When leagues and football clubs were faced with the suspension of football competitions as a result of the Covid-19 outbreak, they had to urgently ensure that leagues could be finalised at a later date than usual, preserving a level playing field for the (extended) duration of the season.
In Portugal, this led to professional football clubs concluding a temporary no-poach agreement, which was found by the Portuguese competition authority (AdC) to have the object of restricting competition. In this blog post, we discuss the Court of Justice’s preliminary ruling on questions referred to it by the Portuguese appeal court. The judgment develops several aspects of the concept of an object restriction, as we discuss below.
Background
As Covid-19 washed over Europe in the first quarter of 2020, the Portuguese league (LPFP) suspended the 2019/2020 First and Second Divisions. This created a problem for professional football clubs, as several contracts with players had termination provisions that were tied to the date of the end of the football season, set for end of June 2020. With the league likely extending into the summer, many players would become free agents before the end of the extended season, which could disrupt the league.
The clubs quickly sought to agree terms with the Portuguese union of football players, but this did not immediately yield an agreement to resolve the issue. At that point, the clubs took matters into their own hands and agreed, at a videoconference on 7 April, “that no club will hire a player who unilaterally terminates his employment contract, citing difficulties caused by the COVID-19 pandemic or by any exceptional decision arising therefrom, and in particular by the extension of the sporting season” (the “no-poach agreement”).
By 30 April, it became clear that it would be possible to finish the 2019/20 league later in the year, and by 4 May, the LPFP and relevant unions agreed to a memorandum of understanding under which employment contracts that were set to expire during the current season were deemed to be automatically extended until the end of the extended season. This was formalised in a collective labour agreement dated 8 June 2020. In the end, the season ended in August 2020, with Porto securing the league title with two games remaining by defeating Sporting CP.
The AdC was not happy with the no-poach agreement, and imposed interim measures on 26 May to suspend the agreement for 90 days. Despite the collective labour agreement overtaking events, the AdC continued its investigation and adopted an infringement decision on 28 April 2022, finding that the no-poach agreement constituted an agreement having as its object the restriction of competition. The LPFP and several clubs appealed that decision to the Tribunal da Concorrência, Regulação e Supervisão (the “referring court”). That court had doubts as to the application of Article 101(1) by the AdC, and decided to put several preliminary questions to the Court of Justice.
Summary of the Tondela ruling
The ruling is essentially split into two sections. The first discusses whether, given the specific context in which the no-poach agreement came into being, the AdC had properly applied the concept of a restriction “by object” in this case (the referring court’s third question). The second discusses how the referring court should apply the Meca Medina and Wouters case-law (the referring court’s first and second questions).
Object of the no-poach agreement
The Court starts by finding that the no-poach agreement does not qualify as a sporting rule solely adopted on non-economic grounds, which would fall outside of the EU competition rules. Clubs’ team sheets are an essential parameter of football competitions, and these competitions “give rise to an economic activity”. As a result, measures concerning recruitment and transfers have “a direct impact on the conditions for engaging in that economic activity and on competition between the professional football clubs engaged in that activity”.
To respond to the referring court’s question for guidance, the Court recalls that the concept of an object restriction refers only to certain types of coordination between undertakings which reveal a sufficient degree of harm to competition. This means that the conduct “must have a manifest anticompetitive rationale”, a new formulation of an existing concept in the Court’s case-law. The Court then goes on to explain that in some markets, the recruitment of highly skilled workers, such as professional football players, is an essential parameter of competition, such that limiting or controlling it would qualify as collusion between competitors in relation to resources they need. To determine whether a restriction of recruitment in such a market qualifies as an object restriction, account must be taken of its content, its economic and legal context, and its objectives.
As regards context, where the collusion is particularly harmful to competition (e.g. market sharing), the analysis may be limited to what is strictly necessary. Where it is not necessarily equally harmful to competition, it can still present a sufficient degree of harm to conclude that it has an anticompetitive object, but “a more in-depth examination” of context is necessary, taking into account the nature of the products or services concerned, and the real conditions of the structure and functioning of the sector(s) or market(s) in question. Such an analysis can lead to a finding that “particular circumstances” are such as to demonstrate that the conduct is not sufficiently harmful to justify a categorisation as having an anticompetitive object.
With respect to objectives, the Court refers to the need to establish the objective aims which the conduct seeks to achieve, from a competition standpoint. The fact that the undertakings acted without a subjective intention to restrict competition or whether they pursued certain legitimate objectives is not decisive.
The Court continues by assessing the content, context, and objectives of the no-poach agreement. As to its content, the Court finds that the football clubs jointly determined their conduct in the upstream market, which is the market for the recruitment of players. In doing so, they refrained from any possibility of deciding independently to recruit a player that fell within the scope of the no-poach agreement. Such an agreement manifestly restricts a competitive parameter which plays an essential role in high-level sport. Without the restriction, clubs could compete with each other for the affected players. A no-poach agreement artificially partitions the resources (i.e., the players). Those players are, in turn, restricted from offering their services to other undertakings, which limits their negotiating power. However, the examination of an agreement’s content alone is not sufficient for a finding that the no-poach agreement had an anticompetitive object.
As to the no-poach agreement’s legal and economic context, the Court notes, first, that football clubs compete with each other on different markets, including ticket sales, acquisition of sponsorships, and the exploitation of rights. Their position on those markets depends on their participation in sporting competitions. Those competitions are based on sporting merit, and depend on preserving a degree of equality of opportunity between clubs. It is legitimate for sports governing bodies to have rules relating to the organisation and conduct of competitions and the participation of players therein. Since sporting merit can only be guaranteed if all teams face each other in homogeneous regulatory and technical conditions, it may be legitimate to seek to ensure, to a certain extent, the stability of player rosters that serve as a pool for the teams which are put together by clubs during a given season. However, such rules must comply with Articles 101 and 102 TFEU. The no-poach agreement “partly ‘freezes’ the competitive relationship between those clubs as economic operators”.
Second, the Court discusses the relevance of the pandemic for the assessment of the object of the no-poach agreement. Most importantly, the Court found that Covid-19 created uncertainty as to the season’s end date and therefore the expiry of the contracts of certain players who could potentially have been snapped up by other clubs before the end of the extended season. This would inevitably have changed the teams involved, undermining the integrity of the remainder of the season. Although a crisis like the pandemic “is not per se such as to justify” an exception to Article 101(1), and this is no different in the field of sport, the referring court must nevertheless take into account the circumstances around the pandemic and its impact on professional football when analysing the context within which the no-poach agreement came into being.
Finally, the Court discusses how to approach the “competition-related objective aims pursued by the agreement”. In this respect, the no-poach agreement aimed at restricting competition on the player recruitment market, but also sought to mitigate the distortion to the proper progress of the Portuguese first and second divisions that would occur if certain players would be free to be recruited before the season ended.
The Court finds that the referring court may validly find that the no-poach agreement pursued both an anticompetitive and a procompetitive aim. Sporting merit “presupposes the maintenance of the integrity of the competition and the stability of the composition of the player rosters that serve as a pool for the teams which are put together by those clubs during a given competition or season”, according to the Court. Where a procompetitive aim is pursued in parallel with an anticompetitive one (as is the case here), this is fundamentally different from a situation where conduct pursues an anticompetitive aim while also pursuing other objectives which are possibly legitimate but neutral from a competition standpoint (as was argued to be the case in both Superleague and Diarra).
The Court concludes that although the no-poach agreement’s content was such as to manifestly restrict competition on the player recruitment market, the examination of the legal and economic context shows that the agreement came about “in a sector where competition features numerous specific characteristics in a very unique context”. Moreover, the agreement does not only pursue an objectively anticompetitive aim, but is also aimed at achieving an objectively procompetitive aim. It will now be for the referring court to determine whether or not, in view of all these aspects, the no-poach agreement presents a sufficient degree of harm to have the object of restricting competition.
Meca Medina and Wouters
The referring court’s first and second questions asked, in essence, whether the prohibition in Article 101(1) does not apply to the no-poach agreement because it was justified by the pursuit of one or more legitimate objectives in the public interest on the basis of the Wouters and Meca Medina case-law.
However, as the Court recalls, that case-law does not apply to object restrictions. The harm of an object restriction is too great for it to be regarded as justified and proportionate to the legitimate objective pursued. Only Article 101(3) can save such object restrictions. Therefore, only if the referring court finds that the no-poach agreement did not have the object of restricting competition, can it find that the Wouters and Meca Medina case-law applies to it. The Court, furthermore, clarifies that this case-law is applicable to any conduct by which an association of undertakings and that association’s members coordinate amongst themselves, irrespective of what form that coordination may take, and not only to decisions by an association of undertakings.
The Court considers, in line with its case-law, that “ensuring the regularity of sporting competitions” is a legitimate objective, which may justify rules on time limits for transfers and rules to ensure a degree of stability in player rosters. It will be for the referring court to determine whether the no-poach agreement is necessary and proportionate to achieve that objective. This guidance is, of course, only relevant if the referring court concludes that the agreement does not have an anticompetitive object
Comment
New(ish) developments in the Court’s approach to object restrictions
One of the most hotly debated points in competition law is what constitutes a restriction of competition by object. This ruling does not break with existing precedent, but it does develop the case-law in interesting ways.
First, the Court introduces a clarification as to what it means for an agreement to present a sufficient degree of harm to competition to conclude that it has an anticompetitive object: it must have “a manifest anticompetitive rationale”. In support, the Court cites Advocate General Emiliou’s Opinion in the case, who had referred to a “manifest anticompetitive economic rationale”, quoting, in turn, Advocate General Kokott’s Opinion in Generics UK, who stated that “the agreement and its context must reasonably clearly reveal the potential to harm competition”. It is inherent in the words “manifest” and “reasonably clearly” that the Court is only willing to accept that agreements fall into the category of “by object” restrictions if their potential to be anticompetitive is obvious.
Second, and relatedly, this does not however mean that the analysis of such agreements can only be “cursory”, and that if such a cursory look does not reveal a sufficient degree of harm to competition, the agreement is off the hook. To the contrary, the Court confirms that while the analysis of particularly harmful conduct can be limited to what is strictly necessary, for conduct that is not necessarily equally harmful to competition, a more in-depth analysis may be needed, which may still reveal that the conduct had an anticompetitive object. This is another echo of the pay-for-delay cases, where the Court also held in relation to the complex arrangements that formed part of the Servier case that the assessment of the object of such arrangements must be carried out “on the basis of objective considerations, where necessary as a result of a detailed analysis of that practice, its objectives and the economic and legal context of which it forms part”, and ultimately found that the arrangements had the object of restricting competition (see also our blog on Servier, here).
Third, the case offers a rare example of an agreement that the Court considers on the basis of its content manifestly to restrict a competitive parameter which plays an essential role in the relevant sector, but which nonetheless based on its context and objectives may not have an anticompetitive object. In this respect, the case can be seen as naturally following the Cartes Bancaires judgment, in which the Court found that the Commission had disregarded the plausible procompetitive rationale of the CB system, as well as ignoring the two-sided nature of the market.
Procompetitive objectives versus legitimate objectives neutral to competition
The famous trio of sports cases Superleague, Royal Antwerp and ISU had clarified that an object restriction could not be “saved” by a legitimate objective that is in the public interest (the Wouters exception or, specifically in sports, the Meca Medina exception). The Tondela case follows that approach as well.
However, it clarifies that the situation is different where the agreement pursues a pro-competitive objective. The Court bases this on the idea that in the analysis of an agreement’s objectives as part of the three factors that must be assessed to establish whether or not the agreement has an anticompetitive object (alongside content and legal and economic context), those objectives must be competition related. In other words, there is no room here for public interest arguments that are neutral to competition. If the agreement pursues procompetitive objectives alongside its anticompetitive aims, then this can mean that, overall, the agreement does not present a sufficient degree of harm so as to qualify as an object restriction. This is not necessarily new (arguably it was already established in Cartes Bancaires), but it does put in sharper contrast how to approach other objectives that the agreement pursues alongside any anticompetitive aims.
It also raises the question when a restrictive agreement can be said to also pursue procompetitive objectives.
In Diarra, FIFA’s rules on transfers were said by FIFA to have the objective of preserving the integrity, regularity and proper conduct of sporting competitions. However, this is treated by the Court as a legitimate-but-competition-neutral objective, not a procompetitive one.
In Tondela, on the other hand, the agreement did have a procompetitive objective. Essentially, this is described as the intention to maintain stability in clubs’ player rosters throughout the indefinite suspension period of the 2019/2020 season and, once it became possible, to enable the resumption of play in conditions guaranteeing the integrity thereof. The problem was that numerous player contracts were due to expire before the end of the extended season so that, in the absence of the no-poach agreement, the clubs would have been free to recruit players of their competitors during the final decisive weeks of the season. Clubs with greater financial means could have strengthened their squads and weakened those of their rivals, thereby irrevocably distorting the proper progress of the league.
Both cases are ultimately about preserving the integrity of the relevant sporting competitions. Of course, in Tondela there was an imminent threat to that integrity caused by the extension of the football season and the impact this would have on players’ contracts, whereas Diarra was about the general rules for football transfers. However, nowhere does the Court discuss that distinction or how it is determinative for establishing whether the concurrent objective that the agreement (also) pursued is procompetitive or legitimate-but-competition-neutral. For practitioners, it will be tricky in practice to draw a clear bright line between these two concepts.
On or offside?
That leads us, finally, to the question whether no-poach agreements are on or offside. Generally speaking, the judgment provides helpful guidance for the assessment of no-poach agreements under Article 101 TFEU. Such agreements can constitute restrictions by object, but a careful assessment is needed for such a conclusion to be valid. In particular, it is necessary to ascertain in each specific case to what extent the employees that fall in the scope of the agreement are a competitive parameter which plays an essential role in the relevant market. It will also be important to assess the specific context within which the agreement came about and whether that context gives rise to the need for some form of regulation of labour mobility (such as the stability of player rosters in this case). Finally, due regard must be given to any other objectives that the agreement pursued, and whether these were procompetitive in nature.
In this specific case, the Court has left the referring court in a difficult position where it will need to balance the anticompetitive objectives of the agreement against its procompetitive aims. Unlike in cases like Superleague, where the Court left little doubt as to its views, the referring court in Tondela is left with significant work still to be done. It will be interesting to see whether it considers that the Portuguese clubs stayed onside or just crossed the line.
The VAR review of the Portuguese no-poach agreement is complete, over now to the referring court to make the call: on or offside?
Leave a Reply