Newsletter
Suscríbete a nuestro Newsletter y entérate de las últimas novedades.
https://centrocompetencia.com/wp-content/themes/Ceco
volver
Various legal systems, when sanctioning collusion, state that this conduct may occur either through an agreement or a concerted practice.
Thus, Article 101 of the Treaty on the Functioning of the European Union prohibits all agreements between undertakings, decisions by associations of undertakings and concerted practices which may affect trade between Member States and which have as their object or effect the prevention, restriction or distortion of competition within the internal market.
Similarly, Chilean law, in letter (a) of Article 3(2) of Decree Law 211 (DL 211), states that “Agreements or concerted practices involving competitors (…)” that affect certain variables are sanctioned.
Likewise, Colombian regulation, in Decree 2153 of 1992, Article 45, defines an agreement as “[a]ny contract, arrangement, understanding, concerted or consciously parallel practice between two or more companies.”
In turn, Article 11 of Ecuador’s Organic Law on Regulation and Control of Market Power includes among the prohibited practices “any agreement, collective decision or recommendation, or concerted or consciously parallel practice, and in general all acts or conducts carried out by two or more economic operators, expressed in any form.”
Finally, the United Kingdom’s Competition Act 1998 also covers concerted practices, stating in Section 2 that, under certain conditions, it prohibits “agreements between undertakings, decisions by associations of undertakings or concerted practices which have as their object or effect the prevention, restriction or distortion of competition.”
This glossary will jointly analyze the concepts of agreement and concerted practice, as doing so clarifies how they are distinct. The analysis will take European legislation as its point of reference, as this is where the doctrinal development on the distinction is most advanced. However, the definitions presented here may be applicable in other jurisdictions as well.
This distinction is relevant, first, because competition law sanctions are justified by their ability to prevent or deter conduct that reduces welfare (Grunberg & Montt, 2017, p. 310). But deterrence assumes that the subjects of a prohibition or mandate know what is prohibited or mandated. This in turn assumes that, in the context antitrust law, economic agents know what constitutes an agreement or a concerted practice.
Second, properly distinguishing between these two concepts is also important because, in Chile, Article 62 of DL 211 provides for criminal sanctions for agreements but not for concerted practices. Therefore, gaining greater clarity regarding the content of the concepts of agreement and concerted practice will also provide greater clarity about what is not included in the criminal sanction for collusion.
That said, it is difficult to clearly draw this distinction. The term “concerted practice” is a legal one that does not exist in natural language. This is not the case for all legal concepts. For example, in the case of homicide, the term is used in non-legal social contexts, and its legal meaning is drawn from that context. The same applies to the term “agreement,” which exists in everyday social life. But this is not the case for “concerted practice.”
Given this situation, it becomes necessary to refer to case law. However, relying solely on case law carries certain risks, as an unsystematic explanation built through casuistry may be driven by specific facts. As case diversity increases, the meaning of the concept loses conceptual unity (Black, 2005a, p. 346). Thus, this glossary attempts to show which doctrinal concepts have developed from jurisprudential practice.
There is a consensus that concerted practices must be distinguished from conscious parallelism (Valdés, 2016, p. 7). The latter occurs when market actors, by observing their competitors’ behavior, adopt similar conduct. Parallelism arises when competitors engage in the same conduct, and it becomes conscious when each competitor acts with the awareness that others will behave the same. However, in this scenario, each actor retains independence (because there is no communication), and the conduct does not constitute a collusive agreement.
Therefore, a collusive act, whether an agreement or a concerted practice, necessarily implies the cessation of independence among actors. This occurs through either an agreement or a concerted practice. The following sections will explore different theses on how these two forms of conduct are distinguished.
According to this view, the difference between an agreement and a concerted practice lies in the fact that, while the former can be proven in court by direct evidence of the agreement, the latter can only be established through circumstantial evidence. Under this interpretation, the concept of concerted practice is synonymous with tacit agreement (in the sense that these agreements can only be discovered indirectly, through indicia, not in the economic sense of tacit collusion). Let’s consider some examples. Under this logic, an agreement could be proven by producing a chat in which two managers agree on the price of a product. By contrast, a concerted practice could be proven, at least in part, by evidence such as images showing that, the day before a price increase in a certain industry, all the managers had dinner together at a restaurant (this would not be direct evidence of the agreement, but evidence that could help infer one).
Based on this, some argue in Chile that criminal sanctions for collusion apply only to express agreements, while administrative sanctions (i.e., fines, but not imprisonment) apply to both express and tacit agreements —that is, concerted practices (Walker, 2020, p. 143). This is because the criminal offense of collusion only refers to agreements, not concerted practices. Thus, Juppet and Morales argue that regarding the crime of collusion the use of circumstantial evidence would be incompatible with the requirements of criminal proceedings (Morales & Juppet, 2018, Chapter II, Section 5).
Similarly, Hernández states that the distinction is not conceptually relevant, but relevant in practice (Hernández, 2016, p. 20), since the figure of concerted practices provides evidentiary advantages (Hernández, 2016, p. 21). He explains that this figure aims to address the difficulties in proving an agreement, allowing both clear-cut agreements, cases where an agreement is presumed but lacks proof, and scenarios involving behavior that goes beyond simple parallel conduct (Hernández, 2016, p. 21). Therefore, a form of consensus or coordination would be required, but it need not be explicit, it may be inferred (Hernández, 2016, p. 21).
In the case of Chile, this thesis presents some difficulties. Above all, any attempt to base the distinction in this way is misguided because it is legislatively redundant: there are already rules that cover these evidentiary considerations. Regarding the admissibility of circumstantial evidence, Article 22 of DL 211 establishes that the judge must assess the evidence of an agreement based on “sound reasoning” (sana crítica), allowing both direct and indirect evidence (Araya, 2015, p. 233). Moreover, the same article mentions as admissible any indicia or information that, in the court’s view, may establish relevant facts (see further critique in Peralta, 2023, pp. 147–149). Additionally, in more general terms the following is true: what is proven and the means used to prove it are different things. The evidentiary thesis confuses these two elements.
According to this position, the difference between agreement and concerted practice is not evidentiary but substantive: they represent two different types of conduct. In this line, Grunberg refers to the definition of concerted practice by the Court of Justice of the European Union in the case ICI v. Commission (§64) and states that concerted practices are a form of coordination between companies which, without reaching the stage of an actual agreement, knowingly substitutes the risks of competition with practical cooperation between them (Grunberg, 2017, p. 21).
Defining concerted practices as a residual form of collusion is a constant theme in the specialized literature. It is stated that concerted practices are a type of collusion designed to capture situations in which the concept of agreement, despite its breadth, cannot be applied (Araya, 2015, p. 246; del Mar, 2011, p. 112; Frenz, 2016, p. 243). According to this definition, concerted practices are a form of collusion characterized by a lack of independence among competitors and the presence of “something less” than an agreement.
This view is vague, as it does not allow market actors to know in advance and with sufficient clarity what is punishable. Thus, the real function of the concept of concerted practice under this interpretation is to operate in those cases where, despite the broad concept of agreement, it cannot be used. In this way, the concept of concerted practice would be something more than mere parallel behavior and something less than an agreement. Nonetheless, even if caselaw could complement this interpretation and provide examples of when we are dealing with a concerted practice, we still lack a concept that can be applied without referring to specific instances already determined to fall under that category (Peralta, 2023, pp. 149–154).
It is therefore useful to turn to Oliver Black’s formula, which seeks to be more precise (Black, 2005b, p. 155):
Let X and Y be persons, and let Ax and Ay be their respective actions. In that case, X and Y act jointly when:
Thus, whereas in an agreement it would suffice for there to be offer and acceptance, in a concerted practice there must be joint action based on mutual reliance and dependence, with the parties involved in a concerted practice sharing a common objective, which in turn depends on communication between them.
From this, a possible hypothesis about the relationship between an agreement and a concerted practice is that while the former is a mere meeting of wills, the latter is the implementation of that meeting of wills. Seen from this angle —and considering agreements and concerted practices beyond positive law, as mere social phenomena, is an executed agreement different from a concerted practice? Black believes it is. On the one hand, there can be a concerted practice without an agreement, since joint action may occur without X committing to do Ax or Y committing to do Ay. In fact, the parties might even explicitly state that they are not committing to anything, yet still rely on each other, accepting that risk (Black, 2005b, p. 155). On the other hand, not all action carried out pursuant to an agreement constitutes a concerted practice. For example, it is not true that every execution of an agreement implies mutual trust and dependence, as it may be that X does not trust Y to do their part—perhaps because they suspect that Y, for whatever reason, will not be able to fulfill the commitment (Black, 2005b, p. 155)
According to the thesis just mentioned, the critical point when evaluating whether parallel conduct also constitutes a concerted practice is whether the mutual dependence and reliance, belief, and knowledge accompanying the parties’ actions were acquired at least in part through communication between rivals (Page, 2007, p. 427) (see also CeCo Glossary “Information Exchange”). In this respect, communication takes different forms, including: indirect communications (mediated by a third party), unspecified communication (where someone makes a general announcement), implicit communication (where the listener must make complex inferences to understand what the speaker means), or non-verbal communication (such as a wink) (Page, 2007, p. 427) Thus, the central idea is that a concerted practice involves communication followed by parallel action (Page, 2007, p. 428). This resolves one of the most challenging issues in modern antitrust: identifying the conditions under which parallel behavior counts as concerted action (Page, 2007, p. 407).
Therefore, the essential elements are: a) uniform, interdependent behavior; b) prior communications that transmit the rival’s intention, without reaching reciprocal assurances; and c) a causal relationship between (a) and (b) (Page, 2017, p. 608). Consequently, the prosecuting authority must argue that the defendant’s conduct suggests coordination through something more than mere interdependence, such that the accusation includes some form of communication between competitors (Page, 2017, p. 602). Thus, this category allows courts to sanction not only agreements but also certain information exchanges that result in interdependent conduct.
One criticism of Black’s approach is that including the communication of intentions and mutual dependence and reliance in the definition of concerted practices does not clarify what type of communication satisfies the definition, nor what kind of evidence would be sufficient to prove the relevant communication (del Mar, 2011, p. 118). That is why it is useful to turn to Page, who offers a view on what type of evidence would be sufficient. The two critical variables are: the recipients and the content of the communications (Page, 2017, p. 611). Thus, courts are more likely to classify conduct as a concerted practice when the communication is private (only among competitors) rather than public, and when it refers to future competitive conditions rather than present ones (Page, 2017, p. 611). Furthermore, communications made by decision-makers carry more evidentiary weight than those made by other company agents (Page, 2017, p. 611).
In light of the above, the crucial role of communication in enabling interdependent conduct among market players becomes even clearer. One analytical tool used to examine this are facilitating practices. These are actions taken by competitors to make anticompetitive coordination easier, without requiring an agreement (Hay, 2006, p. 896). They are especially relevant in markets with structural difficulties that hinder oligopolistic interdependence (Zheng, 2019, p. 410). This could be the case, for example, of public announcements of future pricing intentions, meant to invite the counterpart to enter an interdependent oligopolistic relationship (Hay, 2006, p. 900), or mechanisms for sharing prices (Zheng, 2019, p. 435), such as certain industry association meetings (for example, something similar happened in the “Chicken case,” as detailed in the note “Especial Caso Pollos: El papel de la Asociación Gremial ‘APA’”).
Returning to the core idea, concerted practices primarily involve transmitting information indirectly and without reaching an agreement, but in a way that facilitates oligopolistic interdependence. According to Hay, the goal of the concerted practice figure is to fill the enforcement gap between anticompetitive agreements and pure oligopolistic interdependence (Hay, 2006, p. 896).
On this point, it is worth emphasizing that one must distinguish carefully between information that companies may legitimately exchange and competitively sensitive information. This issue arises in the context of data collaboratives, which are an emerging form of public–private partnership. In these collaborations, actors from different sectors exchange and analyze data or provide data science expertise to create new public value and generate new insights. Nevertheless, a key issue here is avoiding the dissemination of competitively sensitive information, as this could constitute a concerted practice (see CeCo note “Data Collaboratives and the Ban on Concerted Practices in the EU: Compliance Challenges”).
All of the above means that the concept of concerted practice also covers cases in which there is evidence of ongoing communication, even if the specific content of those communications is unclear. In such cases, the existence of communication may be enough to infer that there was a concerted practice (Page, 2017, pp. 610–611). This is because, under this concept, there is no need for specific communications to be proven in order to reach the conviction that there was a concerted practice as long as there is circumstantial evidence allowing one to infer communication (Page, 2007, p. 409).
Even more, the category of concerted practices could be used to sanction those who have not made any verbal statements of intent, i.e., who have not communicated to competitors how they will act. According to Page, this could be the case of a competitor who silently attends several meetings where others declare their intention to act in a certain way (e.g., by setting a certain price), and later all of them—including the silent party—act accordingly (Page, 2007, p. 445). That is why the direct exchange of prices between rival companies is considered illegal, even if only one firm provides such information to another (Tapia, 2009, p. 54).
Finally, one area where the role of communication in anticompetitive conduct may gain new relevance is algorithmic collusion. This refers to collusion implemented by algorithms, either with or without human intervention (see CeCo note “OECD: Algorithmic Competition”). If there is human intervention, we are dealing with an algorithm designed to facilitate the implementation of a previously adopted collusive agreement. For example: three competitors who agree to collude adopt an algorithm to automatically set product prices based on a real-time competitor price database. The algorithm can immediately detect any price deviation and automatically adjust up or down—thus “stabilizing” the cartel.
On the other hand, there may be a case where an algorithm functions as a hub (in a hub-and-spoke scheme; see CeCo Glossary), enabling the exchange of strategic information between competitors. For example: three competitors, who do not reach an agreement among themselves, each independently hire a third-party price-setting software service and provide it with sensitive, non-public commercial information as input (e.g., production forecasts or pricing plans). The algorithm then considers the information from all three and generates an output: a pricing strategy that is the same for all three companies, thus maximizing their profits.
In principle, this scenario would not constitute collusion if the market participants (the spokes) independently chose to hire the algorithmic hub. However, if the hub’s main objective is to maximize its clients’ revenues, it could align the strategies of all its clients. That is, the hub may find it more profitable to optimize profits for all clients collectively, not individually (i.e., benefiting everyone without harming anyone). If this happens, the end result may be a supra-competitive price (in the extreme, a monopoly price). Moreover, the more companies use the hub’s services, the more data it collects about competitors, creating a feedback loop that refines its optimization strategy and consolidates its position (see CeCo note “Ezrachi and Stucke: Tacit Algorithmic Collusion in Secondary Markets”). In this case, there may not be an agreement between competitors, but there is a practice of information sharing, even if it is done through the hub of the collusive scheme (according to the U.S. Department of Justice, this is what occurred in the RealPage case—see CeCo note “DOJ Lawsuit Against RealPage: Price Algorithms in the Housing Market”).
Third, perhaps the most interesting case of algorithmic collusion is tacit algorithmic collusion, which occurs when a complex algorithm, through its own learning process, reaches a collusive strategy without any prior agreement between competitors and without any information exchange. To date, there have been no documented cases where algorithms have autonomously generated collusive conduct. Nevertheless, given the growing use of price-setting algorithms by firms, the OECD Report on algorithmic competition emphasizes the importance of identifying facilitating practices for tacit algorithmic collusion. Some of these practices —potential plus factors— may include the exchange of datasets between competitors or the communication of decision-making parameters that algorithms could use (Capobianco & Westrik, 2023, pp. 15–16). Thus, strictly speaking, there may be no agreement, but there would be an exchange of sensitive information that could be sanctioned under the concept of concerted practice.
The question of whether agreements and concerted practices are synonymous terms was debated in the “Supermarkets case”. In this case, both the Chilean Competition Tribunal (TDLC) and the Supreme Court sanctioned supermarket chains SMU, Walmart, and Cencosud for setting minimum resale prices for fresh chicken. The scheme was carried out through the fresh chicken suppliers: the supermarket chains approached them to reinstate the minimum resale price whenever a deviation by another chain was detected.
In its ruling, the TDLC stated that the distinction between agreements and concerted practices was outdated (¶31), and that current case law focuses on covering all forms of collusion. Thus, it chose to use the term “agreement” to cover both concepts. For its part, the Supreme Court stated in its judgment that regardless of the technical label, the accusation concerns the fact that the companies’ conduct was not determined by free price-setting, but rather by a common will (C.8).
That said, this must be analyzed in light of the fact that, in its accusation, the National Economic Prosecutor’s Office (FNE) charged the supermarket chains with having participated indistinctly in “agreements and concerted practices.” For this reason, during the litigation, some of the chains objected to the interchangeable use of these terms in the FNE’s accusation, arguing that although both share the subjective element, they differ in identity due to the forms in which they are manifested (TDLC Judgment, Expository Section 4.11).
In this regard, the supermarkets case can be analyzed as a hub-and-spoke scenario involving concerted practices. As described, the conduct consisted in maintaining a minimum resale price for fresh chicken, based on the understanding that the other chains would act similarly, and that any deviation might be “punished” (C. 104). Accordingly, the TDLC noted that it was necessary to determine “whether we are facing a collusive agreement known as ‘hub & spoke,’ that is, a situation where one firm (the ‘hub’) maintains several separate vertical agreements with firms that are ostensibly competitors (the ‘spokes’), but which in reality maintain an implicit agreement among themselves” (C. 107). On this point, the TDLC emphasized that the key is to determine when a merely vertical agreement (i.e., between firms at different levels of a distribution chain), which may be legal or unlawful depending on the circumstances, becomes a horizontal agreement (i.e., between competitors) of an illegal nature. To do this, additional elements must be identified to infer horizontal coordination (C. 105).
That is, it is necessary to establish how the horizontal communication required to sanction such conduct occurred. On this point, the judgment concluded that “given the nature of the agreement, the accused conduct is inferred from the parties’ actions, which were carried out through communications between each of them and the suppliers, mainly via email” (C. 168). Thus, the TDLC found that “the three chains adhered to a behavioral rule consisting of a vertical restriction to sell fresh chicken at a minimum price in their various stores. Furthermore, this adherence was conditional on the others doing the same, which is particularly evident from the fact that each supermarket demanded that suppliers ensure the rule was followed by its competitors” (C. 93).
Therefore, returning to the “structural thesis” already discussed in this glossary (section 1.3), this case could be understood as one in which, although there was no agreement, there was a concerted practice. This is because the supermarkets acted interdependently, and this conduct was explained by communication between them, vertically mediated by the chicken suppliers.
In 2008, the European Commission fined producers of paraffin wax (and related products), which are used to manufacture candles, waxed paper, paper cups, among other items. The companies involved had engaged in anticompetitive arrangements over a 13-year period. The agreement involved price fixing, market allocation by geography, and customer allocation.
The case began with a leniency application by Shell and led to dawn raids and subsequent leniency applications by other companies. As a result, the Commission imposed fines totaling €676 million.
The Commission concluded that the proven facts formed part of a single, complex and continuous agreement, punishable under what was then Article 81 of the Treaty on the Functioning of the European Union. Regarding how this agreement operated, it was disguised as “Technical Meetings” regularly attended by employees of most producers. Evidence showed that during these meetings, the producers fixed the prices of paraffin waxes and exchanged commercially sensitive information.
On this information exchange, the Commission stated in its final decision that “[e]ven if not all contacts amounted to agreements in the sense of Article 81(1) EC, it is clear that certain discussions which fell short of constituting/leading to an agreement can still be characterised as a concerted practice. Three undertakings involved in the present case, […], have indicated that usually, Sasol would instigate the discussions about prices, but then prices and pricing strategies were discussed by all the attendees. Even if such […] would not always lead to formal agreements, it would allow the participants to the meetings to understand each others’ intentions in terms of commercial policy and therefore reduced significantly the uncertainty on future market conditions. Given the purpose of the meetings and the way they were conducted, this can be considered as a concerted practice by object. As explained at recital (204) such practice falls within the scope of application of Article 81(1) EC. In any event, the exchange of pricing letters that were sent after the meetings by the various participants at the Technical Meetings to their competitors constitutes coordinated market behaviour.” (Paragraph 239)
Black, O. (2005b). The Conceptual Foundations of antitrust. Cambridge University Press.
Frenz, W. (2016). Handbook of EU Competition Law (C. R. Aird, Trans.). Springer.
Hay, G. (2006). Horizontal Agreements: Concept and Proof. The Antitrust Bulletin, 51(4), 877–914.
Hernández, H. (2016). Informe en Derecho C-304-2016.
Page, W. (2017). Tacit Agreement Under Section 1 of the Sherman Act. Antitrust Law Journal, 593–639.
Zheng, W. (2019). A Knowledge Theory of Tacit Agreement. Harvard Business Law Review, 9, 399–438.
