Competition law systems typically draw a sharp distinction between concerted and independent conduct, with the former being viewed as more fraught with antitrust risk than unilateral conduct. The US and EU systems thus contain broad prohibitions of agreements, or other collusive arrangements, between independent entities that restrain trade or restrict competition (Section 1 Sherman Act and Article 101 TFEU respectively). Neither provision extends, however, to wholly unilateral conduct, or the internal workings, of a single firm, for example arrangements between an employer and employee or other persons within a firm, or between a parent and its wholly-owned subsidiary. In the US, the Supreme Court has held that since parents and their wholly owned subsidiaries always have a unity of purpose, they are incapable of conspiring with one another in an antitrust sense (the activity is viewed as that of a single enterprise, see Copperweld Corp v Independence Tube Corp (1984)). In the EU, where a subsidiary has no economic independence or freedom to determine its course of action on the market, but carries out the instructions issued by its controlling parent, the entities are treated as part of a single ‘economic unit’ and agreements between them fall outside the scope of Article 101 (see Viho Europe v Commission (1996)).
Similarly, it has been held that arrangements between a principal and agent may fall outside of Section 1 and Article 101. The US Supreme Court has held that a manufacturer that simply consigns its products through a genuine agent, disposing of its goods directly to customers and establishing the price at which it sells the product, acts unilaterally (see United States v General Electric (1926)). Consequently, the supplier is able to dictate the price at which the agent sells, and the territory into which and customers to whom it can sell, without creating a conspiracy. In this respect, the agent acts as its representative and an integral part of the principal’s business rather than as a conspirator. Although the doctrine has had less prominence and importance since Sylvania (1977) and Leegin (2007), and their reversal of the per se prohibition of vertical territorial and price restraints (so requiring them to be appraised under a fuller rule of reason analysis), it still applies (Valuepest (4th Cir 2009)).
In the EU, the close economic links existing in an agency relationship may also mean that an agent is so closely interrelated with its principal that the relationship is characterised by ‘economic unity’. Where an agent operates as an integral part of its principal’s undertaking, agreements between them relating to the market on which the agent offers the principal’s goods or services to potential customers, fall outside of Article 101(1) (see Confederación Española de Empresarios de Estaciones de Servicio v Compañia Española de Petróleos (CEES) (2006) and CEPSA Estaciones de Servicio SA v LV Tobar e Hojos SL (CEPSA) (2008)).
In addition to impacting on the substantive reach of Section 1 and Article 101, the doctrine also enables liability and responsibility for an antitrust infringement by an agent to be attributed to its principal in certain circumstances; in the US where the agent(s) acts within the scope of apparent authority (US Supreme Court, American Society of Mechanical Engineers Inc v Hydrolevel Corp (1982)) and in the EU where the agent forms part of the same economic unit as the parent (EU General Court, Minoan Lines v Commission (2003)).
Given the significant consequences that flow from the doctrine, a crucial issue is how to identify genuine agency agreements and to distinguish them from ones between independent entrepreneurs. It is clear in both the US and EU that it is not the parties’ characterisation of the arrangement (the form) which is determinative (so sham agency arrangements are not able to evade application of the antitrust rules). Rather, in the US, it is a question to be determined by the trier of fact taking account of a range of factors, including, whether the supplier retains title to, and control of, the products, whether the supplier retains risk of loss, who insures the products and whether consignment is used as a veneer for a vast anticompetitive scheme (see Simpson v Union Oil Co (1964)). Likewise, in the EU, it is the ‘economic reality’ that is decisive. Although some early cases suggest agents must act as ‘auxiliary organs’ of the principal (and constitute independent entities if acting for a number of principals, see VZW Vereniging van Vlaamse Reisbureaus v VZW Sociale Dienst van de Plaatselijke en Gewestelijke Overheidsdiensten (1987)), modern cases focus on the question of whether the agent bears more than a negligible proportion of the financial and commercial risks linked to sales of goods to third parties (CEES and CEPSA).
The question of how broadly the agency concept extends is important to the question of whether arrangements in digital markets between suppliers and platforms (such as Amazon, or e-Bay,) are caught by Section 1 or Article 101. Although several publishers agreed in settlement arrangements to terminate their distribution agreements for electronic books using an agency, as opposed to a wholesale or reseller, model, with Apple (and other retailers) following European Commission and DOJ investigations (see e-books (2013) and United States v Apple (2d Cir, 2015), there is no explicit ruling in either the US or EU on how such distribution agreements are to be treated. The core question arising in each of these cases was whether the publishers engaged with Apple, in a horizontal concerted practice or conspiracy to raise retail prices for e-books. The question of whether the vertical agreements relating to sales on online platforms would fall within the scope of Article 101 or Section 1 did not therefore arise. Nonetheless, applying the principles set out in the cases, it seems possible that an arrangement between a supplier and a platform could constitute a ‘genuine’ agency one falling outside the reach of Section 1 or Article 101 where, for example, ownership of the products/services is not passed to the platform (the platform does not buy products from suppliers for resale), where contracts concluded are formed between the supplier and customer, and where the platform does not bear any of the risks related to the sale or provision of the goods or services, but receives a commission or remuneration for concluded contracts. If this is the case, some have argued that the antitrust laws may have a ‘platform gap’, immunising potentially restrictive distribution arrangements from scrutiny under these provisions. This has led some to call of a modification or reinterpretation of agency rules.