by both the referring court and the interested parties that the context in which the sharing agreements at issue came into existence is characterised by the considerations which follow. 69. First, the agreements were entered into in the context of an emerging infant market in Romania, namely the compulsory private pension fund market, which was set up in 2007 when the CSSPP granted authorisation to a certain number of companies the sole object of which was the administration of such funds. In the establishment of that market, which, according to the information supplied by the Consiliul Concurentei, affected approximately 4 million people, the initial affiliation period undeniably constituted a decisive stage for the purposes of determining the market structure and the shares held by each authorised management company. 70. From that perspective, the aim of the arithmetical method imposed by law for the random and proportional allocation of those individuals who had not joined a pension fund or who appeared not to have made a valid choice (such as the duplications), the number of which could not be known, was, inter alia, against a background in which competition between managers was being promoted, to preserve and guarantee a structure for that new compulsory private pension fund market that reflected the efforts made by each manager to attract the greatest number of participants during the initial affiliation stage and the choices actually and lawfully made by individuals. 71. Secondly, it should be pointed out that the compulsory private pension fund market was set up over a relatively short period of time, namely four months, at the end of which the position of each of the fund managers potentially had a decisive influence on the share of the market that they would subsequently hold. The very nature of that market and the fact that participants were subject to a substantial financial penalty (transfer penalties provided for by Law No 411/2004) if they should decide to change funds meant that the relevant consumers had limited room for manoeuvre and, to an extent, fixed the market share held by the companies concerned at the end of the affiliation period. 72. Thirdly, although, as observed by ING Pensii, certain managers maintained that the sharing agreements were aimed at alleviating practical difficulties and that it was necessary to remedy a statutory allocation mechanism that was considered, on the facts, to be ineffective, it must be pointed out that the agreements at issue were, in all likelihood, entered into before any difficulties connected with the setting up of the initial affiliation procedure had yet arisen. 73. Even if one accepts the applicant's argument that there were numerous shortcomings in the legislative framework for the establishment of the private pension fund market, that does not change the fact that those managers deliberately circumvented the statutory rules for allocating participants in such a way that was likely to alter the structure of the market as a result of their actions which brought about legally valid affiliations. 74. I would reiterate that, under the relevant legislation, duplications were to be rejected as invalid and allocated in a random manner in proportion to the market shares held at the end of the affiliation validation procedure. By entering into bilateral agreements to share out duplications between themselves, in the absence of any clear legally valid choice on the part of the participants, the fund managers circumvented the application of transparent rules, to the detriment of free competition. 75. As is apparent from these considerations, it does not seem to me that the contextual factors specific to the main proceedings are such as to invalidate the conclusion that the sharing agreements have an anti-competitive object. On the contrary, they serve to reinforce that conclusion. 76. It seems to me that that type of agreement conflicts patently with the concept inherent to the TFEU competition provisions, according to which each economic operator must determine independently the policy which it intends to adopt on the market and compete freely for any client on the market. By entering into those agreements, the fund managers colluded with each other in order to minimise the risks of competition. ( [18]18 ) 77. Consequently, I am of the opinion that the agreements to share out participants who had signed more than one affiliation application, concluded between the managers of private pension funds in breach of the statutory rules, fulfil all the conditions to be categorised as restrictive `by object'. 78. The question which remains to be examined is whether the number of persons actually affected by the sharing agreements may be relevant in determining whether there was an appreciable effect on competition. B - The relevance of the number of persons actually affected to the examination of a significant or appreciable effect on competition 79. All things considered and in view of the circumstances of the case, the question raised is in fact whether the actual effect of the sharing agreements at issue, which do indeed seem to have an anti-competitive object, may be relevant for the purpose of concluding that there is a restriction of competition. 80. I am of the view that that question must be answered in the negative. When an agreement is found to have an anti-competitive object, there is no need to examine the actual effects of the agreement. 81. Once the anti-competitive object of the agreements in question has been established, there is no longer any need to examine whether the agreements have an appreciable effect on market competition. Since anti-competitive conduct is prohibited by reason of its very nature, in the light of the degree of harm that it may cause to competition, it is completely unnecessary to characterise such conduct by reference to its economic impact or precise geographical extent. 82. Here, again, it should be recalled that an agreement that is restrictive `by object' is prohibited, whatever the ultimate consequences of the agreement, simply because of the potential danger to competition that it represents. As the Court stated, inter alia, in the case giving rise to the judgment in Expedia, ( [19]19 ) it must be held that an agreement that may affect trade between Member States and that has an anti-competitive object constitutes, by its nature and independently of any concrete effect that it may have, an appreciable restriction of competition (paragraph 37 of the judgment). 83. It follows from all the above considerations, that in circumstances such as those of the present case, the number of clients (duplications) that were in fact shared out between the private pension fund managers is of no relevance, as this question relates to the actual effects of an agreement, which, in the case of agreements which are restrictive `by object', do not require examination. The anti-competitive object of a sharing agreement, in particular the ability of such an agreement to produce negative effects on the market, does not depend on the actual number of clients actually shared out, but solely on the terms and objective aims of the agreement in the light of the economic and legal context in which it was entered into. 84. While it is unnecessary to carry out a full analysis of the effects of an agreement to establish that there is a restriction of competition `by object', the fact remains that the agreement in question must, objectively, be capable of giving rise to such a restriction. In other words, the courts cannot, when examining the economic and legal context of the agreements, completely ignore their potential effects in the light of the information known at the time the agreements were entered into. 85. I am of the view that the requirement for the restriction in question to be appreciable cannot be disregarded in the case of a restriction `by object', but that consideration of that requirement, to an extent, forms part of the examination of the economic and legal context carried out in order to confirm whether the agreement in question is in fact up capable of having such a restrictive effect. In the event that the court should conclude, at the end of the examination of the economic and legal context, that the agreement in question has the potential to produce only negligible effects on competition, in view, inter alia, of the weak position occupied by the relevant parties on the market for the products in question, it may conclude that the agreement falls outside the scope of Article 101(1) TFEU. 86. In the present case, it does not appear that the legal and economic circumstances in which the agreements came into existence, as referred to above, can cast any doubt on there being a restriction capable of having an appreciable effect on competition. IV - Conclusion 87. In light of the foregoing considerations, I propose that the Court answer the question submitted by the Înalta Curte de Casatie si Justitie as follows: Article 101(1)(c) TFEU must be interpreted as meaning that where there is an agreement to share out clients, such as the agreements entered into between the pension funds in the main proceedings, which, by its nature, constitutes an infringement as the agreement is anti-competitive by object, the final actual number of clients shared is irrelevant for the purpose of determining whether the condition relating to the restriction of competition for the purposes of Article 101(1)(c) TFEU has been fulfilled. __________________________________________________________________ ( [20]1 ) Original language: French. ( [21]2 ) Law No 411/2004 as republished in Monitorul Oficial al României No 482 of 18 July 2007, and amended and supplemented by Emergency Government Decree No 112/2007, published in Monitorul Oficial al României No 710 of 22 October 2007 (`Law No 411/2004'). ( [22]3 ) Published in Monitorul Oficial al României No 503 of 27 July 2007. ( [23]4 ) Published in Monitorul Oficial al României No 746 of 2 November 2007. ( [24]5 ) Following the amendments made by Order No 31/2007, the validation or non-validation of persons who signed more than one affiliation application was carried out monthly on 15 November (for the bi-monthly reports for 1 October, 15 October and 15 November), 15 December (for the reports of 1 December and 15 December) and on 17 January 2008 (for the bi-monthly reports of 1 January and [1]5 January 2008). ( [25]6 ) Law as amended, supplemented and re-published (Monitorul Oficial al României No 240 of 3 April 2014). Under that provision, `[e]xpress or tacit agreements between economic operators or associations of economic operators, decisions adopted by associations of economic operators and concerted practices shall be prohibited where they have as their object or effect the restriction, prevention or distortion of competition on the Romanian market or any part thereof and, in particular, where they are directed towards: ... (c) the sharing of markets or sources of supply ...' ( [26]7 ) See, in particular, judgments in Bronner (C-7/97, [27]EU:C:1998:569, paragraphs 12 to 22); Confederación Española de Empresarios de Estaciones de Servicio (C-217/05, [28]EU:C:2006:784, paragraphs 13 to 23); ETI and Others (C-280/06, [29]EU:C:2007:775, paragraphs 19 to 29) and Allianz Hungária Biztosító and Others (C-32/11, [30]EU:C:2013:160, paragraphs 17 to 23). ( [31]8 ) See judgment in Allianz Hungária Biztosító and Others (C-32/11, [32]EU:C:2013:160, paragraph 20 and the case-law cited). ( [33]9 ) Emphasis added. ( [34]10 ) 56/65, [35]EU:C:1966:38. ( [36]11 ) C-67/13 P, [37]EU:C:2014:2204, paragraph 49. ( [38]12 ) Judgment in CB v Commission (C-67/13 P, [39]EU:C:2014:2204, paragraphs 49 to 51 and the case-law cited). ( [40]13 ) Judgment in CB v Commission (C-67/13 P, [41]EU:C:2014:2204, paragraphs 53 and 54 and the case-law cited). ( [42]14 ) C-67/13 P, [43]EU:C:2014:1958, point 40. ( [44]15 ) Ibid., points 44 and 45. ( [45]16 ) According to the information received during the Consiliul Concurentei's investigation, one of the managers suggested to the competing companies on 15 August 2007 that they enter into bilateral and multilateral agreements establishing mutual validation criteria in the event of multiple affiliation applications to more than one fund and imposing financial penalties on companies that did not agree to those criteria. Agreements to share out duplications which departed from the statutory provision applicable were also proposed during a meeting of the Asociatia pentru Pensiile Administrate Privat din România (Association for Private Pensions in Romania). ( [46]17 ) See, in particular, judgments in LTM (56/65 [47]EU:C:1966:38, p. 359) and Football Association Premier League and Others (C-403/08 and C-429/08, [48]EU:C:2011:631, paragraph 135 and the case-law cited). ( [49]18 ) See, to that effect, judgment in Beef Industry Development Society and Barry Brothers, C-209/07, [50]EU:C:2008:643, paragraph 34. ( [51]19 ) C-226/11, [52]EU:C:2012:795. References 1. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0001 2. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0002 3. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0003 4. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0004 5. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0005 6. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0006 7. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0007 8. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0008 9. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0009 10. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0010 11. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0011 12. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0012 13. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0013 14. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0014 15. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0015 16. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0016 17. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0017 18. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0018 19. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#t-ECR_62014CC0172_EN_01-E0019 20. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0001 21. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0002 22. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0003 23. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0004 24. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0005 25. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0006 26. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0007 27. file:///../../legal-content/redirect/?urn=ecli:ECLI%3AEU%3AC%3A1998%3A569&lang=EN&format=pdf&target=null 28. file:///../../legal-content/redirect/?urn=ecli:ECLI%3AEU%3AC%3A2006%3A784&lang=EN&format=pdf&target=null 29. file:///../../legal-content/redirect/?urn=ecli:ECLI%3AEU%3AC%3A2007%3A775&lang=EN&format=pdf&target=null 30. file:///../../legal-content/redirect/?urn=ecli:ECLI%3AEU%3AC%3A2013%3A160&lang=EN&format=pdf&target=null 31. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0008 32. file:///../../legal-content/redirect/?urn=ecli:ECLI%3AEU%3AC%3A2013%3A160&lang=EN&format=pdf&target=null 33. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0009 34. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0010 35. file:///../../legal-content/redirect/?urn=ecli:ECLI%3AEU%3AC%3A1966%3A38&lang=EN&format=pdf&target=null 36. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0011 37. file:///../../legal-content/redirect/?urn=ecli:ECLI%3AEU%3AC%3A2014%3A2204&lang=EN&format=pdf&target=null 38. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0012 39. file:///../../legal-content/redirect/?urn=ecli:ECLI%3AEU%3AC%3A2014%3A2204&lang=EN&format=pdf&target=null 40. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0013 41. file:///../../legal-content/redirect/?urn=ecli:ECLI%3AEU%3AC%3A2014%3A2204&lang=EN&format=pdf&target=null 42. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0014 43. file:///../../legal-content/redirect/?urn=ecli:ECLI%3AEU%3AC%3A2014%3A1958&lang=EN&format=pdf&target=null 44. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0015 45. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0016 46. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0017 47. file:///../../legal-content/redirect/?urn=ecli:ECLI%3AEU%3AC%3A1966%3A38&lang=EN&format=pdf&target=null 48. file:///../../legal-content/redirect/?urn=ecli:ECLI%3AEU%3AC%3A2011%3A631&lang=EN&format=pdf&target=null 49. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0018 50. file:///../../legal-content/redirect/?urn=ecli:ECLI%3AEU%3AC%3A2008%3A643&lang=EN&format=pdf&target=null 51. file:///tmp/lynxXXXX2NsHdR/L90884-5802TMP.html#c-ECR_62014CC0172_EN_01-E0019 52. file:///../../legal-content/redirect/?urn=ecli:ECLI%3AEU%3AC%3A2012%3A795&lang=EN&format=pdf&target=null