Newsletter
Suscríbete a nuestro Newsletter y entérate de las últimas novedades.
https://centrocompetencia.com/wp-content/themes/Ceco
volver
The essential facilities doctrine refers to infrastructure that is indispensable for the continuity of a business or for a company to be able to reach its consumers. Infrastructure is considered essential when its duplication is unfeasible or extremely difficult, due to the existence of physical, geographical, legal, or economic restrictions (EC, 2002).
A typical example of an essential facility is the national electricity distribution network, used by power generators to supply consumers. In this sense, electricity producers depend on access to the existing infrastructure to serve their users.
The essential facility concept is of economic nature, and it refers to situations in which a supplying company owns an input that is essential for the provision of a service. In general, the supplying company provides access services to companies that compete in the downstream market that is situated closer to the final consumer (Serra and González, 2010).
Refusal to grant access to an essential facility may be considered an abuse of a dominant position by the entity that controls it, especially when this obstacle hinders competition in the downstream market (EC, 2002).
The essential facilities concept is not legally enshrined in positive law. Rather, it found its origin and has evolved from European and north American caselaw. In this sense, cases related to the essential facilities doctrine belong to the more general case law category of refusal to deal.
The first case in which the essential facilities doctrine was applied ocurred in 1912, when a group of railroad companies prevented competing firms from offering transport services in the St. Louis region (224 U.S. 383).
The Supreme Court considered that this conduct was an attempt to monopolize the rail transport market and therefore ordered the defendants either to grant non-discriminatory access to third parties or else divest themselves of the essential infrastructure.
In 1973, the essential facilities doctrine was used again in a case related to denial of access to an electricity transmission system.
Years later, as a result of a dispute in the telecommunications industry (the MCI/AT&T case), what is now known as the “MCI test” was created. This test sets out four conditions under which an asset must be considered an essential facility (Serra and González, 2010):
Regarding the first condition, it is important to define the relevant market precisely in order to determine whether the company truly has monopoly power or not. Regarding the second condition, some consider it is probably the most difficult to prove. Furthermore, the third condition not only includes direct refusal to grant access, but also situations in which access is granted under abusive or discriminatory conditions. Lastly, the fourth condition features arguments linking access to the essential capacity with capacity limits or the deterioration of the services it provides.
The first time that the European Commission used the term “essential facility” was in its decision in the Sea Containers/Stena Sealink case in 1992. In that case, the complainant accused Sealink, a port operator that owned a ferry service, of assigning less favorable slots to companies competing with its own ferry service.
A second relevant milestone was the Oscar Bronner case, in which the Court of Justice of the European Union established a three-step test (“Bronner criteria”) to determine whether refusal to grant access does or does not constitute a violation of the law (Serra and González, 2010):
What is striking about this test is that it applies regardless of the size of the company seeking access, ruling out the possibility that the claimant might justify the impossibility of replicating the facility based on its relative share of the market.
In 2005, the essential facilities doctrine was discussed in the context of Article 82 of the Treaty of the European Union. In that discussion, five criteria were established for considering that a refusal to grant access is abusive (Serra and González, 2010):
The main distinction in these criteria is that they are based on the concept of dominance rather than monopoly, recognizing that a refusal of access may have effects in other markets.
In Chile, in 2006, the Tribunal de la Libre Competencia (TDLC) applied the essential facilities doctrine in the Voissnet v. Telefónica (Ruling TDLC No. 97/2010) case, sanctioning Compañía de Telecomunicaciones de Chile S.A. for having refused to allow its wholesale clients to use its broadband platform, with the aim of protecting its own business. In response, the court ordered the company to amend the restrictive clauses (Serra and González, 2010).
Argentina has also seen cases involving the essential facilities doctrine. A notable example involved a vertical merger where an international shipping company sought to acquire the concessions for six terminals in Puerto Nuevo (Serra and González, 2010).
The first criterion for defining whether a facility is essential consists in defining the final market, which corresponds to the market that companies can access by means of the essential facility. The definition of the final market must include all substitutes that provide services equivalent to those provided by the alleged essential facility.
The second criterion consists in assessing how feasible it is for a new market entrant to duplicate the essential facility. Duplication does not mean creating an exact replica of the essential facility, but rather reproducing the services provided by it.
In some cases, a new facility may have disadvantages when compared to the essential facility, due to differences in cost or quality. However, case law tends to consider that such differences are insufficient to establish that an input should be considered essential.
Duplication of an essential facility must be profitable for the company undertaking it. Essential facilities are usually associated with high investment costs that not all companies can bear. For example, the profitability of a new facility may depend on supplying a minimum percentage of the market (“entry threshold”). If the company interested in building the new facility is unable to serve that portion of the market, construction of the facility would not be economically viable.
Another case is that in which a third party wants access to an essential facility that has already been duplicated (for example, two ports or two communication platforms serving the same area). One might think that if duplication is possible, then the facility should not be considered essential. Nonetheless, if duplication is not profitable for the third company, it is not clear whether access to the input should be mandatory for one or both existing facilities.
The foregoing suggests that the number of facilities serves as a proxy of the degree of competition in the provision of essential facilities. Thus, the more facilities there are, the less necessary it would be to regulate their respective access conditions.
There are considerations that favor regulating access to an essential facility. One of these is the divergence between the social cost and the private cost of duplicating an essential facility, due to the presence of externalities. In general, the construction of new essential facilities tends to face more opposition from residents than the existing facility does. For example, construction of the first telecommunications transmission tower may be well received by a community, but the same might not be the case for the second or third tower. Thus, the new essential facility may face higher entry costs than the existing facility, creating an entry barrier that grants market power to those who have access to the existing infrastructure.
There are legitimate commercial reasons that could make it very costly for the owner of an essential facility to grant third parties access to its infrastructure. Some of these reasons are technical incompatibility, capacity limitations, and congestion costs.
In theory, the general rule is that it would be advisable to grant access to an essential facility insofar as doing so increases social welfare. However, in practice, regulatory authorities do not have all the necessary information —such as the costs and demand faced by market participants— to quantify the effect that granting access would have on social welfare.
When duplication of an essential facility is not economically viable, introducing competition in the activities that use the service provided by the facility could, in the short term, generate improvements for consumers in the downstream market.
Nonetheless, if access is mandated by the regulatory authority, its implementation and regulation would require an adequate framework. In this sense, day-to-day control of access would be costly for the regulatory body.
On the other hand, some authors argue that granting access to essential facilities could discourage investment in such facilities (Hausman, 1999). This could occur in cases in which access is offered at the ex post average cost of maintaining the infrastructure, a tariff which does not take into account the risks assumed by the facility’s owners at the time of investment.
Determining whether it is desirable to make agents offer access to an essential facility depends, on the one hand, on how mature the market in which it operates is; and, on the other hand, on the origin of the essential facility. Below are four cases that take these criteria into account.
When the essential facility is built by the government, it must decide whether to privatize the company as an integrated monopoly or to sell separately the essential facility component and other potentially competitive components.
In these cases, it is of utmost importance that the government determine the regulatory framework that will govern the facility before privatizing the infrastructure: if the government wishes to introduce competition, it must establish the conditions of access to the facility. Moreover, access prices are also determined based on the regulatory framework. Alternatively, the government could specify service provision standards, making the facility’s owner responsible for making the investments necessary to comply with those standards (Serra and González, 2010).
The government might be interested in auctioning the facility with open access. As in the case of privatization, the concession holder will know in advance the terms under which access prices are set.
One way to regulate price setting is to design the auction so that the facility is awarded to the bidder that can offer the lowest access price. In this sense, companies compete for the essential facility, relieving the authority of part of the regulatory oversight burden. Once the concession period expires, the authority may repeat the same procedure to again delegate operation of the essential facility (Serra and González, 2010).
There are cases in which an essential facility, instead of being acquired or granted, is created by a company, thereby establishing an unregulated monopoly. In such cases, regulating access is more complicated because access prices must incorporate both current production costs and the risk borne by the company when it invested in developing the facility.
The difficulties that arise in setting access prices for an essential facility could be anticipated. In particular, companies planning to develop an essential facility could submit to the competition authority the pricing scheme they intend to apply, requesting that it be approved in advance.
In addition, the company could request that the authority grant it a period during which the facility is not subject to access regulations. This could reduce uncertainty, thereby encouraging investment in socially desirable facilities (Serra and González, 2010).
Joint ownership of an essential facility can be seen as evidence that the infrastructure in question is indeed essential, as it reveals that joint ownership generates economies of scale that make its provision more efficient (Serra and González, 2010).
In the United States, courts tend to apply a less stringent test in declaring a facility essential when refusal to participate comes from a group of companies rather than from a single firm.
One of the main features of the digital economy is that it is not based on the physical infrastructure that gave rise to the essential facilities doctrine. Now, access to data or the way in which a platform ranks commercial users (“ranking”) are aspects that are essential to competition in these markets. From this perspective, development of digital markets could require rethinking the essential facilities doctrine.
In the European Union, the essential facilities doctrine has not been applied in competition law matters in recent years. However, it is often mentioned in political discussion, as it is viewed as a tool that could promote competition in digital markets, where “technology giants” act as “gatekeepers” for companies and consumers (for example, see our note on the European Union Digital Markets Act here).
Under the Digital Markets Act of the European Union (see details in our CeCo note here), a company is considered a “gatekeeper” for one or more platform services when the following conditions are met: 1) it operates a core platform service that serves as a gateway for business users to reach end users; 2) it has a significant impact on the internal market; 3) it enjoys an entrenched and durable position in its operations, and it is likely that these conditions will be maintained in the near future (Büchel & Rusche, 2021).
The following section presents two cases in which the remedies imposed by the European Commission are quite similar to decisions in earlier cases in which the essential facilities doctrine was applied. Academic Inge Graef argues that, instead of applying this doctrine, the Commission extends the scope of other abuses in order to impose remedies based on other theories of harm. In doing so, the authority bypasses the more stringent requirements of the essential facilities doctrine (Graef, 2019).
According to Graef, the essential facilities doctrine could be more appropriate than other theories of harm in the context of digital markets. This is due to market failures that are inherent in this type of markets: network effects (for example, combined use of consumer data) and switching costs (for example, the costs incurred by users when changing platforms).
According to this academic, the existence of these market failures, which are specific to digital markets, allows companies to artificially increase their dominant position. For this reason, the use of the essential facilities doctrine could better align with the underlying economic incentives in these markets.
In June 2017, the European Commission imposed a fine of 2.4 billion euros on Google for illegally favoring its own price comparison service (“Google Shopping”) in the online search market (Abellán, 2017; AT.39740). The Commission found that the positioning of this service was abusive because it diverted traffic from competing services and could have anticompetitive effects in national markets. In particular, Google is said to have leveraged its dominant position from one market to another — a related but separate “downstream” market (see details of the case in our CeCo note here).
Although Google defended the relevance of using the “Bronner” criteria, the European Commission held that, in this case, its application was irrelevant. This, due to two reasons: first, Google did not refuse third-party access to its search market; and second, termination of the infringement did not entail requiring the dominant company to transfer an asset or enter into an agreement with a third party.
The remedy imposed by the Commission to bring the infringement to an end was to require Google to comply with the principle of equal treatment. Thus, the positioning and display of “Google Shopping” on the Google search engine should be subject to the same algorithm applied to its competitors’ services. According to academic Inge Graef, in this case, application of the essential facilities doctrine would have led to a similar result. In fact, this case can also be interpreted as a refusal to grant access to the prominent placements on a platform (Graef, 2019). When the Commission frames the case as a self-preferencing practice by the dominant company, it overlooks the fact that access to the Google platform is essential for price comparison services to be able to compete.
In July 2018, the European Commission again sanctioned Google, imposing a fine of 4.3 billion euros (AT.40099). In this case, the authority alleged that Google had used Android to consolidate its dominance in the online search market.
Google was accused of imposing three abusive contractual restrictions on Android device manufacturers. One of these was requiring manufacturers to preinstall the “Google Search” application and the “Chrome” browser as a condition for granting licenses for “Google Play Store” (Graef, 2019). The European Commission considered that this restriction amounted to a strategy by Google to tie or bundle products. According to the authority, since “Google Play Store” is an indispensable application for users (it is used to download other applications), Google used it as a “tying product” to position Google Search and Google Chrome in the online search market.
According to experts, in this case, Google’s infringement is more closely related to preventing devices from accessing the “Google Play Store” application than to the tying practice at issue. In this sense, the potential unlawful conduct would be more akin to refusal to deal than to an unlawful tying arrangement.
