The European Antitrust Review 2014 • Section 1: EU substantive areas
Non-Pricing Conduct by Dominant Firms in the EU: The Case of Exclusive Dealing
After years of focus on pricing conduct by dominant firms under EU law, non-pricing conduct is receiving the attention it deserves. The Google investigation, the many issues surrounding standard essential patents, and refusals to deal are now on the front burner. They promise to move the law forward in this important area.
It is hardly surprising that pricing conduct has received so much attention. Modern thinking co-existed uneasily with a long line of often incompatible judicial precedent. While enforcers began considering whether a dominant firm’s pricing precluded equally efficient rivals from effectively competing, a sometimes complex but necessary economic assessment, judicial precedent placed at least as much value on legal certainty. Yet legal certainty, though unquestionably valuable in itself, is not without cost, particularly when it threatens to chill the discounting that competition laws are meant to encourage. Through recent enforcement actions, and especially its published Guidance on Article 102, the European Commission moved the analysis from the expensive comfort of virtually per se rules toward an economic assessment more consistent with the goals of competition law.
Fortunately, much of the thinking that developed around pricing conduct is applicable to non-pricing conduct, where the stakes can be just as high. While discouraging discounting is an obvious and immediate harm, the same is true for other pro-competitive conduct in this age of innovation rivalry. The cost of ‘false positives’ is high.
This article applies current thinking to exclusive dealing, a common, often beneficial and familiar practice from which fundamental principles can be extrapolated to non-pricing conduct generally. Our focus is on situations where buyers agree to buy all or almost all of their requirements from a single seller (‘single branding’, in EC parlance). Regardless of whether it is assessed as a vertical agreement under article 101 or an abuse of dominance under article 102, exclusive dealing requires the same distinction between anti-competitive foreclosure and competition on the merits that continues to propel development of EU competition law. Modern precedents are scarce, but the recent judgment from the EFTA Court in Norway Post provides useful insight and an opportunity for the law to develop in this key area.
Norway Post is the national postal operator in Norway and a leading provider of business-to-consumer (B-to-C) parcels. In 2000 and 2001, the company entered into agreements with major retail chains NorgesGruppen (grocery stores and kiosks), Shell (petrol stations), COOP (grocery stores) and ICA (grocery stores) for the exclusive right to establish a Post-in-Shop (PiS) network for a range of postal and financial services. The agreement with NorgesGruppen and Shell provided that they would be Norway Post’s preferred partner in exchange for exclusive access to all its outlets, regardless of whether there was a PiS in the store or not. The agreements with COOP and ICA similarly granted Norway Post exclusivity but only in outlets where a PiS was established.
After a complaint in 2002 from recent entrant Swedish Privpak, the EFTA Surveillance Authority (ESA) opened an investigation into these exclusive agreements. The Authority noted that Norway Post was the only supplier of B-to-C parcel services with a network covering the whole of Norway. Its purported share remained about 98 per cent in the market for over-the-counter B-to-C parcel delivery. The ESA concluded that Norway Post had abused its dominant position by preventing new entrants from access to leading grocery store, kiosk and petrol station chains and, as a consequence, prevented them from establishing a delivery network capable of competing effectively with Norway Post’s network. Norway Post was fined €12.89 million.
On appeal, the EFTA Court confirmed the Authority’s decision but reduced the fine to €11.11 million because ESA’s administrative procedure had taken too long.
Norway Post argued that to establish anti-competitive fore-closure, it is not enough to show that one path of access to the market was closed off. ESA must demonstrate that effective access to the market was hindered or eliminated and that the conduct made it possible for the dominant undertaking to exercise market power. Moreover, ESA did not specify the degree of potential foreclosure or show that it was substantial. Since the exclusivity was imposed on distributors rather than end-users, competitors were still able to compete for the entire market. Norway Post contended that its conduct was intended to secure a fast and efficient roll-out of a new distribution model. The exclusive agreements aimed to ensure effective implementation of its public service obligations and protect a significant investment in a new delivery network.
The Court firmly rejected these arguments, saying that ‘it was sufficient for ESA to show that the conduct in question was liable to distort competition by raising barriers to entry and, therefore, to the maintenance or growth of the competition still existing in the market.’1 It further noted that ‘for the same reason, it is immaterial whether an ‘as efficient competitor’ could have competed effectively with Norway Post.’2Entry barriers precluded efficient competitors from competing at all.
On the market coverage, the Court found it apparent from the decision that ESA had established that the retail outlet foreclosure was ‘substantial’.3 ESA had shown that access to the leading outlet chains was important for providers of B-to-C parcel services and Norway Post’s conduct had significantly limited rivals’ possibilities of establishing delivery services at these chains. Because of the weak competitive situation in the market for over-the-counter B-to-C parcel delivery, the Court agreed that ‘even a limited degree’ of foreclosure was liable to restrict competition in the market.4
Norway Post argued that its agreements were lawful because they only foreclosed 40 per cent of the market, and that a competitor could profitably set up a delivery network consisting of outlets other than those used by Norway Post. In dismissing these arguments the Court noted that the foreclosure rate was actually 50 per cent, if not more, which it deemed ‘substantial’.5 More importantly, the Court concluded that foreclosure of a substantial part of the market cannot be justified by showing that the remaining part of the market is sufficient for a limited number of competitors; ‘it is not for the dominant undertaking to decide how many viable competitors will be allowed to compete for the remaining contestable portion of demand.’6 Customers should benefit from whatever degree of competition is possible on the market and competitors should be able to compete for the entire market and not just for a part of it.
Norway Post also argued that although the alleged infringement continued for five-and-a-half years, there was no evidence of actual anti-competitive effects. They pointed to evidence that there was no price decrease after the exclusivity ended. Rejecting this argument, the Court reiterated established precedent that an analysis of actual anti-competitive effects is not required. It is enough that the ‘practices tended to restrict competition or that their conduct was capable of having that effect’.7 Even though ESA went one step further and supplemented its analysis by considering the likelihood of actual anti-competitive effects, it was not required to do so. The Court nonetheless acknowledges that ‘in some cases, the persistent lack of actual negative effects on competition may cast doubt on a finding by ESA that a certain conduct is liable to restrict competition’.8 But Norway Post’s claims in this particular case are not convincing enough to cast such doubt. Where an undertaking is already capable, as a result of its dominant position, to set the price for its products and services independently of those of its competitors, the creation of such additional barriers does not necessarily lead to an increase in prices but merely can perpetuate its existing dominant position. Similarly, the removal of such additional barriers to entry does not necessarily lead to a decrease in prices.
Finally, Norway Post claimed its conduct was objectively justified and the ESA applied too strict a test when assessing objective justifications and efficiencies, especially considering that Norway Post provides a service of general economic interest. It claimed a legitimate interest in pursuing chain exclusivity, outlet exclusivity and its renegotiation strategy.
The Court quickly dismisses all these justifications, partly because Norway Post introduced new pleas that were inadmissible at this stage of the procedure. On the efficiencies, the Court concluded that the conduct was not objectively necessary and that Norway Post in any event failed to show that efficiency gains were passed on to consumers. The fact that the state, and thereby taxpayers, may have avoided some expenses was irrelevant.
Analysing exclusive dealing
The elements of an infringement for exclusive dealing are well-established and comparable to other abuse of dominance cases:
- market power by at least one party to the arrangement;
- conduct that forecloses rivals from the opportunity to compete; and
- the absence of an overriding legitimate business justification.
The market power requirement is, of course, hugely important. Not only is it a statutory requirement under article 102, but it is a prerequisite for a firm to have the ability to foreclose rivals. Yet whether a firm actually has market power is rarely knowable without the sort of extensive economic analysis undertaken during the course of in-depth investigations or litigation, not when merely contemplating entering into a contract. Defining markets is notoriously uncertain, as are the likelihood of entry or expansion, effective strategies by power buyers to protect themselves and the many other variables relevant to whether a large firm actually has market power. Except in the most obvious cases, a prudent analysis sometimes must assume that market power exists for the sake of argument and focus instead on whether the conduct is foreclosing.
Whether exclusive contacts are foreclosing often presents the more interesting issue. Courts and regulators typically look to some combination of the percentage of the available customer base to whom the contracts applied and their duration. Over a decade ago, the European Commission and the courts found unlawful the exclusive contracts in Langnese-Iglo9and Schöller Lebensmittel.10 Langnese and Schöller were incumbent ice cream suppliers who had exclusive contracts with retailers representing more than 30 per cent of the market. The duration of Langnese’s agreements averaged two-and-a-half years, while Schöller’s agreements were considered indefinite since they were subject to tacit renewal.
More recently, in Tomra,11 the EU Courts upheld the Commission’s finding that Tomra’s contracts were unlawful in markets for ‘reverse-vending machines.’ The contracts involved volume discounts, (which unfortunately were assumed without apparent analysis to be the same as exclusive contracts), but they covered about 40 per cent of the relevant markets.
Similarly, in the United States, the Microsoft court said that ‘a monopolist’s use of exclusive contracts, in certain circumstances, may give rise to a section 2 violation even though the contracts foreclose less than the roughly 40 per cent or 50 per cent share usually required in order to establish a section 1 violation.’12 By contrast, the appellate court in Gilbarca found lawful exclusive contracts where the defendant held a 55 per cent share of the market for retail gasoline dispensers and entered into exclusive arrangements covering 38 per cent of all outlets in the market.13 In reversing a finding of liability by the district court, the Court of Appeals for the Ninth Circuit observed that the arrangements were short-term and easily terminated.
Although these cases turned on market coverage and duration, it would be a mistake to conclude that they lay down hard and fast thresholds for these variables, where three-year contracts are fine but not 37 months, or contracts covering 40 per cent of the market are fine but not 41 per cent. Nor are the EC Guidelines on Vertical Restraints helpful in saying that single branding obligations exceeding five years are ‘for most types of investments not considered necessary’ to achieve the claimed efficiencies, given the broad range of circumstances.14Each case turns on its facts, but what facts are relevant? The answer begins with an examination of the basic principles of anti-competitive foreclosure under EU law, many of which are recently developed in pricing cases such as Intel,15 TeliaSonera,16 Tomra17 and Post Denmark.18
Foreclosure deprives rivals from the opportunity to compete
Rivals are not foreclosed merely because they lose the competition to win contracts. Foreclosure requires rivals to be denied an opportunity to compete, placed at a material disadvantage for reasons other than the quality of their product or the merits of their service. Foreclosure occurs when ‘actual or potential competitors are completely or partially denied profitable access to a market.’19 Merely winning business from rivals is not foreclosure, but rather the sort of competition on the merits that the competition laws are meant to protect. At the same time, the fact that a contract can be tacitly renewed, as in Schöller, does not make its duration infinite for these purposes since rivals can compete for it and customers can switch without breaching the contract.
Foreclosed rivals must be equally efficient
If competitors can compete for the entire exclusive contract, merely losing is not foreclosure. As the EC’s Vertical Guidelines explain: ‘If competitors can compete on equal terms for each individual customer’s entire demand, single branding obligations [exclusive contracts] of one specific supplier are generally unlikely to hamper effective competition unless the switching of supplier by customers is rendered difficult due to the duration and market coverage of the single branding obligations.’20 This can happen when one firm is a ‘must have’ for only part of the customers’ demand, but insists on an ‘all or nothing’ contract to foreclose competition from the contestable portion of the customer’s needs.
This issue has played out most recently in volume discount cases, since they can (sometimes) operate as exclusive contracts. The issue there is whether an equally efficient rival can compensate customers for forgoing across-the-board discounts, but on the basis of a smaller supply. By allocating the entire discount to the contestable portion of demand, we can determine whether the discount operates effectively as an exclusive contract. Switching part of the business during the course of the arrangement is precluded in both cases; one because of a contractual commitment and the other because equally efficient rivals cannot compete for the contestable portion. But once volume discounts are shown to be operate like exclusive contracts, it is still necessary to assess whether they are anti-competitive.
Dominant firms may not dictate the size of rivals
This is the comment made by both the Tomra and Norway Post courts in response to arguments that only a portion of the market is foreclosed. Some commentators have wrongly interpreted the comment to suggest that all exclusive contracts by dominant firms might be suspect since they necessarily limit, to some extent, the amount of the market for which rivals can compete at any given time. But the statement means no such thing and, as discussed below, is instead a useful principle when assessing whether rivals are foreclosed from the market.
With these principles in mind, it is possible to give context to the market coverage and duration holdings noted above. The issue in those cases and others is straightforward when assessing foreclosure: how much business comes up for competition each year? The amount of the market covered by exclusive contracts and their duration are thus not conclusive in themselves but relevant to this key fact. The fewer the number of contracts and the shorter their duration, the more business comes up each year. This is the variable determining whether equally efficient rivals have an opportunity to compete, regardless of whether they actually win. If so, they are not foreclosed.
The amount of business coming up each year is also the variable that gives content to the sensible point that dominant firms should not be allowed to dictate the size of their rivals. Of course, the mere existence of contracts in any market means that not all business is up for grabs at every moment. But this does not mean that rivals are ‘foreclosed’ from the market or that they are relegated to only a portion of it. Rivals may have competed for these same contracts, and they can certainly do so again when the contracts expire. This is the ‘opportunity’ to compete that precludes a finding of foreclosure. It is what places a competitive constraint on dominant firms and thus constitutes competition. And when competition takes the form of intermittent bidding events, this ability to compete for every contract as it expires means rivals are not confined to the portion of the market not currently covered by the contracts. They can compete in the entire market as each competitive event occurs.
It therefore is necessary to determine how much business coming up for bid is enough. The answer of course is highly fact-based, but several common questions are relevant.
The first is whether equally efficient rivals had an opportunity to compete for the same exclusive contracts. If the customer’s entire demand is contestable, not just a portion, regulators should at least hesitate before condemning the firm that won it. On the other hand, concerns may arise even in these circumstances if exclusive contracts cause too little business to come up for bid each year to constitute a real ‘opportunity’ for rivals to compete. The same is true if the concern is that ongoing contracts are precluding potential new entrants from an opportunity to secure initial business.
The second is how much is enough business to come up for bid each year in light of the market realities. Foreclosure principles require that we look at whether the amount constitutes a real ‘opportunity’ to compete in the market. And the question is not whether the business is merely enough to create a small competitor, but whether to allow it to maintain its presence as a competitive constraint as contracts expire. One relevant factor will be the rivals’ minimum viable scale, not because it is permissible to cap their growth at this level but because reaching it (or, more accurately, having the opportunity to reach it) allows rivals to remain a competitive force regardless of whether it wins.
Also relevant is the extent to which rivals can remain viable bidders and for how long, even if winning business might take some time. Rivals currently competing in neighbouring markets, for example, may have a greater ability and incentive to wait for exclusive contracts to expire than start-up firms who need a quick win to survive. So the key question is understanding the qualities of the potential competitive constraints to assess whether those rivals have an opportunity to compete for each customer as the events arise, regardless of whether they win. If so, the contracts do not foreclose rivals or cause the incumbent to have market power.
All these factors must be assessed in light of the efficiencies and customer benefits created by the exclusive arrangements. It is now well-established in the EU, as in the US, that even dominant firms are permitted to compete hard for business, even if doing so maintains their dominance, so long as they are competing on the merits rather than through foreclosure tactics. But competition on the merits is not necessarily limited to better service and lower prices. Exclusive contracts often create the sort of efficiencies that benefit customers and they should not be rejected without a real competitive assessment.
To be sure, a thorough risk assessment turns on more than the legal analysis set forth above; it includes the probability of complaints, the certainty of the facts, the magnitude of the benefits and a range of business factors. But recent cases like Post Norway, one of the few exclusive dealing judgments, are a useful addition to the analysis. While uncertainty is never eliminated entirely, the law on exclusive dealing has developed to allow companies to pursue legitimate business objectives with manageable risk.
- Case E-15/10 Posten Norge AS v EFTA Surveillance Authority, 18 April 2012, paragraph 131. ↑
- Paragraph 132. ↑
- Paragraph 136. ↑
- Id. ↑
- Paragraph 160. ↑
- Id. ↑
- Paragraph 187. ↑
- Paragraph 192. ↑
- Case T-7/93 Langnese-Iglo v Commission  ECR II-01533. ↑
- Case T-9/93 Schöller Lebensmittel v Commission  ECR II-01611. ↑
- Case C-549/10 P Tomra and Others v Commission, 19 April 2012, not yet reported. ↑
- United States v Microsoft Corp, 253 F.3d 34 (D.C. Cir. 2001) ↑
- Omega Envti, Inc v Gilbarca, Inc, 127 F.3d 1157, 1163 (9th Cir. 1997). ↑
- EC Vertical Guidelines (2010), paragraph 133. ↑
- Commission Decision of 13 May 2009, COMP/C-3/37.900 – Intel. ↑
- Case C-52/09 Konkurrensverket v TeliaSonera Sverige  ECR I-00527. ↑
- Case C-549/10 P Tomra and Others v Commission, 19 April 2012, not yet reported. ↑
- Case C-209/10 Post Danmark A/S v Konkurrencerådet  ECR I-0000. ↑
- DG Competition discussion paper on the application of Article 82 of the Treaty to exclusionary abuses (December 2005) (‘EC Discussion paper’), paragraph 58. ↑
- Paragraph 133. ↑
Latham & Watkins LLP
Boulevard du Régent 43-44
Tel: +32 2 788 6000
Fax: +32 2 788 6060
Latham & Watkins is a full-service international law firm with approximately 2,100 lawyers in 32 offices around the world. Widely recognised as one of the global elite law firms, Latham’s practice spans a wide range of transactional, litigation, corporate and regulatory areas, with extensive expertise and specialist knowledge in numerous industries.
Latham’s global antitrust and competition practice group includes over 150 dedicated lawyers skilled at working across specialties and jurisdictions to offer clients the integrated and sophisticated response required to meet their global challenges. The firm’s European competition lawyers work out of our offices in the major jurisdictions throughout Europe, including Brussels, Frankfurt, Hamburg, London, Madrid, Milan and Paris. They are consistently ranked among the premier lawyers in their field.
The firm handles all aspects of EU competition law before the European Commission, the member states and the courts. Its work includes cutting-edge cases involving alleged abuses of dominance, complex merger clearances and multinational cartel investigations.
Next Chapter: State Aid