Jump to Content Jump to Main Navigation

8 The Law And Economics Of Vertical Restraints

Damien Geradin, Anne Layne-Farrar, Nicolas Petit

From: EU Competition Law and Economics

Damien Geradin, Dr Anne Layne-Farrar, Nicolas Petit

From: Oxford Competition Law (http://oxcat.ouplaw.com). (c) Oxford University Press, 2015. All Rights Reserved. Subscriber: null; date: 27 October 2020

Guidelines on Vertical Restraints — Basic principles of competition law

(p. 463) The Law And Economics Of Vertical Restraints

  1. I.  Introduction 8.01

  2. II.  Types of Vertical Restraint 8.06

    1. A.  The Issue 8.06

    2. B.  The Exclusive Contractual Relationship Group 8.08

    3. C.  The Resale Price Maintenance Group 8.18

    4. D.  The Limited Distribution Group 8.21

    5. E.  The Market-Sharing Group 8.29

    6. F.  The Buyer Power Group 8.38

  3. III.  A Step-by-Step Method for the Self-Assessment of Vertical Restrictions 8.45

    1. A.  Screening of Vertical Restraints 8.47

    2. B.  Full-Blown Competition Analysis of Vertical Restraints 8.86

  4. IV.  Online Distribution 8.100

    1. A.  The Context 8.100

    2. B.  The New Framework for Online Distribution 8.104

  5. V.  Conclusions 8.112

I.  Introduction

8.01  Each manufacturer seeking to convey goods or services to customers faces a basic, standard choice.1 It may either distribute them directly, through downstream vertical integration (integrated distribution), or it may entrust this task to specialized agents via the conclusion of a vertical agreement (independent distribution).2

(p. 464) 8.02  This chapter focuses on independent distribution and how it is treated under EU competition law. Since the 1960s and the Consten and Grundig case,3 it is acknowledged that vertical agreements can entail restrictions of competition—generally called ‘vertical restraints’—which deserve competition law scrutiny.4 While the early case law and Regulations adopted in the field focused primarily on restrictions of intra-brand competition,5 and condemned many contractual clauses per se, regardless of their actual effects,6 a more liberal and economic approach was introduced with the promulgation of Regulation 2790/1999.7 For a long time, legal and economic studies had cast light on the fact that vertical agreements had as many pro-competitive effects as anticompetitive ones.8 Following one of the most vociferous debates in the history of EU competition law,9 a consensus emerged that competition authorities needed to focus on cases raising real competition concerns, that is, those where inter-brand competition (between competing goods or services) is actually hampered, so that any further restriction of intra-brand competition becomes problematic. Vertical restraints on tight oligopolistic markets were a case in point. On those markets, which are prone to collusive outcomes, competition between suppliers tends to be weak.10

8.03  In response, the Commission launched studies in 1997 with a view to reforming the law of vertical agreements.11 This process culminated in 1999 with the adoption of block exemption Regulation 2790/1999. The new legal framework, which represented a radical departure from the previous approach,12 rested on a basic economic premise: the ability of a vertical agreement to produce anticompetitive effects hinges predominantly on the market power of the parties to the agreement (especially the supplier’s market share, which reveals the degree (p. 465) of inter-brand competition). In turn, the system created a safe-harbour mechanism whereby any agreement between parties holding less than a predetermined ‘market share’ (and which did not include a (shorter) list of black clauses and conditions) would be presumed to benefit from an exemption under Article 101(3) TFEU.13 Above the relevant market share threshold, an individual assessment needed to be carried out in light of the principles mentioned in a set of complementary Guidelines.14 In addition, the new legal framework also governed internet distribution, which was not covered previously.15

8.04  With the expiry of Regulation 2790/1999 on 31 May 2010, but also with the growth of large retailers throughout Europe and the rise of internet distribution (B2B or B2C), the Commission initiated a review process in July 2009 which culminated in the adoption of Regulation 330/2010 and of a new set of Guidelines.16

8.05  Against this background, the purpose of this chapter is to provide an overview of the new legal regime applicable to vertical agreements. Following this introduction, Section II sets out the different types of vertical restraint and the theories of competitive harm ascribed to them. Section III offers a step-by-step overview of the method that should be followed by firms and their counsel in the post-notification era with a view to self-assessing vertical agreements under EU competition law. Section IV deals with the issue of online distribution, which sparked intense controversy during the stakeholder consultation process. Finally, a brief conclusion is provided in Section V.

II.  Types of Vertical Restraint

A.  The Issue

8.06  With the paradigmatic shift towards an economic approach in 1999, and its confirmation in the recently adopted texts, EU law has migrated from a ‘form-based approach’ to a so-called ‘effects-based approach’. Pursuant to Regulation 330/2010 and the Guidelines, the crux of the matter is to determine whether a vertical agreement (or part of it) has actual or potential anticompetitive effects that are not outweighed by pro-competitive effects (or objective justifications).

8.07  Within the new regulatory framework, six groups of vertical restraints, with distinct possible anticompetitive effects, can be distinguished.17 In line with the way economists work, the Guidelines ascribe one or more theories of competitive harm to each of these types of (p. 466) restriction and identify their possible countervailing objective justifications as well as procompetitive effects.

B.  The Exclusive Contractual Relationship Group

(1)  Notion of exclusive contractual relationships

8.08  In an exclusive contractual relationship, a party to a vertical agreement relinquishes its freedom to contract with a third party. Its most drastic variant can be found in ‘single branding’ arrangements, which limit a buyer’s ability to buy, resell, or use as inputs competing goods or services.18 Less extreme declinations of exclusive contractual relationships include quantity forcing,19 conditional rebate schemes (although in some cases they may effectively amount to exclusive agreements), two-part tariffs (fixed fee plus a price per unit), tying arrangements (where the sale of one product is conditional upon the purchase of another),20 or any other clauses (eg ‘English clauses’)21 or penalties which rigidify the supplier/buyer relationship by encouraging the buyer to concentrate its purchases of goods or services with one supplier.22

8.09  The Van den Bergh Foods case is a good illustration of an exclusive contractual relationship.23 A supplier of ‘impulse’ ice cream had made freezers available to its Irish distributors for free. In return, the distributors were contractually required to refrain from storing other brands of ice cream in those freezers.

8.10  An exclusive contractual relationship may also result from exclusive supply arrangements. Here, the supplier is obliged (or incentivized) to sell the contractual goods or services only (or mainly) to one buyer, in general or for a particular use.24

(2)  Theories of competitive harm

8.11  Drawing on economic theory, the Guidelines ascribe three theories of competitive harm to exclusive contractual relations, namely foreclosure, collusion, and reduced consumer choice. First, exclusive contracts may foreclose competitors’ access to outlets (eg in the case of single branding or long-term contracts) or inputs (eg in the case of exclusive supply). In a market subject to widespread single branding arrangements, a new supplier hoping to enter the market has no other choice but to set up its own distribution network (with the attendant costs and risks that this involves).25 Such arrangements may thus constitute a barrier to entry.

(p. 467) 8.12  Second, collusion may be facilitated when all competitors make use of exclusive contracts.26 Collusion describes a situation in which rival oligopolists explicitly or tacitly agree to align their commercial policies. Because exclusive contracts rigidify the market shares of rival suppliers, they undermine oligopolists’ incentives to cheat from the collusive price through a price cut.27

8.13  Finally, single branding may harm consumer welfare when the buyer is a retailer which deals directly with the final consumer.28 In such a setting, single branding reduces consumer choice within the point of sale.

(3)  Objective justifications and pro-competitive effects

8.14  The economic literature is replete with articles arguing that exclusive contractual relationships have often objective justifications and pro-competitive effects.29 The Guidelines on vertical restraints take several of those findings on board. They focus, in particular, on three types of welfare-enhancing effects that may arise from exclusive contractual relationships.

8.15  First, exclusive contractual relationships can neutralize free-riding amongst rival manufacturers. A free-rider problem arises when a manufacturer finances the pre-and post-sales investments of its retailers (advertisement, promotional expenses, training of workforce, etc). This situation generates a ‘positive externality’ which benefits rivals. Those consumers that have been drawn to the relevant point of sale thanks to the supplier’s promotional efforts may eventually purchase a less expensive competing product, since its manufacturer did not incur the promotional expenses.30 Exclusive contractual relationships, and in particular single branding,31 prevent competitors from free-riding on each others’ investment suppliers.32

8.16  Second, in some sectors which often involve branded or positional goods/services, exclusive contractual relationships limit ‘certification’ issues. A manufacturer willing to introduce a new ‘premium’ product/service needs to sell primarily through retailers whose reputation is to stock only ‘quality’ products. If the manufacturer does not limit its sales to such premium stores, its product/service may be undervalued by customers and its marketing strategy may be put into jeopardy. To convince ‘premium’ stores to sell the premium product, (p. 468) manufacturers may thus have recourse—at least for a certain period—to exclusive contractual relationships (such as exclusive supply, etc).33

8.17  Third, exclusive contractual relationships are often said to solve ‘hold up’ problems.34 Such issues arise when a buyer makes a ‘specific’ investment for the performance of a vertical contract (an oil refinery builds a pipeline linking its facilities to those of a particular oil supplier). The investment is ‘specific’, because apart from the particular contractual relationship, it has no other value. It is, as economists would say, a ‘sunk cost’. Hostage of its own investments, the buyer ends up locked into a commercial relationship with the supplier, which may for instance engage in opportunistic behaviour by discontinuing the supply of oil or by raising prices. To prevent such a risk, the parties may enter into exclusive contractual relationships, for instance by entering into a long-term exclusive contract, which specifies given quantities and a ceiling price.35

C.  The Resale Price Maintenance Group

(1)  Notion of resale price maintenance

8.18  Resale price maintenance (RPM) arises when a supplier imposes—directly or indirectly—a resale price on its buyers.36 RPM may take the form of a fixed, minimum, or maximum resale price. RPM may also arise in disguise when a supplier recommends a resale price but ensures its observance through incentives (payment of premiums, discounts, or other benefits to its distributors) or penalties (eg threats to terminate the contract). A rare example of the Commission dealing with this type of restriction is provided by the B&W case.37 In 2002, the Commission approved B&W Loudspeakers’ notified selective distribution network on condition that the company remove several hardcore restrictions, such as a disguised resale price maintenance clause stipulating minimum retailer prices and margins.

(2)  Theories of competitive harm

8.19  The theories of competitive harm ascribed to RPM assume that the supplier imposes the same price on all of its buyers. They borrow heavily from the economics of horizontal collusion. First, in ways similar to horizontal price collusion amongst purchasers,38 RPM eliminates (p. 469) price competition at the distribution level.39 However, this should only be a cause of concern if the share of the market covered by RPM is significant. An assessment of the aggregate market share of the retailers subject to RPM would thus seem necessary to determine whether it has adverse effects on consumer welfare. Second, RPM facilitates horizontal price collusion amongst suppliers (tacit or explicit).40 Fixing resale prices makes it easier for parties to a horizontal cartel to monitor adherence to a joint line of action. The reason is that when prices are fixed at the resale level, a reduction of market shares at the upstream level can only be explained by the deviation of a cartelist.41 Finally, RPM often entails a uniform resale price, which prevents efficient price discrimination by resellers (price discrimination leading to an increase in output and a more efficient recovery of fixed costs).

(3)  Objective justifications and pro-competitive effects

8.20  The works of Lester Telser42 in the 1960s—and more generally of Chicago scholars—have shed light on the objective justifications, and possible pro-competitive effects, of RPM.43 First, RPM arguably allows suppliers to protect themselves against the risk that retailers lower prices at the expense of quality of service.44 Second, RPM protects the image of certain branded or positional products, which would otherwise be less valued by customers (eg luxury goods). Third, suppliers will often use RPM to convince new operators to join a distribution network. Absent RPM, new retailers may hesitate joining a distribution network for fear of facing aggressive price competition from incumbent retailers. Fourth, when both the upstream and downstream markets are subject to a monopoly, RPM eliminates risks of ‘double marginalization’.45 Fifth, when some retailers offer pre-and post-sales services and others do not, resale price maintenance removes risks of free-riding.

(p. 470) D.  The Limited Distribution Group

(1)  Notion of limited distribution

8.21  In a limited distribution system, a supplier restricts the number of distributors to which it sells goods or services.46 In a first variant, the supplier selects a limited number of distributors on the basis of a set of quantitative and/or qualitative criteria.47 This type of limited distribution is referred to as ‘selective distribution’. In a second variant, the supplier decides to appoint one distributor for a given geographic area (or for a category of customer). This type of limited distribution is referred to as ‘exclusive distribution’ (or ‘customer exclusivity’).48

(2)  Theories of competitive harm

8.22  Limited distribution systems are likely to trigger two types of restrictive effect on competition. First, limited distribution generates foreclosure concerns. Those buyers that do not belong to a limited distribution network are unable to obtain inputs from the relevant supplier. There is therefore a risk of ‘input foreclosure’ which, depending on the market power of the manufacturer, may in turn reduce intra-brand competition at the buyer level.49

8.23  Second, because it is easier to coordinate amongst a small number of entities, limited distribution facilitates collusion—tacit or explicit—amongst purchasers .50 In addition, limited distribution facilitates collusion—tacit or explicit—amongst suppliers, because the monitoring of deviations becomes easier with a limited number of retail outlets. This risk is particularly acute when several suppliers appoint the same distributor for the same territory.

(3)  Objective justifications and pro-competitive effects

8.24  Most of the objective justifications for limited distribution are predicated upon the economic theory of incentives.51 In particular, Chicago School scholars have long argued that suppliers grant contractual protection to their distributors to stimulate their incentives to invest. On markets for durable and complex goods (cars, computers, stereos, etc), distributors often provide essential pre-and post-sales services such as consulting, testing, demonstrations, explanations of the relevant documentation, etc.52 Those services come, however, at a cost, which in turn translates into higher prices. Distributors providing such services are thus vulnerable to the risk that consumers select a product within their point of sale, but subsequently purchase it from a different distributor who offers a more attractive price (precisely because it has not incurred similar costs in pre-and post-sales services).53

(p. 471) 8.25  To maintain retailers’ incentives to invest in pre-and post-sales services, and avoid this freerider problem, suppliers may grant territorial protection, customer exclusivity, or enter into an exclusive supply commitment. More generally, such contractual protection is equally useful when a distributor must make an ‘initial investment’ on a new market,54 undertake ‘promotional efforts’,55 or when it has a certain reputation for quality on the market.56

8.26  In addition to such ‘incentive’ effects, limited distribution engenders a range of ‘costs’ effects. First, limited distribution triggers economies of scale, that is, a reduction in the average total costs of production (the typical cost of each unit produced).57 For instance, in exclusive supply relationships, the supplier only deals with one buyer. Assuming that the supplier finances (part of) a buyer’s equipment, the fixed financial burden of this investment can be allocated over its entire production scale. In contrast, if the supplier has contractual dealings with several buyers, its investments in equipment would be multiplied and the financial burden of each fixed investment would have to be spread over fewer quantities. As a result, its average total cost would be higher.58

8.27  Second, limited distribution reduces the number of contractual partners of the supplier, and thus limits transaction costs (negotiations, delivery, billing, monitoring, etc).59

8.28  Finally, some types of limited distribution arrangements, and in particular selective distribution, allow suppliers to prevent ‘principal–agent’ problems. Some retailers subject to competition may neglect the quality of post-sale service, and in turn harm the uniform reputation/brand image of the suppliers’ good. To alleviate such concerns, suppliers may apply ex ante selection systems (selective distribution),60 or contractually require that the distributors comply with a list of specifications in terms of know-how and the image they wish to convey (franchising).61

E.  The Market-Sharing Group

(1)  Notion of market sharing

8.29  In a market-sharing arrangement, a supplier restricts the venues where its buyers can purchase or sell contractual goods/services.62

(p. 472) 8.30  A first form of market sharing is exclusive purchasing.63 Here, a buyer commits to purchase exclusively from one particular supplier (eg from a regional wholesaler) in order to meet its requirements of a given product/service. Thus, the buyer cannot purchase from other suppliers of the same product (other wholesalers in other geographical regions).64

8.31  A second conventional form of market sharing involves the resale side of the market, and often arises in the context of exclusive distribution systems (where each buyer is primarily responsible for the resale of the product/service within a given territory or to a designated type of customer).65 There is market sharing when a supplier restricts its distributors’ freedom to resell outside the assigned territory/designated customer base.66 In this context, a distinction is usually drawn between restrictions of ‘active sales’ (the buyer cannot actively solicit customers outside its territory)67 and restrictions of ‘passive sales’ (the buyer cannot meet unsolicited orders from customers outside its territory).68

8.32  Market sharing is a polymorphous concept. In its purest form, a contractual clause may directly forbid the resale of products/services outside the relevant territory (or designated type of customer). Alternatively, the supplier may use indirect incentives (financial rewards or penalties) to encourage distributors to confine their deliveries to their assigned territory/customer base.

8.33  The Nintendo case provides a good illustration of unlawful market sharing. In a 2002 decision, the Commission found that Nintendo and several of its EU distributors had colluded to artificially keep high price differentials across several Member States.69 Under the collusive arrangement, each distributor was required to prevent parallel trade from its territory to other territories (parallel trade involves exports from low price countries to high price countries).70 Nintendo and several distributors had taken active steps to stem parallel trade. Distributors that had allowed parallel exports were punished through supply (p. 473) reductions (or even boycott). As a result of such conduct, the Commission meted out a €167.8 million fine on Nintendo and seven of its official distributors.

(2)  Theories of competitive harm

8.34  In a market-sharing system, each buyer enjoys a monopoly over the resale of a product/service to a particular territory/type of customer. Intra-brand competition is thus entirely eliminated. If inter-brand competition is limited, each buyer thus enjoys ‘significant market power’ (the ability to raise prices significantly and durably above the competitive level).71

8.35  In addition, market sharing can thwart the integration of the internal market when resale restrictions partition markets along national lines. In this variant, differentiated prices will prevail across Member States. This goes against the philosophy of EU market integration, which seeks to ensure, whenever possible, homogeneous conditions of trade for customers across the EU.72

8.36  Finally, market sharing facilitates collusion. With market sharing, a buyer’s price cuts do not generate larger sales. Hence, buyers’ incentives to deviate from a tacit or explicit collusive equilibrium are limited.

(3)  Objective justifications and pro-competitive effects

8.37  Even more clearly than exclusive distribution, market sharing eradicates free-rider issues, and may thus encourage buyers to invest in pre-and post-sales services. Importantly, such systems are often imposed by suppliers on—possibly reluctant—buyers who are prone to compete aggressively on price at the expense of quality.73

F.  The Buyer Power Group

(1)  Notion of buyer power

8.38  In mainstream competition economics, buyer power is traditionally seen as a disciplining factor against the market power of large suppliers.74 This is because strong buyers have the ability and incentive to bring new sources of supply on the market in response to a small but permanent increase in relative prices.75

8.39  In this context, the Guidelines delve into relatively unchartered territory by turning buyer power into theories of competitive harm. The Guidelines seek in particular to offer guidance on two novel areas, namely upfront access payments (ie, payment of fixed fees by suppliers to retailers in order to gain access to their shelf space—known as slotting allowances) and category management agreements (ie, agreements where the distributor entrusts a given (p. 474) supplier—a ‘category captain’—with the marketing of a category of products, which include rival products).

8.40  Such issues have, however, been dealt with at Member State level. The UK Office of Fair Trading (OFT), for instance, grappled with the issue of category management in the acquisition by United Biscuits (UK) Ltd of the Jacobs Bakery Ltd. In this case, some competitors raised concerns about the power of the merged firm to further its own sales at the expense of its rivals using its control over the supermarkets through the category management process. In the context of this merger, the OFT did not, however, expect a retailer (particularly a major supermarket chain) to permit itself to be disadvantaged by its choice of category manager, or be bound by its recommendations.76

(2)  Theories of competitive harm

8.41  The Guidelines ascribe two theories of harm to upfront access payments. First, they may lead to anticompetitive foreclosure of (i) other distributors if such payments ‘induce the supplier to channel its products through only one or a limited number of distributors’77; (ii) of other suppliers where the widespread use of upfront access payments increases barriers to entry for small entrants.78 Second, the Guidelines state that upfront access payments may reduce competition and facilitate collusion between distributors. According to the Commission, upfront access payments are likely to increase the price charged by the supplier for the contract products since the supplier must cover the expense of those payments. In turn, and without much explanation in relation to the collusion issue, the Guidelines consider that these higher supply prices may reduce retailers’ incentives to compete on price downstream, while the profits of distributors are increased as a result of access payments.79

8.42  According to the Commission, category management agreements generally do not raise competition law concerns. Yet, they may occasionally lead to the foreclosure of other suppliers ‘where the category captain is able to limit or disadvantage the distribution of products of competing suppliers’.80 In addition, such agreements may facilitate collusion between distributors when the same supplier serves as a category captain for all or most of the competing distributors on a market. In a rather terse manner, the Guidelines state that in such cases the category captain will provide a common point of reference for the distributors’ marketing decisions.81 More convincingly, the Guidelines consider that category management may also facilitate collusion between suppliers through increased opportunities to exchange sensitive market information.82

(p. 475) (3)  Objective justifications and pro-competitive effects

8.43  As far as upfront access payments are concerned, the Commission recognizes that they may lead to efficiencies such as the efficient allocation of shelf space for new products. In addition, they reduce the risk of free-riding by suppliers on distributors’ promotional efforts.83 This is relevant in particular when suppliers are tempted to launch new products, which are suboptimal. With upfront access payments, the risk of commercial failure does not bear entirely on the buyer. In other words, upfront access payments represent a risk-sharing mechanism, which decreases the risk that suppliers will launch suboptimal products at the expense of buyers.

8.44  As far as category management agreements are concerned, the Commission considers that they may allow distributors to have access to the supplier’s marketing expertise for a certain group of products. In particular, since such agreements are based on customers’ habits, they lead to increased customer satisfaction by satisfying demand expectations. Put simply, they ensure that the optimal quantity of products is presented directly on the shelves in a timely manner.84 Moreover, such agreements generate cost savings for the buyer, which outsources the management of a category of products to a supplier. Finally, category management agreements generate economies of scale, as the cost of managing a category of products is merely incurred once (by the category captain), and can be spread over a wide range of products.

III.  A Step-by-Step Method for the Self-Assessment of Vertical Restrictions

8.45  With the increased risks stemming from the enforcement of the EU competition rules (ie, heavy fines, annulment and damages actions, damage to reputation, etc), firms should regularly self-assess vertical agreements through the lens of Article 101 TFEU.85

8.46  To this end, the Guidelines on Vertical Restraints suggest that ‘the assessment of a vertical restraint involves in general … four steps’,86 which requires a preliminary delineation of the relevant market.87 In our opinion, the assessment of vertical restraints can be whittled down to a simpler two-step method. First, firms and their counsel should screen their agreement against a set of compatibility and incompatibility presumptions, which can be found in the Regulation and Guidelines (Section A). Second, only those agreements that do not fall under such presumptions must be subject to a full-blown individual competition analysis (Section B). (p. 476) In the following sections, we only deal with vertical agreements covered by Regulation 330/2010,88 and we take it as a given that trade between Member States is affected.

A.  Screening of Vertical Restraints

8.47  Since 1999, the law of vertical agreements relies extensively on presumptions.89 The Regulation and the Guidelines establish two sets of presumptions, which from the outset permit one to ascertain whether or not their purported vertical agreement falls foul of EU competition rules. Pursuant to those documents, parties to a vertical agreement must first verify whether their purported agreement falls within the presumption of incompatibility provided for in Article 4 of the Regulation (Section 1). Vertical agreements which fall outside this presumption of incompatibility may in turn be presumed compatible with Article 101 TFEU if certain conditions defined in the Regulation and the Guidelines are fulfilled (Section 2). As will be seen, only those vertical agreements that do not benefit from those presumptions are subject to a full-blown competition analysis.

(1)  The presumption of incompatibility

(a)  Preliminary remarks

8.48  Article 4 of the Regulation sets down a list of five ‘hardcore restrictions’, (sometimes referred to as ‘black clauses’) whose presence in a vertical agreement ipso jure leads to (i) a presumption that the agreement as a whole restricts competition within the meaning of Article 101(1) TFEU; (ii) the exclusion of the application of the block exemption to the entire agreement;90 and (iii) a presumption that ‘the agreement is unlikely to fulfil the conditions of Article 101(3)’.91 (p. 477) Such restrictions are thus subject to a quasi per se prohibition. There is therefore no need to undertake a painstaking assessment of the economic effects of the agreement. Moreover, the incompatibility rule affects the entire agreement, which is thus deemed null and void as a whole under Article 101(2).

8.49  Interestingly, and despite internet players’ calls to that effect,92 Article 4 does not specifically target restriction(s) on buyers from selling goods/services on the internet. In our opinion, the silence of Article 4 in relation to online sales is precautionary in nature. Absent any track-record on this issue, the Commission has seemed reluctant to cast in stone a restrictive solution, which could soon have become obsolete in sectors driven by rapid technological change.93 Rather, the Guidelines have favoured a case-by-case approach, comparable to the approach taken notably in the French case law.94

(b)  Resale price maintenance

8.50  The first hardcore restriction targets RPM. It declares incompatible agreements which have as their object ‘the restriction of the buyer’s ability to determine its sales price’.95 Under such agreements, the buyer must observe a fixed (or minimum) resale price, set in the contract.96 Importantly, indirect RPM mechanisms are also caught by the prohibition.97 For instance, agreements setting a minimal profit margin98 or a maximum discount level are presumed incompatible.99 In contrast, a maximum100 or recommended101 price (or price level) is not in (p. 478) principle a hardcore restriction unless it disguises an indirect RPM mechanism (when linked to the exertion of pressure, penalties, or incentives).102

8.51  Since the US Supreme Court Leegin ruling of 2007, the strict incompatibility rule applicable to RPM has sparked intense controversy in the EU.103 In Leegin, the US Supreme Court abandoned the century old per se prohibition rule applicable to RPM,104 and subjected those practices to a rule of reason standard (which entails the balancing of the pro-and anticompetitive effects of RPM).105

8.52  Possibly as a result of this controversy, the Guidelines have manifestly relaxed the rigidity of the incompatibility presumption enshrined in Article 4(a) of the Regulation.106 The Guidelines explicitly state that the parties have the ‘possibility to plead an efficiency defence under Article 101(3) in an individual case’.107 The Commission thus apparently exhibits greater receptiveness to the efficiency gains arising from RPM.108

8.53  This impression is further confirmed by the Guidelines, which without claiming to be exhaustive, mention three types of justification for RPM.109 First, RPM may ‘be helpful during the introductory period of expanding demand to induce distributors to better take into account the manufacturer’s interest to promote the product’ by, for example, incentivizing them to redouble their promotional efforts.110 Second, RPM may

be necessary to organise in a franchise system or similar distribution system applying a uniform distribution format a coordinated short term low price campaign (2 to 6 weeks in most cases) which will also benefit the consumers.111

Finally, the Guidelines endorse the Telserian line of justification, in stating that ‘in some situations the extra margin provided by [resale price maintenance] may allow retailers to provide (additional) pre-sales services, in particular in case of experience or complex products [sic]’ and ‘prevent … free riding at the distribution level’ in relation to the provision of such services.112

(p. 479) 8.54  In theory, therefore, the parties are entitled to invoke some of the efficiencies identified by Chicago scholars to counter a finding of incompatibility. That said, in practice, the flexibility introduced in the Guidelines could prove ineffectual. The majority of the efficiency benefits arising from RPM are ‘qualitative’ in nature. The various positive effects on distributors’ incentives are thus not easily amenable to economic quantification as generally required under Article 101(3) TFEU. In turn, it will often prove complex to balance the harmful quantitative effects of RPM against their qualitative benefits under Article 101(3). As a result, the balancing exercise will inevitably hinge on a value judgment—which by its nature is variable, imprecise, and subjective.

(c)  Territorial resale prohibitions

(i)  Principle

8.55  The second hardcore restriction catches measures directly or indirectly restricting the freedom of a buyer from selling goods or services in certain geographical regions or to certain customers.113 Such restrictions are akin to ‘market partitioning’, and should therefore be presumed incompatible. Article 4(b) catches direct obligations not to sell to certain customers/territories, as well as obligations to dismiss orders from other customers/territories (or to refer to other distributors).114 Its scope of application also covers indirect measures which have the same effect, such as incentive schemes (conditional bonuses or discounts) and pressures (refusals to supply, threats to terminate the agreement).115 According to Article 4(b), the vertical agreement may, however, define the ‘place of establishment’ of the buyer.

(ii)  Exceptions

8.56  The Regulation brings four exceptions to the presumption of incompatibility.116 First, restrictions of ‘active sales’ into a territory (or to a customer group) granted exclusively to another buyer are not presumed incompatible. A sale is ‘active’ when the buyer solicits customers located within the exclusive territory (or customer base) assigned to another buyer. In contrast, restrictions on ‘passive sales’ are presumed incompatible. A sale is ‘passive’ when it originates from unsolicited orders of customers located in the exclusive territory (or belonging to the customer base) of another buyer.117 Restrictions of sales via the internet,118 which the Commission deems to be passive and not active sales, are also presumed incompatible

8.57  Pursuant to the three other exceptions enshrined in the Regulation, a supplier can: (i) restrict ‘sales to end users by a buyer operating at the wholesale level of trade’; (ii) prevent ‘members of a selective distribution system to sell to unauthorised distributors within the territory reserved by the supplier to operate that system’; and (iii) restrict ‘a buyer’s ability to sell (p. 480) components, supplied for the purposes of incorporation, to customers who would use them to manufacture the same type of goods as those produced by the supplier.’119

(d)  Restrictions on active and passive sales in selective distribution networks

8.58  The third hardcore restriction concerns selective distribution agreements. Pursuant to Article 4(c) of the Regulation, suppliers cannot restrict the territories/customers (in)to which selective distributors may sell to end-users.120 This presumption of incompatibility covers both active and passive sales. It thus goes further than the hardcore restriction enshrined in Article 4(b).

8.59  Contrary to a common misconception, Article 4(c) does not forbid exclusive-selective distribution networks. Suppliers can freely select distributors and assign to them specific territories/customers (which means that suppliers will not sell products/services to other distributors within the same territory/customer base).121 What is incompatible is to limit the selective distributors’ ability to make active or passive sales to end-users located within the territory of other selective distributors. Similarly, suppliers can impose a restriction on the dealer’s freedom to determine the location of its business premises.122 This latter possibility permits suppliers significantly to impede the ability of pure internet players (those without a physical infrastructure) to join a selective distribution network.

(e)  Restrictions of cross-deliveries in selective distribution networks

8.60  The fourth hardcore restriction also concerns selective distribution agreements.123 Selective distributors must remain free to purchase the contract products/services from other members of the distribution network, operating either at the same or at a different level of trade.124

(f)  Restrictions on component suppliers to sell to end-users, repairers, and independent service providers

8.61  The last hardcore restriction covers agreements between component suppliers and buyers who incorporate them into their own products.125 Original equipment manufacturers (OEMs) may seek to reserve for themselves markets for repair and maintenance. In practice, they may restrict component suppliers’ ability to sell to end-users, repairers, or independent service providers.126 Article 4(e) thus presumes such restrictions incompatible with Article 101(1) TFEU. The presumption of incompatibility also encompasses indirect restrictions, such as prohibitions on the component supplier to provide certain technical information to endusers.127

(p. 481) (2)  The presumptions of compatibility

8.62  Agreements devoid of hardcore restrictions may be presumed compatible with Article 101 TFEU if they fall within one of the three safe harbours established under EU competition law.

(a)  Agreements of minor importance

(i)  The appreciability rule

8.63  In the Franz Völk v SPRL Ets J Vervaecke judgment of 1969, the Court of Justice held that ‘an agreement falls outside the prohibition when it has only an insignificant effect on the markets, taking into account the weak position which the persons concerned have on the market of the product in question’.128 The Commission codified this principle—the so-called ‘appreciability rule’ or de minimis doctrine—in its Notice on agreements of minor importance of 2001.129 Pursuant to the Notice, vertical agreements in which no party holds a market share in excess of 15 per cent are presumed not to appreciably restrict competition within the meaning of Article 101(1) TFEU.130 This presumption of compatibility does not apply if the agreement contains a hardcore restriction.131

(ii)  The practice

8.64  The determination of the supplier and buyer’s market shares requires a prior definition of the relevant product and geographic markets.132 The market share held by a firm fluctuates with the size of the market under consideration. To take a simple example, Coca-Cola’s market share is likely to be small if one takes the view that Coca-Cola operates on the wholesale market for soft drinks. Its market share is obviously much higher if one takes the view that Coca-Cola operates on the wholesale market for carbonated soft drinks with a cola flavour.

8.65  Under EU competition law, the relevant product market comprises all products that consumers consider ‘substitutable by reason of the products’ characteristics, their prices and their intended use’.133 The relevant geographic market ‘comprises the area in which the undertakings concerned are involved in the supply and demand of products or services, in which the conditions of competition are sufficiently homogenous’.134

8.66  In practice, an analysis of ‘substitutability’ must be undertaken to delineate product and geographic markets. To this end, the most conventional technique consists in simulating the effect of a small, but significant and non-transitory increase in prices (5–10 per cent) on the demand for the product (the so-called ‘SSNIP test’). If demand shifts to other products and/or neighbouring geographic areas, it can readily be assumed that these products and/or neighbouring geographic areas belong to a same relevant market.

(p. 482) 8.67  Despite its apparent simplicity, the definition of the relevant market is far from being an exact science. To take again our above example, how can one say with certainty whether Coca-Cola operates on the wholesale market for soft drinks, rather than on the wholesale market for carbonated soft drinks with a cola flavour? Of course, practitioners generally find assistance in decisional precedents, and in particular in the numerous decisions adopted by the Commission under the EU Merger Regulation. That said, a large number of markets still remain to be defined by competition authorities.135

8.68  Moreover, the calculation of the market shares poses a significant challenge. To compute market shares, parties need data on the total sales (for the supplier) and purchases (for the buyer) achieved on the relevant market.136 In principle, however, such data is unavailable to the parties, which do not—and should not—know the amount of sales and purchases achieved by their rival (and their customers/suppliers).137

(b)  Agreements between small and medium-sized enterprises

8.69  The Guidelines exhibit a noticeable degree of sympathy towards SMEs (undertakings which have fewer than 250 employees, and have either an annual turnover not exceeding €40 million or an annual balance-sheet total not exceeding €27 million).138 Vertical agreements between SMEs are deemed ‘rarely capable of appreciably affecting trade between Member States or of appreciably restricting competition within the meaning of Article 101(1) TFEU’, and thus generally fall outside the scope of Article 101(1).139 Should such agreements, however, satisfy the conditions for the application of Article 101(1) TFEU,

the Commission will normally refrain from opening proceedings for lack of sufficient interest for the European Union unless those undertakings collectively or individually hold a dominant position in a substantial part of the internal market.140

(c)  Block exemption mechanism

8.70  The third presumption of compatibility is the main feature of Regulation 330/2010.141 Vertical agreements which observe a dual set of market share thresholds (Section (i)), as well as a range of conditions (Section (ii)) are deemed automatically to fulfil the conditions for the application of Article 101(3).142 In some exceptional circumstances, the benefit of the block exemption may, however, be withdrawn (Section (iii)).

(p. 483) (i)  The double market-share thresholds

8.71  Pursuant to Article 3 of the Regulation, a vertical agreement is presumed to benefit from the Article 101(3) TFEU exception:

on condition that the market share held by the supplier does not exceed 30% of the relevant market on which it sells the contracts goods or services


the market share held by the buyer does not exceed 30% of the relevant market on which it purchases the contract goods or services.143

8.72  This ‘double market-share threshold’ is a novelty. Interestingly, in 1999 the Commission tried to introduce a similar mechanism. The proposal attracted, however, widespread stakeholder opposition (lawyers seemed reluctant to undertake what was seen as overly complex economic assessments) and the Commission eventually opted for a single market-share threshold .144 Ten years later, criticism of market share thresholds has faded. Firms and their legal counsel are often said to appreciate the legal certainty afforded by such ‘safe harbours’.145 Despite this, however, it remains open to question whether (i) market shares are good proxy for inter-brand competition146 and (ii) whether all firms contemplating the conclusion of a vertical agreement, and in particular small ones, enjoy sufficient expertise to undertake the intricate economic analysis (market definition and market share calculation) prescribed under Regulation 330/2010.

8.73  In substance, the reason behind the introduction of an additional market-share threshold is predicated upon a variant of the ‘buyer power’ (or monopsony) theory. As explained previously, in mainstream competition economics, buyer power is primarily viewed as a (p. 484) welfare-enhancing factor mitigating the effects of significant market power,147 often to the direct benefit of end consumers (particularly so when the buyer is a retailer).148

8.74  However, with the vast expansion of retail distribution and the rise of extremely large retailers,149 concerns over the exploitation of monopsonistic buyer power have become more acute. In the context of vertical agreements, large distributors may impose on suppliers low purchase prices, payment of listing fees, or other (non) price advantages (upfront access payments). Under the previous Regulation, such agreements automatically benefited from a presumptive exemption as long as the supplier’s market share did not exceed 30 per cent. To bring such agreements under in-depth competition law scrutiny, the new texts introduce an additional buyer’s market-share threshold. Only those agreements in which the buyer’s market share remains below 30 per cent are presumed to fulfil the Article 101(3) TFEU conditions. Other agreements, which possibly give rise to anticompetitive buyer power, must be subject to a full-blown, individual assessment. In other words, the new Regulation entails an extension of the substantive scope of EU competition law to new categories of agreements. This is further confirmed by the various new sections—discussed above—of the Guidelines devoted to up-front access payments and category management agreements.

8.75  That said, the economics of buyer power are far from settled.150 As indicated previously, since the works of J.K. Galbraith, mainstream economic theory views buyer power as a (p. 485) ‘countervailing’ factor, which leads generally to lower resale prices. In a recent study, E. Pfister asserts that ‘invariably, buyer power is considered a factor of competitive strength’.151 In contrast, the main theories of harm associated with buyer power—albeit intuitively and theoretically valid—have not been confirmed empirically.152

8.76  Moreover, those theories of harm often seem predicated upon disputable assumptions. With respect to category management agreements, for instance, the Guidelines state that they may result in anticompetitive foreclosure of other suppliers where the category captain is able to limit or disadvantage the distribution of products of competing suppliers.153 As other commentators have noted, it is however open to question why a retailer would allow a category captain to limit competition through the foreclosure of rival upstream suppliers.154 A retailer has no interest in limiting upstream distribution as it may result in increased input prices. Rather, the retailer, which strives to offer lower prices to end-users, may simply use a category management scheme as an incentive device to stimulate price competition amongst suppliers. In this setting, the retailer will appoint as the category captain the supplier which grants the largest price reductions. As long as equally efficient rival suppliers can compete for shelf management with the category captain, there is no foreclosure concern.155

8.77  Finally, the fact that additional market shares must be calculated raises an informational problem. Each party only enjoys ‘perfect’ information on its own market share (but not on the other’s). Of course, the parties can exchange information on their market shares. Yet, one cannot guarantee that the exchanged information is accurate. In this context, economic theory shows that in situations of information asymmetry, ‘moral hazard’ issues may arise. A retailer willing to conclude—or maintain—a vertical agreement at all costs may, for example, be tempted to share incorrect information with its potential supplier.156 In such case, however, both parties, including the one which acts honestly, may be held liable for infringing Article 101(1).

8.78  Because—from a legal policy perspective—normative standards should ideally be based on robust economic evidence, it would arguably have been wiser to (i) maintain the simple market share threshold of Regulation 2790/1999 for all agreements and (ii) provide, exceptionally, that when—in certain sectors—buyer power is likely to give rise to anticompetitive effects (eg through the exploitation of suppliers), the Commission and national competition authorities (NCAs) can withdraw the benefit of the block exemption.157

(p. 486) (ii)  Conditions

8.79  All agreements devoid of hardcore restrictions and which meet the double market-share threshold are deemed compatible with Article 101 TFEU.158 However, the parties’ contractual freedom is not absolute. Article 5 of Regulation 3320/2010 identifies three types of restraint which occasionally appear in vertical agreements, and which must comply with specific conditions. If these conditions are met, the restraint is deemed compatible, and this is the end of the self-assessment. If these conditions are not met, the restraint—and the restraint only—is deemed incompatible with Article 101(1) and must therefore undergo a full-blown competition analysis. The rest of the agreement remains, however, covered by the presumption of compatibility.159

8.80  Article 5 first targets ‘direct or indirect non-compete obligations’160 (ie, single-branding clauses and exclusive purchasing obligations).161 It provides that the block exemption only covers non-compete obligations for a period of no more than five years.162 Any such obligation with an indefinite duration; of more than five years; or tacitly renewable beyond a period of five years, is excluded from the benefit of the block exemption.163

8.81  Second, Article 5 focuses on ‘clauses prohibiting a buyer, after termination of the agreement, from manufacturing, purchasing, selling or reselling goods or services’.164 Such clauses are in principle not eligible for a block exemption.165 The Regulation does, however, provide for an exception to this if the clause is: (i) indispensable to protect know-how transferred by the supplier to the buyer; (ii) limited to the retail outlet from which the buyer has operated during the contract period; and (iii) limited to one year following the expiry of the agreement. In fact, this exception primarily concerns franchising agreements, where the franchisor transfers important trade secrets to the franchisee.

8.82  Third, Article 5 excludes from the block exemption clauses which impose ‘any direct or indirect obligation causing the members of a selective distribution system not to sell the brands of particular competing suppliers.’166 The purpose of this provision is to ensure that suppliers making use of selective distribution schemes do not foreclose access to specific competitors. In other words, the block exemption does not apply to practices which are akin to (p. 487) collective boycott.167 By contrast, the block exemption covers general non-compete obligations in the context of selective distribution networks.

(iii)  Withdrawal of the block exemption

8.83  Pursuant to Article 6 of Regulation 330/2010, the Commission is empowered to declare the Regulation—and in particular the block exemption—inapplicable to ‘vertical agreements containing specific restraints’. This exception applies only to situations ‘where parallel networks of similar vertical restraints cover more than 50% of a relevant market’.168 The Commission must issue a Regulation to this end.169

8.84  In such cases, the Commission will carry out a full competition analysis of the agreement under Article 101 TFEU.170 This provision seeks to avoid so-called type II errors, which occur when a rule fails to regulate conduct that harms consumer welfare (false negatives or false acquittals).

8.85  Finally, pursuant to Article 29(1) and (2) of Regulation 1/2003, the Commission and the NCAs can respectively withdraw the benefit of a block exemption in particular cases, if an agreement has effects incompatible with Article 101(3) TFEU.171 Insofar as NCAs are concerned, this is only possible if those effects ‘occur in the territory of that Member State, or in a part thereof, and where such territory has all the characteristics of a distinct geographic market.’172

B.  Full-Blown Competition Analysis of Vertical Restraints

(1)  Preliminary remarks

8.86  Agreements which fall short of the above-mentioned screening principles must be subject to a full-blown competition analysis.173 The principles governing the individual analysis of such agreements are laid down in the Guidelines. Importantly, those principles must be applied on a case-by-case basis and not mechanically.174

(p. 488) (2)  Method

8.87  Parties undertaking a full-blown competition assessment must first identify the anticompetitive features of their agreement, and the related theories of competitive harm (Section (a)). Second, the parties must test whether those theories of competitive harm are plausible in light of the market characteristics, and whether they are likely to give rise to a restriction of competition within the meaning of Article 101(1) TFEU (Section (b)). Third—and only if there is a restriction of competition under Article 101(1)—the parties must verify if their agreement produces efficiencies and objective justifications which trigger the benefit of an exemption under Article 101(3) (Section (c)).

8.88  In this section, we take, as a matter of illustration, a vertical agreement which contains a three-year single branding clause and involves a supplier controlling nearly 40 per cent of the relevant market.

(a)  Selection of the theories of harm

8.89  Not all vertical restraints have a similar effect on competition. Economic theory ascribes specific scenarios of competitive harm to the various types of vertical restraint. A prerequisite of any meticulous self-assessment is therefore to ‘frame’ the analysis by selecting a relevant theory of harm.

8.90  As explained previously, the Guidelines provide useful guidance for this self-assessment. For each category of vertical restraint, the Guidelines articulate a range of possible theories of harm. To take the example of single branding, the Guidelines particularly mention risks of (i) ‘foreclosure of the market to competing suppliers and potential suppliers’ (customer foreclosure);175 (ii) ‘collusion in the case of cumulative use by competing suppliers’ (supplier collusion);176 and (iii) a ‘loss of in-store inter-brand competition where the buyer is a retailer selling to final consumers’.177 We assume, here, that there is no cumulative use of single branding in the market and that the buyer is not a retailer. In such a case, none of the latter two assumptions is relevant. The individual analysis can thus focus exclusively on the risk of foreclosure.

(b)  Assessment of the theories of harm

8.91  The next step involves testing the plausibility of the theory of competitive harm in light of the market features. In assessing any potential risk resulting from a single-branding obligation the Guidelines consider that the following list of factors should be examined: the market position of the supplier,178 the needs of individual customers covered (100 per cent or less),179 the market coverage of the single branding commitment (the tied market share),180 (p. 489) the duration of the non-compete obligation,181 the market position of competitors,182 barriers to entry,183 countervailing power (or purchasing power),184 and the level of trade.185

8.92  The Guidelines provide details on those various factors. For instance, a single-branding obligation the duration of which is less than one year is deemed unlikely to generate anticompetitive effects. In contrast, single-branding obligations entered into by non-dominant companies for a duration of between one and five years will usually require a proper balancing of pro-and anticompetitive effects. At any rate, single-branding obligations are more likely to result in anticompetitive foreclosure when entered into by dominant companies.

8.93  The same applies to the tied market share. It is important to ascertain whether the point of sale that is foreclosed from rivals constitutes an important sales channel. If a supplier with a 40 per cent market share benefits from a single branding commitment with a customer that accounts for 10 per cent of its sales, the tied market share (ie, 4 per cent of the relevant market) is relatively limited. It is thus unlikely that the agreement restricts competition. If, however, the customer represents half of the sales of the supplier, the tied market share is much higher (ie, 20 per cent of the relevant market). Here, the restrictive effect of the agreement is serious.

8.94  In relation to the countervailing power of buyers, the Guidelines recognize that ‘powerful buyers will not easily allow themselves to be cut off from the supply of competing goods or services’.186 Buyers may, for instance, request financial compensation, through lower purchasing prices. In such cases, the Guidelines do not discard competition concerns. Rather, they stress that whilst this may be beneficial to certain individual buyers, it ‘would be wrong to conclude automatically from this that all single branding obligations, taken together, are overall beneficial for customers on that market and for the final consumers’.187

8.95  Finally, in relation to the level of trade, the Guidelines draw a dividing line between agreements concerning final products, where foreclosure is in general more likely ‘given the significant entry barriers for most manufacturers to start retail outlets just for their own products’.188 In contrast, the Guidelines take a more positive view of agreements concerning the supply of a final product at the wholesale level. They consider that there is ‘no real risk of anticompetitive foreclosure if competing manufacturers can easily establish their own wholesaling operation’.189

(p. 490) (c)  Efficiencies and objective justifications

8.96  Once a vertical agreement is found to create actual or likely anticompetitive effects, the parties need to verify whether this agreement can benefit from an individual exemption under Article 101(3) TFEU.190 As explained, vertical restraints may have a number of objective justifications and produce redeeming efficiencies. For each group of vertical restraint, the Guidelines provide clarifications on admissible objective justifications and pro-competitive efficiencies.

8.97  As far as single branding is concerned, the Guidelines recognize that the supplier may seek

(i) protection from free-riding by other suppliers (eg on promotional efforts);191 (ii) protection of a relation-specific investment made by the supplier (eg in equipment which can be used only to produce components for a particular buyer);192 (iii) protection from hold-up problems that may arise with the transfer of substantial know-how (which cannot be taken back by the supplier);193 or (iv) to overcome capital market imperfections (eg it is more efficient for the supplier to provide a loan than it is for a bank).194 Interestingly, the Guidelines also declare that quantity forcing is as equally efficient as single branding, but generates less restrictive effects on competition.195

8.98  With only very few exceptions, the Guidelines provide scant guidance on the level of sophistication required in the assessment of objective justifications and efficiency benefits.196 Parties should thus seek guidance from the Commission’s Guidelines on the application of Article 101(3) of the Treaty (the ‘General Guidelines’).

8.99  In practice, parties should proceed with caution. Competition authorities are often conservative when it comes to efficiency arguments. To be fair, their scepticism is understandable. Most objective justifications for vertical restraints do indeed hinge on behavioural speculations (incentives of buyers), rather than on structural efficiencies (quantitative costs savings). Such justifications are thus inevitably tainted with value judgment. To take a simple example, a supplier imposing a single-branding obligation may simply, but legitimately, over-estimate the risk of free-riding. Hence, it is imperative that firms conduct an objective and rigorous assessment of the possible efficiencies of their vertical agreements. In addition, firms should keep all supportive evidence of any efficiency benefits (should subsequent administrative or judicial proceedings be launched).

(p. 491) IV.  Online Distribution

A.  The Context

8.100  Up until the adoption of the new EU framework on vertical restraints, the issue of online distribution triggered intense debate.197 In essence, pure internet players (eg eBay),198 argued that under Regulation 2970/1990 firms operating selective distribution networks were free to undermine online distribution through various types of vertical restraint.199 For instance, it was reported that suppliers operating selective distribution systems had occasionally prohibited the setting up of a website, subordinated online sales to the observance of a recommended price, placed a cap on quantities sold through the internet, etc.200 In a document entitled ‘Empowering Consumers by Promoting Access to the 21st Century Market, A Call for Action’, eBay thus proposed to eradicate such restraints through a new hardcore restriction, whose proposed wording would read as follows:

The exemption provided for in Article 2 shall not apply to vertical agreements which, directly or indirectly, in isolation or in combination with other factors under the control of the parties, have as their object: … (f) the restriction of the ability of the buyer or any of his customers to sell the contract goods or services on the Internet without prejudice to the exceptions permitted under paragraph b.

8.101  Unsurprisingly, those proposals were fiercely challenged by suppliers of branded products (cosmetics, luxury goods, jewellery, watches, etc) which rely primarily on ‘brick and mortar’ shops for the distribution of their products. They argued in particular that pre-and post-sales services (advice, testing, etc) at the point of sale represent significant investments at both supplier and buyer levels. Members of such networks may in turn be reluctant to incur them, absent protection from internet distributors.201 This would be all the more problematic (p. 492) given the contribution of such services to demand growth.202 That said, for many products, free-riding appears in reality to take place the other way around (eg cars, electronics), with consumers first searching online, and then buying at a brick and mortar shop.203

8.102  Suppliers of branded products thus viewed the existing regulatory framework as satisfactory.204 They argued that the upcoming regulatory framework should confirm that selective distribution entails the freedom to condition online sales upon the existence of a physical outlet (and other qualitative requirements related, eg, to download rates, payment interfaces, search tools).205 In reality, suppliers operating a selective network seemed amenable to cumulative distribution (physical + online). Their primary area of concern related to pure internet distribution, which involves players without a physical presence.

8.103  The debate over the new regulatory framework also concerned other notions, such as the concept of passive and active sales in exclusive distribution networks. According to the existing framework, internet sales were generally deemed passive sales, which suppliers could not restrict. Yet, in light of technological progress, a number of suppliers argued that internet sales could no longer be regarded as passive, in particular when they entail a website targeted at specific customers, customer tracking, etc.206 Those suppliers thus argued that amendments to the regulatory framework needed to be introduced, so as to entitle suppliers to restrict certain internet sales.207

B.  The New Framework for Online Distribution

8.104  Remarkably, the new Regulation does not devote a single line to the issue of online distribution (as was the case with Regulation 2790/99). The issue is entirely left to the Guidelines which take a favourable stance vis-à-vis that form of distribution.208 Paragraph 52 of the Guidelines unambiguously declares that the internet is a powerful tool to reach a greater number and variety of customers than by more traditional sales methods, which explains why certain restrictions on the use of the internet are dealt with as (re)sales restrictions. In principle, every distributor must be allowed to use the internet to sell products…. (p. 493) This positive stance has direct consequences on the rules governing selective distribution (Section 1) and on the notions of active and passive sales (Section 2).

(1)  Selective distribution and online commerce

8.105  In conformity with Article 4(c) of the Regulation, the Guidelines declare that within a selective distribution system, dealers should be free to sell, both actively and passively, to all end-users, also with the help of the internet.209 That said, the Guidelines provide suppliers with some degree of control over internet sales. Just as they may require quality standards for brick and mortar shops, suppliers may require quality standards for the use of internet websites.210 Drawing inspiration from the solutions promoted by the French NCA and the Paris Court of Appeals,211 the Guidelines even acknowledge that a supplier may require its distributors to have one (or more) brick and mortar shop(s) as a condition for joining the distribution system. In the same vein, a supplier may request distributors that use third party platforms to sell their products, to display the logo and brands of the contractual product on the website.212

8.106  The Commission also considers as a ‘hardcore restriction’ any obligation which dissuades dealers from using the internet to reach a greater number (or variety) of customers by imposing criteria for online sales which are not equivalent to those imposed for sales in a brick and mortar shop.213 Importantly, this does not mean that the criteria for online and offline sales should be uniform.214 For example, in order to prevent sales to unauthorized dealers, a supplier may place a limit on the quantities sold by its selected dealers to an individual end-user. In such cases, the cap placed on sold quantities may have to be stricter for online sales, if it is easier for unauthorized dealers to obtain products through the internet.215

(2)  Active and passive sales

(a)  Internet sales are passive

8.107  The Guidelines seek to provide guidance on what constitutes a passive and active sale in the online world. As explained previously, the concept of active and passive sales is primarily relevant in relation to exclusive distribution. Suppliers can restrict a distributor’s freedom actively to sell products in a territory that has been granted to a different distributor. However, unsolicited, passive sales cannot be restricted.

8.108  In principle, setting up a website to sell a product is viewed as a passive sale, since it is deemed a reasonable way to allow customers to reach the distributor.216 Surely, the use of a website may have effects beyond the distributor’s own territory (or customer group). However, this stems from the technology itself, which allows easy access from everywhere.

(p. 494) 8.109  The Guidelines provide illustrations of passive online sales. If a customer visits the website of a distributor and contacts the distributor and if such contact leads to a sale, including delivery, then that sale is considered passive. The same holds true if a customer opts to be kept (automatically) informed by the distributor and this leads to a sale. Finally, the fact that a distributor offers different language options on its website (including languages not used in its territory) does not, of itself, alter the passive nature of the sale.217

8.110  Given, therefore, that the Guidelines consider internet sales to be passive sales, sales via the internet to another territory (or customer base) cannot be restricted on pain of falling within the presumption of incompatibility set out in Article 4 of the Regulation. In this context, the Guidelines provide four specific examples of hardcore restrictions of passive internet selling: (i) agreements according to which an (exclusive distributor) is required to prevent customers located in another exclusive territory from viewing its website or automatically to re-route its customers to the manufacturer’s or other (exclusive) distributors’ websites;218 (ii) agreements whereby an (exclusive) distributor is required to terminate an internet transaction if the credit card details reveal an address that is not within its exclusive territory;219 (iii) agreements which require that the distributor limit its proportion of overall sales made over the internet;220 and (iv) agreements whereby the buyer pays a higher price for products intended to be resold online (‘dual pricing’).221

(b)  In exceptional circumstances, internet sales may be considered active

8.111  The Guidelines merely lay down a presumption that internet sales are passive. In exceptional cases, internet sales may be considered active, and can thus be restricted. This is the case, for instance, if a distributor sends emails to consumers located in the exclusive territory of another distributor. Similarly, the Guidelines consider online advertisement specifically addressed to certain customers as a form of active selling to those customers.222 For instance, territory-based banners on third party websites are active sales into the territory where these banners are shown.223 Similarly, paying a search engine (or an online advertisement provider) to have advertisements displayed specifically to users in a particular territory is active selling into that territory.

(p. 495) V.  Conclusions

8.112  Whilst the recently adopted regulatory framework provides some useful guidance on inter-net distribution, it remains—to say the least—optimistic as regards the ability of firms to juggle with complex economic assessments, such as market definition and market share computation. In addition, it paints a bleak picture of buyer power, which (i) marks a departure from conventional antitrust economics; and (ii) relies on fragile and untested assumptions. Practice will tell whether those extensions of the EU rules on vertical restraints are to be welcomed.


1  See R. Coase, ‘The Nature of the Firm’ (1937) 4 Economica 386; O. Williamson, Market and Hierarchies: Analysis and Antitrust Implications (New York: Free Press, 1975). Economists sometimes speak of ‘hierarchy’ (the firm) and ‘delegation’ (the market) strategies. The choice of one or other method of distribution obviously varies according to the relevant sector, timing, and a company’s individual preferences. Eg with regard to the distribution of personal computers, though traditionally carried out using the independent distribution method, the past ten years have shown a growing trend towards integration. Clearly, the manufacturer may still decide to combine these two forms of distribution (‘dual distribution’). Economic research and, in particular, the ‘theory of the firm’ has identified the main determinants of such choice. According to R. Coase, recourse to independent distribution leads to transaction costs (or governance costs) in terms of research, negotiation, and performance problems. In contrast, under the integrated model, firms avoid (some) transaction costs as there exists a relationship of ‘authority’. However, their production costs increase. There is, therefore, a trade-off between transaction costs and production costs.

2  An agreement governing the relationship between the supplier and a distributor is generally termed a ‘vertical agreement’ because it involves the cooperation of non-competing undertakings active at different stages of the value chain. Distribution agreements between operators at the same level of the production process are horizontal agreements. The integrated distribution model also raises competition law issues, which are generally assessed under the merger control or abuse of dominance rules.

3  See Joined Cases 56 and 58/64 Etablissements Consten SARL and Grundig-Verkaufs GmbH v Commission, 13 July 1966 [1966] ECR 299.

4  A vertical agreement is defined in Art 1(1)(a) of Regulation 330/2010 as covering:

an agreement or concerted practice entered into between two or more undertakings each of which operates, for the purposes of the agreement or the concerted practice, at a different level of the production or distribution chain, and relating to the conditions under which the parties may purchase, sell or resell certain goods or services.

According to Art (1)(1)(b) of the Regulation a ‘vertical restraint’ means a restriction of competition in a vertical agreement falling within the scope of Art 101(1) TFEU—see Commission Regulation 330/2010 of 20 April 2010 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of vertical agreements and concerted practices, OJ L 102, 2010 (‘the Regulation’).

5  A large number of contractual clauses hampering competition between distributors of the same good or service (territorial exclusivity, etc) were thus prohibited.

6  During this early period, and for a considerable length of time thereafter, the assessment of vertical agreements was characterized by a complete lack of economic analysis. Agreements between parties that did not enjoy market power could thus be prohibited by the Commission without any meaningful economic analysis being performed to substantiate such findings.

7  See in this context S. Bishop, ‘Unfinished Business: The New Approach to Assessing Vertical Restraints’ (2002) 37(1) Intereconomics: Review of European Economic Policy 12.

8  See on this point J. Tirole, The Theory of Industrial Organization (Cambridge, MA: MIT Press, 1988), ch IV.

9  See B.E. Hawk, ‘System Failure: Vertical Restraints and EC Competition Law’ (1995) 35 Common Market L Rev 973.

10  Network industries subject to natural monopolies or oligopolies are typically an area where inter-brand competition is limited.

11  See F. Wijckmans, F. Tuytschaever, and A. Vanderelst, Vertical Agreements in EC Competition Law (Oxford: Oxford University Press, 2006), paras 1.51ff.

12  And more generally, marks the beginning of the ‘effects-based’ approach in EU competition enforcement.

13  From a legal practitioner’s point of view the lists of black, grey, and white clauses applicable under the former regime were quite convenient and their replacement by a system relying much more heavily on economic analysis may be somewhat disconcerting.

14  See Commission Notice—Guidelines on Vertical Restraints, OJ C 130, 2010, at 1 (‘Guidelines’), para 96.

15  With the internet’s rapid development and easy accessibility it became imperative to review the system of control of independent distribution agreements in order to align it with commercial realities.

16  See nn 4 and 14.

17  The Guidelines also devote some space to other types of vertical restraints, such as franchising at paras 189–91 and tying at paras 214–22. These issues will not be dealt with in the present chapter.

18  The idea here is to prevent the buyer from buying products of another brand. See Guidelines, n 14, at para 129. This explains why the Commission mentions exclusive purchasing in the same breath as vertical restrictions belonging to the exclusive distribution group. The concept of a non-compete obligation would appear to cover both single branding and exclusive purchasing.

19  The Commission still treats purchase quotas with a greater degree of leniency than exclusivity commitments. See eg Guidelines, n 14, at para 154.

20  See Guidelines, n 14, at paras 214–22.

21  Ibid, para 129.

22  Contracts with penalty clauses may create a risk of foreclosure and deter entry.

23  See GC, Case T-65/98 Van den Bergh Foods v Commission [1998] ECR II-2641.

24  See Guidelines, n 14, at para 192. In the Guidelines, exclusive supply is treated, in and of itself, as a type of vertical restraint.

25  This theory can be seen, in its strategic version, as a variation of the raising rivals’ costs theory, developed by Professor Salop. See V. Korah and D. O’Sullivan, Distribution Agreements under the EC Competition Rules (Oxford: Hart Publishing, 2002), 19. See T.G. Krattenmaker and S. Salop, ‘Anticompetitive Exclusion: Raising Rivals’ Costs to Achieve Power Over Price’ (1986) 96 Yale LJ 209.

26  See Guidelines, n 14, at para 151.

27  See N. Petit, Oligopoles, collusion tacite et droit communautaire de la concurrence (Brussels: Bruylant, 2007), chs I and IV.

28  See Guidelines, n 14, at para 130.

29  See B. Klein and K.M. Murphy, ‘Exclusive Dealing Intensifies Competition for Distribution’ (2008) 75 Antitrust LJ 433; G.F. Mathewson and R.A. Winter, ‘The Competitive Effects of Vertical Agreements: Comment’ (1987) 77 Am Economics Rev 1057–62; H.P. Marvel, ‘Exclusive Dealing’ (1982) 25 J Law and Economics 1; I.R. Segal and M.D. Whinston, ‘Exclusive Contracts and the Protection of Investments’ (2000) 31 RAND J Economics 603; B. Klein and A.V. Lerner, ‘The Expanded Economics of Free-Riding: How Exclusive Dealing Prevents Free-Riding and Creates Undivided Loyalty’ (2007) 74 Antitrust LJ 473; D. de Meza and M. Selvaggi, ‘Exclusive Contracts Foster Relationship-Specific Investment’ (2008) 38 RAND J Economics 85.

30  See P. Rey and J. Tirole, ‘The Logic of Vertical Restraints’ (1986) 76 Am Economics Rev 921.

31  A single-branding clause allows a supplier to ensure that its distributor focuses its efforts on marketing its product and only its product to the exclusion of the products of other suppliers. In addition, a distributor achieves economies of scale when it distributes a single product. See Tirole, n 8, at 185.

32  See H. Marvel, ‘Exclusive Dealing’ (1982) 25 J Law and Economics 1.

33  Not for so long as to hinder large-scale dissemination, however. See Guidelines, n 14, at para 107(c). Such benefits are more likely with ‘experience’ or complex goods that constitute expensive purchases for the final consumer.

34  See Guidelines, n 14, at para 107(d). For a discussion of the hold-up problem see O. Hart and J. Tirole, ‘Vertical Integration and Market Foreclosure’, Brookings Paper on Economics Activity: Microeconomics, 1990, at 205.

35  There are other ways to avoid this risk, eg by pooling the investment in a joint venture. See Korah and O’Sullivan, n 25, at 32.

36  A system of price thresholds or minimum prices has the same effect.

37  See Commission Press Release, 24 June 2002, ‘Commission clears B&W Loudspeakers distribution system after company deletes hard-core violations’, IP/02/916, Brussels.

38  The prohibition imposed on resale price maintenance is thus consistent with the priority accorded by the Commission to the fight against horizontal collusion. The Commission—via its Guidelines—would appear to agree that resale price maintenance has detrimental consequences on competition by stating that it maintenance reduces or even eliminates intra-brand competition. See, for arguments in favour of a per se prohibition, F. Van Doorn, ‘Resale Price Maintenance in EC Competition Law: The Need for a Standardised Approach’, 6 November 2009, available at <http://ssrn.com/abstract=1501070>.

39  Additionally, a system of maximum or recommended prices, in which resellers remain free to choose a lower or different price, acts as a focal point around which distributors can converge. See Guidelines, n 14, at para 227:

The possible competition risk of maximum and recommended prices is that they will work as a focal point for the resellers and might be followed by most or all of them and/or that maximum or recommended prices may soften collusion between suppliers.

In other words, the Commission sees in this a strategy based on revealing information to competing suppliers regarding the optimal price level to be reached.

40  See Guidelines, n 14, at para 224. See, in particular, B. Jullien and P. Rey, ‘Resale Price Maintenance and Collusion’ mimeo, University of Toulouse, 2002; M. Motta, Competition Policy—Theory and Practice (Cambridge, MA: Cambridge University Press, 2004). See also F. Mathewson and R. Winter, ‘The Law and Economics of Resale Price Maintenance’ (1998) 13 Rev of Industrial Organizaton 57.

41  Systems of resale price maintenance increase price transparency, a corollary of which is an increase in the risk of tacit collusion in concentrated markets. See Guidelines, n 14, at para 224.

42  See L. Telser, ‘Why Should Manufacturers Want Fair Trade’ (1960) 3 J Law and Economics 86.

43  See on these points, H. Hovenkamp, Federal Antitrust Policy—The Law of Competition and Its Practice, 2nd edn (St Paul, MN: West Group, 1999), paras 11.3–11.3.c.

44  In order to make cost savings. On this point, the Chicago authors recognize that control of resale prices eliminates price competition. However, such systems do lead to competition on other equally crucial parameters.

45  In certain markets, where both the supplier and buyer have market power, it is possible that supracompetitive profits are made twice. In this context one normally speaks of ‘double marginalization’. A supplier, wishing to protect its sales profits but also avoid a reduction in quantities sold at resale level, can control the profits of its distributor by fixing downstream prices. See Tirole, n 8, at 174.

46  See Guidelines, n 14, at para 174.

47  Conditions relating to personnel, premises, etc. The supplier may also impose an entry fee (eg in franchise systems). See Tirole, n 8, at 184.

48  Exclusive supply arrangements, which were discussed above, are drastic forms of limited distribution.

49  See Guidelines, n 14, at para 175.

50  Collusion between downstream buyers is facilitated because an understanding of the terms of coordination is simpler; monitoring activities are easier; and the risk of entry is reduced.

51  See J.J. Laff ont and D. Martimort, The Theory of Incentives: The Principal-Agent Model (Princeton, NJ: Princeton University Press, 2001).

52  Pre-sales services are recognized in Guidelines, n 14, at para 107(a).

53  See Tirole, n 8, at 183.

54  See Guidelines, n 14, at para 107(b), which states ‘to open up or enter new markets’.

55  Ibid, at para 107(a), which refers to the ‘free-rider problem’. Also note that the usual providers of capital (banks, equity markets) may provide capital sub-optimally when they have imperfect information on the quality of the borrower or there is an inadequate basis to secure the loan. Where the buyer provides the loan to the supplier, this may be the reason for exclusive supply or quantity forcing vis-à-vis the supplier. See Ibid, at para 107(h).

56  Ibid, para 107(c), referring to the ‘certification free-rider issue’.

57  Ibid, para 107(g).

59  See Korah and O’Sullivan, n 25, at 37, noting that such considerations were at the heart of the old laws which have since been abolished.

60  See Guidelines, n 14, at para 174.

61  In this type of distribution network, it is common to prohibit the resale of goods/services to unauthorized distributors outside of the network.

62  See Guidelines, n 14, at para 168. These territorial systems are more frequent when the downstream party is a wholesaler (as opposed to a retailer).

63  See eg the many cases concerning exclusive supply clauses in the beer industry: CJ, C-23/67 SA Brasserie de Haecht v Consorts Wilkin-Janssen [1967] ECR 525; CJ, C-234/89 Delimitis [1991] ECR I-935.

64  However, unlike single branding, it retains the ability to buy and sell competing products. Single branding and exclusive purchasing are often grouped together under the concept of non-compete obligations.

65  On exclusive distribution, see Section II(B).

66  The supplier may also restrict its distributors’ ability to resell products/services to certain types of customers (eg business customers or private clients).

67  ‘Active’ sales mean actively approaching individual customers inside another distributor’s exclusive territory or exclusive customer group by for instance direct mail or visits; or actively approaching a specific customer group or customers in a specific territory allocated exclusively to another distributor through advertisement in media or other promotions specifically targeted at that customer group or targeted at customers in that territory; or establishing a warehouse or distribution outlet in another distributor’s exclusive territory—see Guidelines, n 14, at para 50.

68  ‘Passive’ sales mean responding to unsolicited requests from individual customers including delivery of goods or services to such customers. General advertising or promotion in media or on the Internet that reaches customers in other distributors’ exclusive territories or customer groups but which is a reasonable way to reach customers outside those territories or customer groups, for instance to reach customers in non-exclusive territories or in one’s own territory, are passive sales—see Guidelines, n 14, at para 50.

69  See Commission Decision of 30 October 2002, Video Games, Nintendo Distribution, OJ L 255, 2003, at 33.

70  In this case, there were striking price differences across Europe. According to the Commission, in early 1996, some Nintendo products were up to 65 per cent cheaper in the UK than in the Netherlands and Germany.

71  Limited by inter-brand competition, both from other goods or services.

72  See Guidelines, n 14, at para 100(d).

73  Assuming that several distributors of a supplier are in a position to compete, one means of control at the supplier’s disposal is to segment the market between the relevant distributors by carving up sales territories. See Tirole, n 8, at 472.

74  See R. Inderst and G. Shaffer in ‘Buyer Power in Merger Control’ in W.D. Collins (ed), ABA Antitrust Section Handbook, Issues in Competition Law and Policy, who define buyer power as ‘the ability of buyers to obtain advantageous terms of trade from their suppliers’.

75  See Guidelines, n 14, at para 116.

76  Such that it resulted in a failure to provide an optimum product mix, which would have maximized the overall profitability of its biscuit range. See UK OFT decision of 10 September 2004 (Anticipated acquisition by United Biscuits (UK) Ltd of the Jacobs Bakery Ltd) where it was also stated that category management involves a leading supplier providing expertise to the retailer to help it to maximize the profitability of each of its product ranges. This may involve, eg, providing research on the best way to market products and at what time.

77  See Guidelines, n 14, at para 204.

78  Ibid, at para 205.

79  Ibid, at para 206. For a discussion of slotting allowances and their impact on the competitive process, see O. Foros and H.J. Kind, ‘Do Slotting Allowances Harm Retail Competition’ CESIFO Working Paper No 1800, Industrial Organisation, September 2006.

80  See Guidelines, n 14, at para 210.

81  Ibid, at para 211.

82  The French NCA has issued an opinion on this subject, see <http://www.autoritedelaconcurrence.fr/user/avisdec.php?numero=10-A-25>.

83  See Guidelines, n 14, at para 208.

84  Ibid, at para. 213.

85  For some interesting proposals on this point see, see P. Lugaard and J. Haan, ‘Ten Points to Consider when Reviewing Regulation 2790/1999’ (2009) Global Competition Policy Online, March.

86  See Guidelines, n 14, at para 110.

87  Ibid: ‘First, the undertakings involved need to establish the market shares of the supplier and the buyer on the market where they respectively sell and purchase the contract products’.

88  See Art 2(1) of the Regulation, n 4. We therefore do not discuss agreements that fall outside the scope of the Regulation and the Guidelines, ie: (i) agreements between competing firms (with the exception of section 2(4) of the Regulation on non-reciprocal vertical agreements); (ii) agreements concluded within the framework of an association of retailers of goods (other than section 2(2) relating to relations between the association and its members); (iii) vertical agreements falling within a specific block exemption (see Art 2(5) of the Regulation), eg motor vehicle distribution agreements; and (iv) agreements pertaining to leases and rental agreements (see Guidelines, n 14, at para 26). We also exclude real agency contracts. A ‘real’ agency contract does not fall within the purview of Art 101(1) TFEU. A ‘false’ agency contract does on the other hand fall within the ambit of Art 101 TFEU. The dividing line between these two types of contract is drawn by reference to the criterion of imputability of financial and commercial risks associated with the contract. A real agency contract is one by which the agent bears no financial risk. Conversely, a false agency contract is a contract where the financial risk and the risks associated with the non-performance of contractual obligations is imputed to the agent. See the Guidelines, n 14, at paras 13–16. Article 101(1) TFEU is applicable to agency contracts when the agent assumes one or more of the following risks: the agent contributes to the costs associated with the supply of the goods or services (eg transport costs), the agent invests in promotional activities, the agent sets up and operates at its own expense an after-sales service or warranty system, the agent makes market-specific investments in equipment, facilities, or staff training; and the agent assumes liability vis-à-vis third parties for products sold, the agent assumes responsibility for the non-performance of the contract by the customer etc.

90  They are of such gravity that the illegality of the clause in question affects the validity of the entire agreement, even where the market-share threshold (to which we shall return) is not exceeded. See the Guidelines, n 14, at paras 47–59. Such restrictions are not severable from the remainder of the agreement. This is important because other restrictions deemed inconsistent with Art 101(1) TFEU (and not covered by Art 101(3)) after an individual assessment remain severable from the agreement (these restrictions are mentioned in Art 5 of the Regulation). Only such clauses are incompatible and the remainder of the agreement survives. See our comments below.

91  See Guidelines, n 14, at para 47.

92  See eBay’s ‘Empowering Consumers by Promoting Access to the 21st Century Market, A Call for Action’, p 11.

93  See LVMH submission concerning the review of the EU competition rules applicable to vertical restraints of 24 September 2009. The luxury product industry generally considers that the growth of online sales of luxury products, the continual adoption of new technologies for internet advertising and sales, make it impossible to adopt exhaustive legislation in this context.

94  See Decision of the French Conseil de la Concurrence, Decision no 08-D-25 of 29 October 2008 and the Paris Court of Appeals judgment of 29 October 2009. The Paris Court of Appeals referred to the Court of Justice a request for a preliminary ruling in C-439/09 Pierre Fabre Dermo Cosmétique SAS.

95  Understood as the objective ability of the agreement and not the intention of the parties. See Art 4(a) of the Regulation and the Guidelines above, n 14, at para 48. For a discussion of RPM see O. Foros, H.J. Kind, and G. Schaffer, ‘Resale Price Maintenance and Restrictions on Dominant Firm and Industry-Wide Adoption’, CESifo Working Paper Series No 2032, 2007; see also V. Verouden, ‘Vertical Agreements: Motivation and Impact in Issues in Competition Law and Policy’ in W.D. Collins (ed), ABA Section of Antitrust Law (2008). See also F. Alese, ‘Unmasking the Masquerade of Vertical Price Fixing’ (2007) 28 European Competition L Rev 514 and M. Kneepkens ‘Resale Price Maintenance: Economics Call for a More Balanced approach’ (2007) 28 European Competition L Rev 656.

96  See CJ, C-243/83 SA Binon & Cie v SA Agence et messageries de la presse [1985] ECR 2015; CJ, 107/82 AEG/Telefunken v Commission [1983] ECR 3151.

97  For an example of a clause permitting an undertaking to scrutinize the wording of dealers’ advertisements as regards selling prices and to prohibit such advertisements, see CJ, C-86/82 Hasselblad v Commission [1984] ECR 883.

98  See in particular Commission Decision of 16 July 2003, Case COMP/37.975 Po/Yamaha, paras 81–2 and 144, not yet published.

99  The prohibition extends to mechanisms that ensure the monitoring and detection of distributors who do not respect the set price. The existence of such mechanism raises suspicion of a concerted practice or vertical price fixing. It may be, eg, an obligation imposed on a distributor to denounce other distributors who depart from the standard price. See also with regard to agency the special case pertaining to a prohibition on the agent to share its commission with the customer. See Guidelines, n 14, at para 48, codifying CJ, C-311/85 Vereniging van Vlaamse Reisbureaus [1987] ECR 3801.

100  See Commission Decision of 5 July 2000, Nathan Bricolux, OJ L 54, 2001, at 1, para 87.

101  The GC has drawn a distinction between mere price recommendations and the imposition of strict rules relating to retail prices: see GC, T-67/01 JCB Service v Commission [2004] ECR II-49. See CJ, C-191/84 Pronuptia de Paris GmbH v Pronuptia de Paris Irmgard Schillgallis [1986] ECR 353, at 25.

102  See Guidelines, n 14, at para 48. And more generally, threats, intimidation, sanctions, suspensions. or delivery delays discouraging a buyer from deviating from a recommended price.

103  US Supreme Court, Leegin Creative Leather Products, Inc v PSKS, Inc 127 S Ct 2705 (2007). For further discussion on this subject see W.S. Grimes, ‘The Path Forward after Leegin: Seeking Consensus Reform of the Antitrust Laws of Vertical Restraints’ (2008) 75 Antitrust LJ 467.

104  Which had been established in USC, Dr Miles Medical Co v John D Park & Sons Co, 220 US 373 (1911).

105  See USC, Standard Oil Co v United States, 221 US 59–68 (1911).

107  See Guidelines, n 14, at para 223. See also para 225: ‘RPM may not only restrict competition but may also, in particular where it is supplier-driven, lead to efficiencies, which will be assessed under Article 101(3).’

108  The Commission, at a roundtable discussion organized by the OECD in 2008, seemed prepared to explore all options within the framework of its review of vertical restraints. The Commission indicated that the 1999 texts did not necessarily reflect its state of thinking on the issue. However, the Commission did express its doubts regarding claimed efficiencies arising out of resale price maintenance practices. OECD Competition Committee, Paris, 21–23 October 2008.

109  See Guidelines, n 14, at para 225.

110  Ibid.

111  Ibid.

112  Ibid. See also E. Gippini-Fournier, ‘Resale Price Maintenance in the EU: In Statu Quo Ante Bellum?’ in B. Hawk (ed), 36th Annual Conference on International Antitrust Law and Policy 2009: Fordham Corporate Law Institute (London: Sweet & Maxwell, 2010).

113  See Art 4(b) of Regulation 330/2010, n 4.

114  See Guidelines, n 14, at para 50. The underlying rationale of such incompatibility is that buyers must be able to sell anywhere they wish.

115  For other examples see Guidelines, n 14, at para 48.

116  See Art 4(b) of Regulation 330/2010, n 4, and Guidelines, n 14, at para 51. These exceptions do not mean that the relevant restrictions are valid but that they can, if the relevant conditions are met, qualify for an exemption.

117  See Guidelines, n 14, at para 51. See also Commission Decision, Video Games, Nintendo Distribution, n 69, at para 331.

118  See discussion in greater detail under Section IV.

119  See Guidelines, n 14, at para 55.

120  See Art 4(c) of Regulation 330/2010, n 4, and Guidelines, n 14, at para 55.

121  See Guidelines, n 14, at para 57.

122  Ibid.

123  See Art 4(d) of Regulation 330/2010, n 4, and Guidelines, n 14, at para 55.

124  See Guidelines, n 14, at para 58.

125  Ibid, at para 59.

126  It may also seek to prevent mere repairs/maintenance of its product.

127  See Guidelines, n 14, at para 59. The supplier may, however, impose on its own repair and maintenance network an obligation to purchase spare parts from it, and lay down a prohibition on dealing directly with the component manufacturer.

128  See CJ, C-5/69 Franz Völk v SPRL Ets J Vervaecke [1969] ECR 295 at 7.

129  See Commission Notice on agreements of minor importance which do not appreciably restrict competition under Article 81, paragraph 1, of the Treaty establishing the European Community (de minimis Notice), OJ C 368, 2002, at 13.

130  Ibid, at para 7.

131  Ibid, at para 11(2).

132  See Guidelines, n 14, at para 88.

133  See Commission Notice on the definition of relevant market for the purposes of Community competition law, OJ C 372, 1997, at 5, para 7; and see Chapter 4.

134  Ibid, at para 8.

135  And where a relevant market analysis has taken place, it may relate to a different/obsolete economic context.

136  See Art 7(a) of Regulation 330/2010, n 4. In the absence of reliable data on the value of sales, it is possible to rely on ‘estimates based on other reliable information concerning the market, including market sales volume’. The market share should be calculated on the basis of data for the preceding calendar year. See Art 7(b) of Regulation 330/2010, n 4.

137  See P.M. Louis, ‘Le nouveau règlement d’exemption par catégorie des accords de transfert de technologie: une modernisation et une simplification’ (2004) 3-4 Cahiers de droit européen 377, 385 and 403.

138  See Annex to Commission Recommendation 96/280/EC.

139  See Guidelines, n 14, at para 11.

140  Ibid.

141  See Guidelines, n 14, at para 23, which refers to a ‘presumption of legality’.

142  Technically, the Regulation relies on Art 101(3) TFEU to declare Art 101(1) TFEU inapplicable (Art 2 of Regulation 330/2010, n 4, recalls that a finding of inapplicability of Art 101(1) TFEU finds its origin in the application of Art 101(3)). The Regulation assumes that the conditions for exemption under Art 101(3) TFEU are met. Recital 8 of the Preamble to the Regulation states that agreements that do not exceed the market-share threshold ‘generally lead to an improvement in production or distribution and allow consumers a fair share of the resulting benefits’.

143  In the event that the market share exceeds the threshold by 5 per cent at the end of the initial assessment of the agreement, the Regulation (Art 7) provisionally allows the exemption to remain in place. If the market share is less than 35 per cent, the agreement can benefit from the exemption for a period of two years. If the market share is greater than 35 per cent, the exemption is only valid for one year.

144  Under the legal regime in place under Regulation 2790/1999 the 30 per cent market-share rule generally only applied to the supplier (except in cases of exclusive supply).

145  The term ‘safe harbour’ is borrowed from Guidelines, n 14, at para 23.

146  It is open to debate whether, from an economic point of view, the market-share thresholds constitute an effective screening mechanism for assessing vertical agreements. The market-share thresholds are based on a structural reasoning, which borrows heavily from the teachings of the Harvard School (in particular, the Structure Conduct Performance (SCP) correlation between market share, market power, and supra-competitive prices). For an illustration of such structuralism see recital 4 of the Preamble to the Regulation, n 4, which states that an individual assessment requires that ‘account [be] taken of several factors, and in particular the market structure on the supply and purchase side’. In a market of differentiated products (economists speak of monopolistic competition), market shares below 30 per cent are not incompatible with the existence of significant market power. Moreover, it is open to question whether a rule based on the market shares held by the parties is really that effective for assessing the risks of oligopolistic tacit collusion. In fact, tacit collusion is a rare economic phenomenon which requires that—in addition to the presence of a tight oligopoly—four cumulative conditions be met: mutual understanding of the terms of coordination (C1), detection of any risks of deviation (C2), the existence of a retaliatory mechanism (C3), and an absence of any challenge by forces that are exogenous to the oligopoly (C4). The issue of market share held by each party is of no particular importance. A situation of oligopolistic tacit collusion is indeed possible below the relevant thresholds. On the other hand, even in cases above the relevant thresholds it is possible that there is no risk of tacit collusion—if one of the conditions is not satisfied (C1 and C2, eg where the market is not transparent). One may therefore question the need to carry out a complex market-share threshold assessment with regard to the risks of collusion, while the agreement could simply be screened via a check against these four conditions.

147  See, eg, the Communication from the Commission—Guidance on the Commission’s enforcement priorities in applying Article 82 of the EC Treaty to abusive exclusionary conduct by dominant undertakings, OJ C 45, 2009, at 7, para 18:

Competitive constraints may be exerted not only by actual or potential competitors but also by customers. Even an undertaking with a high market share may not be able to act to an appreciable extent independently of customers with sufficient bargaining strength. Such countervailing buying power may result from the customers’ size or their commercial significance for the dominant undertaking, and their ability to switch quickly to competing suppliers, to promote new entry or to vertically integrate, and to credibly threaten to do so. If countervailing power is of a sufficient magnitude, it may deter or defeat an attempt by the undertaking to profitably increase prices. Buyer power may not, however, be considered a sufficiently effective constraint if it only ensures that a particular or limited segment of customers is shielded from the market power of the dominant undertaking.

See also the draft Guidelines at para 116:

In some circumstances buyer power may prevent the parties from exercising market power and thereby solve a competition problem that would otherwise have existed. This is particularly so when strong customers have the capacity and incentive to bring new sources of supply on to the market in the case of a small but permanent increase in relative prices.

148  In this sense, there has been only little, if no enforcement, of Arts 101 and 102 TFEU against monopsonistic practices. See CJ, C-95/04 P British Airways v Commission, 15 March 2007 [2007] ECR I-2331 where there was a dominant position on a purchasing market. See also the draft Guidelines on the applicability of Article 101 of the Treaty on the Functioning of the European Union to horizontal cooperation agreements, Brussels, SEC(2010) 528/2 which devote a full section to the competitive risks associated with collective purchasing.

149  See Dobson Consulting, ‘Buyer Power and Its Impact on Competition in the Food Retail Distribution Sector of the European Union’, prepared for the European Commission—DGIV Study Contract No IV/98/ETD/078 who reports that the largest Belgian company, Delhaize ‘Le Lion’, is a retailer, Britain’s Tesco and J. Sainsbury both appear in the UK top 10 companies; Germany has the giant Metro group; whilst Wal-Mart Stores, number four in the US, is the eighth largest company in the world with US$119bn turnover and 825,000 employees (Fortune, 3/8/98).

150  For a discussion of buyer power see eg R. Inderst and C. Wey, ‘Buyer Power and Supplier Incentives’ (2007) 51 European Economic Rev 647; see also R. Inderst, ‘Leveraging Buyer Power’ (2007) 25 Int’l J Industrial Organisation 908.

151  See E. Pfister ‘Buying Power and Competition Policy’ [2009] 1 Concurrences 34, para. 45.

152  Ibid, at para. 46. To the best of our knowledge, there is no case where the Commission has grappled with the anticompetitive effects of up-front access payments and category management agreements.

153  See Guidelines, n 14, at para 210.

154  See P. Lugaard and T. van Dijk, ‘The New EC Block Exemption for Vertical Restraints: A Step Forward and a Missed Opportunity, Global Competition Policy Online, June 2010, at 5.

155  In contrast, there may well be exploitation concerns. However, competition authorities across the EU often consider that such risks should not be dealt with as a matter of priority.

156  Besides which, the system envisaged by the Regulation leads to an increased flow of commercially sensitive information at the distribution stage insofar as the distributor must, in order to determine its market share, know how much sales are made by its competitors.

157  Under the former system, if the manufacturer held a market share below 30 per cent—but the market share held by the distributor was above this threshold—the Commission had to use the individual withdrawal mechanism and was required to satisfy the heavy burden of proof for the application of Art 101(1) TFEU. The introduction of a 30 per cent market-share threshold in relation to distributors has the effect of freeing the Commission from this burden of proof and allows it to deny the benefit of a block exemption for this type of agreement.

158  In such circumstances, the parties can craft their vertical agreement as they see fit.

159  The clauses in question are severable from the rest of the agreement. They themselves cannot benefit from the exemption. See Guidelines, n 14, at para 71.

160  See Art 1(d) of Regulation 330/2010, n 4:

any direct or indirect obligation causing the buyer not to manufacture, purchase, sell or resell goods or services which compete with the contract goods or services, or any direct or indirect obligation on the buyer to purchase from the supplier or from another undertaking designated by the supplier more than 80% of the buyer’s total purchases of the contracts goods or services and their substitutes on the relevant market, calculated on the basis of the value or, where such is standard industry practice, the volume of its purchases in the preceding calendar year.

161  Ibid.

162  Either its duration does not exceed five years, or its renewal beyond five years requires the express consent of both parties.

163  The Regulation provides for a derogation from the maximum duration of five years when the contract goods or services are sold from premises and land which the vendor owns. As long as the buyer occupies the premises, a non-compete obligation is justified. See Art 5(2)(a) of Regulation 330/2010, n 4.

164  See Ibid, Art 5(1)(b) and Guidelines, n 14, at para 68.

165  See Guidelines, n 14, at para 60.

166  See Art 5(1)(c) of Regulation 330/2010, n 4, and Guidelines, n 14, at para 69.

167  See Guidelines, n 14, at para 69.

168  See Wijckmans et al, n 11, at para 9.08. Using the example of a market on which there are four competing manufacturers. Each has a market share of 25 per cent and imposes on its respective distributors an identical single-branding clause. The cumulative effect of this network of agreements entirely forecloses new entrants. However, an assessment pursuant to the Regulation would have found the agreement valid because (i) the relevant thresholds are respected and (ii) single branding does not as such constitute a black clause.

169  See Art 6 of Regulation 330/2010, n 4. The ability to do this is based on Art 7 of Regulation 17/65, which has now been replaced by Art 29(1) of Council Regulation (EC) No 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, OJ L 1, 2003, at 1.

170  See, for an illustration, Commission Decision of 23 December 1992, Langnese Iglo GmbH, OJ L 183, 1993, at 19 and Commission Decision of 23 December 1992, Schöller Lebensmittel GmbH Co KG, OJ L 183, 1993, at 1.

171  See paras 13 and 14 of Regulation 330/2010, n 4.

172  See also Art 29(2) of Regulation 1/2003. See Guidelines, n 14, at para 78. In this situation the Commission is also able to intervene if the case raises a particular interest such as the novelty of the relevant issues (see para 80).

173  Those are either (i) agreements where one of the parties has a market share in excess of 30 per cent or (ii) restrictions introduced into agreements which do not meet the requirements of Art 5 of the Regulation.

174  See recital 3 of the Preamble to the Guidelines.

175  See Guidelines, n 14, at para 130.

176  Ibid.

177  Ibid.

178  Ibid, at para 132.

179  Ibid, at para 133.

180  Ibid.

181  Ibid, at para 133:

Single branding obligations shorter than one year entered into by non-dominant companies are generally not considered to give rise to appreciable anti-competitive effects or net negative effects. Single branding obligations between one and five years entered into by non-dominant companies usually require a proper balancing of pro-and anti-competitive effects, while single branding obligations exceeding five years are for most types of investments not considered necessary to achieve the claimed efficiencies or the efficiencies are not sufficient to outweigh their foreclosure effect.

182  Ibid, at para 134.

183  Ibid, at para 136.

184  Ibid, at para 137.

185  Ibid, at para 138.

186  Ibid, at para 137.

187  Ibid.

188  Ibid, at para 140

189  Ibid, at para 139. This will in turn depend on the type of product.

190  On condition that that the supplier does not occupy a dominant position. See Ibid, at para 127.

191  See Ibid, at para 144. It is of course understandable that a supplier who has funded significant promotional efforts may seek to defend itself against free riders. In this context, the appropriate vertical restraint will be (i) of the non-compete type or quantity-forcing type when the investment is made by the supplier and (ii) of the exclusive distribution, exclusive customer allocation, or exclusive supply type when the investment is made by the buyer.

192  See Ibid, at para 146. In the same vein, in the case of client-specific investments where, eg, an investment made by the supplier—after termination of the agreement—cannot be used by the supplier to supply other customers and can only be sold at a significant loss, a single branding obligation covering the amortization period of the investment is likely to meet the conditions of Art 101(3) TFEU.

193  Ibid, at para 148.

194  Ibid, at para 147.

195  Ibid, at para 145.

196  The Guidelines nonetheless provide two examples of self-assessment for non-compete obligations and quantity forcing.

197  The debate on this issue started relatively early—even before the formal review process of Regulation 2790/1999 began—when the Competition Commissioner set up in 2008 a roundtable entrusted with discussing the future of online commerce. The work of the roundtable centred on the question of distribution via the internet of audiovisual content protected by intellectual property rights (the iTunes case showed that European consumers could not be freely supplied throughout the EU for copyright reasons). However, the roundtable also intended to discuss the online commerce of goods/services which are not protected by such rights. Following the publication of an ‘Issues paper’ containing a list of questions for interested parties, operators active in the sector –whether internet or physical operators—gave their contribution to the public debate. The majority of the contributions submitted tackled the issue of a possible review of the legal framework laid down by Regulation 2790/1999 and the Guidelines. These comments gave impetus to the debate on the issue of vertical restraints and distribution via the internet. A further round of consultation was subsequently organized during the formal review of Regulation 2790/1999. The debates that took place gave rise to a wealth of additional contributions, which can be found at <http://ec.europa.eu/competition/consultations/2009_vertical_agreements/index.html>. Most of the contributions quoted in the following footnotes can be found at this URL address.

198  Others players in this context are <www.rueducommerce.fr>, Interactive Software Federation of Europe, Amazon, etc.

199  According to eBay, selective distribution networks could be used in an anticompetitive manner, eg by rendering on line distribution economically unattractive.

200  See K. Mahlstein, ‘Vertical Restraints and Competition Policy—Internet Sales, a New Dimension to be Considered’, Global Competition Policy Online, March 2009; S. Kinsella and H. Melin, ‘Who’s Afraid of the Internet? Time to Put Consumer Interests at the Heart of Competition’, Global Competition Policy Online, March 2009.

201  See on this the contribution made by the Federation of the Swiss Watch Industry, at 1–2.

202  See the LVMH contribution, at 38 and 40. This is in particular true for a ‘touch and feel’ product (cosmetics, etc), which customers necessarily test in a brick and mortar shop prior to purchase.

203  See D. Carlton and J. Chevalier, ‘Free Riding and Sales Strategies for the Internet’, NBER Working Paper 8067, 2001; Y. Bakos, ‘The Emerging Landscape for Retail E-Commerce’ (2001) 15(1) J Economic Perspectives 69–80.

204  According to Estée Lauder in its contribution, the legal framework authorized the exclusion of ‘pure players’ from selective distribution networks.

205  See the LVMH contribution. This solution draws inspiration from French competition law where the supplier can require from its physical distributors that wish to make online sales that they (i) already have a physical infrastructure in place which meets relevant qualitative criteria and (ii) make the necessary investments so that their website has the requisite level of prestige as required within the network, etc.

206  Chanel in its contribution considers that sales via the internet should not be considered passive sales but active sales. See also the contribution made by the Premier League which challenges the relevance of the distinction between active and passive sales in the field of internet distribution.

207  In contrast see the contribution made by Which. This consumer organization considers that there have been few cross-border internet sales to the detriment of consumers.

208  This would appear to make sense as the Commission cannot rely on any legal precedent in this context. There is in fact no European case law or decisional practice on the issue of the prohibition of online sales.

209  See Guidelines, n 14, at para 56.

210  Or for selling by catalogue or for advertising and promotion in general. Ibid, at para 54.

211  See n 94.

212  See Guidelines, n 14, at para 54.

213  Ibid, at para 56.

214  In practice, this restriction is likely to give rise to significant interpretation problems.

215  See Guidelines, n 14, at para 56.

216  Ibid, at para 52.

217  Ibid, at para 51

218  Ibid, at para 52(a).

219  Ibid, at para 52(b).

220  Ibid, at para 52(c). However, a supplier may—without limiting the online sales of the distributor—require that the buyer sells at least a certain absolute amount (in value or volume) of the products offline to ensure an efficient operation of its brick and mortar shop.

221  Ibid, paras 52(d) and 64. This does not exclude the situation whereby the supplier agrees with the buyer a fixed fee (ie, not a variable fee where the sum increases with the offline turnover as this would amount indirectly to dual pricing) to support the latter’s offline or online sales efforts. With regard to this latter restriction the Commission does consider, however, that in some specific circumstances such an agreement may fulfil the conditions of Art 101(3). Such circumstances may be present where a manufacturer agrees such dual pricing with its distributors because selling online leads to substantially higher costs for the manufacturer than offline sales. The Commission provides the example of a situation in which this may be the case: where offline sales include home installation by the distributor but online sales do not, the latter may lead to more customer complaints and warranty claims for the manufacturer. See also in this context the judgment handed down by the Rechtbank Zutphen, 8 August 2007, 79005/HA ZA 06-716. The Dutch court held that a dual pricing scheme pursuant to which a supplier of built-in kitchen equipment offered less attractive pricing conditions to online distributors did not infringe Dutch or EU competition laws.

222  See Guidelines, n 14, at para 53.

223  In general, efforts to be found specifically in a certain territory or by a certain customer group is active selling into that territory or to that customer group.(p. 496)