Summary of EC Competition Law (Monti)

This Summary of EC Competition Law (Monti) is written in 2015 and covers all major subjects of competition law

Chapter 1: The main values of EU competition law

1.1 Introduction

EU competition policy is undergoing a significant shift in economic theory and political ideology as well as in its institutional enforcement structure. In this chapter you will see that, from a political perspective, competition policy in the EU originally promoted core values: competition, the integration of the internal market and economic efficiency. Today economic efficiency and consumer welfare play a more significant role in motivating competition law enforcement than other values. The role of economic analysis in resolving competition cases takes centre stage. Author suggests that three themes underpin the economic approach favoured by the Community: 1) consumer welfare, 2) market power and 3) pluralism. In the following chapters author explores the extent to which this economic paradigm is activated in the ECJ case law. Significant number of public policy considerations affect decisions in competition cases. The transformation of the political and economic paradigms began in the 1990s, but has accelerated in recent years. The reason for the speeding up is that institutionally, the focus of enforcement is shifting from the Commission and towards National Competition Authorities, regulatory agencies and courts. These institutional changes are brought about by Regulation 1/2003. The exclusive application of economics to resolve competition cases can ensure greater coherence among institutions enforcing the competition law.

1.2 Competition law and economic freedom

There are three conceptions of competition. First, competition may exist when there is rivalry among f undertaking s. Rivalry among undertakings is the means through which a number of socially desirable ends (economic efficiency, economic freedom or consumer welfare) are pursued. To see if there is a distortion of competition you need to look to see whether the ends are met rather than whether undertaking s are rivals, because this way provides a more precise method to determine whether there is a market failure. The reason for this is that undertaking s invent new ways of competing every day. Competition law does not prevent some forms of cooperation among undertaking (mergers), that's why rivalry as the benchmark of competition is incomplete.

Accordingly, most economists support a second conception of competition based on the effects of the behaviour of undertakings on economic welfare. This is preferable for two reasons: first, it provides a realistic benchmark by which to measure the presence of competition. Second, it is more precise, because there can be rivalry but no competition.

The question about whether there is competition is not whether the market is characterized by rivalry, but whether the market yields economic welfare. Defining competition by judging the effects on economic welfare (the neoclassical concept) is traditionally associated with economic approaches to competition law.

A third approach is the economic freedom conception of competition, whose roots you can find in ordoliberalism (political philosophy). Under this model, the aim of competition policy is protection of individual economic freedom of action as a value in itself, or the restraint of undue economic power. Two key notions of this economic philosophy, first, that the economic system should allow all individuals to participate unhampered by the economic power of others (pluralism). Second, that economic freedom is not guaranteed by an unregulated market – the risk of monopolies or cartels necessitates laws to sustain economic freedom.

Translated to competition policy, ordoliberalism necessitates rules that safeguard economic freedom in the marketplace by imposing obligations of fair conduct and suppressing economic power. This conception of competition is henceforth labelled the economic freedom approach.

1.2.1 The concept of competition in Treaty

The economic freedom interpretation of competition comes closer to explaining Article 81. In an early treatise on EU competition law, scholar took the view that competition is distinguished by two characteristics: 1) freedom of action of the individual enterprises; 2) the possibility that market participants may make a choice. The market participants in question are those who enter into the agreement to reduce the number of market players, this action increases the degree of monopolisation in the relevant market. Here, a restriction of competition is a restraint between the parties to the agreement, and it is irrelevant whether the market is competitive. This position is reflected in lots of decisions of the Commission. In discussing exclusive distribution agreements, the Commission noted, that the exclusive nature of a contractual relationship between a producer and a distributor is viewed as restricting competition since it limits the parties freedom of action in the territory covered (Twenty-third Report on Competition policy (1993) para. 212.)

The same analytical structure is found in many other decisions, an approach which would baffle any economist: the facts that explain why the agreement is declared anticompetitive under Article 81(1) are the same facts which are used by the Commission to show that the agreement yields economic benefits under Article 81(3). Commission is concerned about the risk other market participants might be foreclosed from market.

Article 81(3) is incompatible with the economic freedom: 1) if an agreement has an undesirable effect on economic freedom under Article 81(1), no exemption should be granted. The exemption provision in Article 81(3) is a way of reconciling an ordoliberal conception of competition with other important values. The use of other values in an agreement that restricts the economic freedom of undertakings may only be tolerated under the strict conditions provided in Article 81(3): that consumers benefit that economic freedom is not completely eliminated, and that the restriction of economic freedom is the least necessary to achieve those other public interest goals. This way the objective of economic freedom is sacrificed in part to other public interest goals.

1.2.2 The rule of reason

The rule of reason is a concept that arose early on in the history of US antitrust law, its advocates pointed out that the uncertainties were less problematic than the procedural delays and arbitrary decisions of the Commission at present. US antitrust law distinguishes between practices that are per se unlawful and practices that are analysed under a rule of reason – that is, by a fact-specific inquiry to determine whether the agreements cause anticompetitive effects – to decide on their legality.

Procedurally the rule of reason eliminates the need for ex ante scrutiny, and substantively parties are able to identify for themselves whether the agreement causes anticompetitive effects necessitating notification.

The arguments in favour of adopting a rule of reason have been weak. First, its proponents advocate an exclusively economic analysis to competition. Their arguments do not fit within the political scope of EU competition policy. Second, an economic welfare interpretation of Article 81(1) would mean either that Article 81(3) is irrelevant or that Article 81(3) provides for exemption in cases where an inefficient agreement may nonetheless yield desirable political benefits, a conclusion that is the exact opposite of what the advocates of a rule of reason wish to achieve. The rule of reason suits US law but same legal method is unnecessary in the EU, because Article 81(3) provides the same functional of the rule of reason.

In some cases the Court takes the view that a restriction of the economic freedom of market players is a necessary element in the identification of a restriction of competition but is not sufficient: the degree to which economic freedom is restricted must be measured. The Courts’ contribution has been to identify parameters by which such measurement takes place: the economic context in which the undertakings in question operate, the products covered and the market structure. A recent example of this is the Court of First Instances (CFI’s) review of the Commissions Television par Satellite decision. The fact that the agreement restricted the economic freedom of other potential market participants was enough to find that the agreement restricted competition for the purposes of Article 81(1). In upholding this aspect of the decision, the CFI held that the Commission had not found a restriction of competition in the abstract, but had proved that the agreement foreclosed access to a lucrative market by other potential broadcasters, enough to show that the restriction of economic freedom was significant.

A similar approach can be seen in the Commissions Visa International decision. The Commission stated expressly that the efficiency justifications had no role to play in Article 81(1), noting that the impugned agreement would have affected the degree of price competition for acquiring banks and issuing banks, for their respective costs and revenues were largely fixed by the Visa agreement.

The Court has explained that a restriction of economic freedom in a contract would not infringe Article 81 when it did not constitute the major part of the agreement but was necessary to allow for the implementation of an agreement that did not infringe competition. For example non compete clause. In the Commission’s understanding the non-compete clause is an ancillary restraint. The legal effect of this is that if the major part of the agreement is not anticompetitive, then the ancillary restraint is lawful, without any competition law analysis of the restraint. The application of this doctrine requires an analysis of two matters: 1) determining first whether the restraint is subsidiary to the main transaction; 2) whether the restraint is necessary for the commercial operation of the main transaction. The Court has given some pointers on the second aspect, indicating that the restraint must be proportionate to the objective being achieved. The Court has given less guidance how to identify if a clause is ancillary, the case law although suggests that a restraint is ancillary when it concerns the behaviour of the parties outside the framework of the agreement (the non-compete clause is about restricting the sellers activities).

The doctrine serves to allow the enforcer to focus on the principal aspects of a transaction in question in determining its legality. If the main agreement is found not to restrict competition or is exempted, the ancillary restraint is valid automatically. (See Gøttrup-Klim and Metropole decisions.) The doctrine is merely about administrative convenience, allowing the Commission to focus on the principal effects of an agreement, but also to control minor aspects if the ancillary restraints are excessive in duration or scope.

The law on restraints of trade forbids a contract term like the non-competition clause discussed above on two grounds: first because the clause deprives one party of his freedom to use his skills, and second because society suffers as it is unable to reap the benefits of that person’s skills.

Phrase ancillary restraint was introduced in US antitrust law to describe restrictions that were reasonable and outside the scope of the Sherman Act. Now this interpretation has waned, but the doctrine of ancillary restraints has remained in US. It applies when restraints in joint ventures are assessed. Ancillary restraints are lawful if they are necessary for the transaction and they do not diminish the efficiency of the transaction. Ancillary restraints in EU competition law is another misplaced label from the US. In Gøttrup-Klim ECJ attempted to apply an approach which is similar to US law on ancillary restraints. EU’s approach is to inquire whether the restriction is necessary for the commercial success of the main agreement. If it is, and if the agreement as a whole is lawful: it is declared lawful even if it restricts the parties' economic freedom.

1.2.3 Agreements that create competition

In discussing the legality of an exclusive distribution contract (where a producer uses one distributor as exclusive distributor in his geographical area) the Court said: The competition in question must be understood within the actual context in which it would occur in the absence of the agreement in dispute. In particular it may be doubted whether there is an interference with competition if the said agreement seems really necessary for the penetration of a new area by an undertaking.( Societe Technique Miniere) From this decision follows that if the agreement creates more competition by introducing new products on the market, then the restriction on other distributors is acceptable, if the restriction is necessary to create new competition.

The Commission used the same reasoning in Inntrepreneur and Spring. The partial restriction on wholesalers economic freedom was more than offset by the significant market opening possibilities for brewers created by the agreement. Underlying this analysis is belief that by providing easier market access, consumers benefit from more choice and lower prices. Thus there is some recognition of the positive economic benefits of the agreement, although the analysis is premised upon creation of economic opportunities for other market players, not on an economic cost–benefit analysis.

1.3 Internal market

Since the beginning of EU competition policy has been used as mechanism for integrating the single market. The importance of competition law to create or maintain single market is reflected in the priority given by the Commission to apply competition law to practices that disintegrate the internal market.

Commission have stated that market integration is the first principle of EU competition policy. Consten and Grundig judgment is often used as an illustration that market integration is more important than economic efficiency. This case is also an illustration of the Community’s irrational understanding of market integration. To sum up, while market integration is an important end, it is not clear yet whether the enforcement policy in Grundig facilitates market integration in practice or whether it acts as a barrier to greater integration.

1.4 Economic gains

Economic welfare is one of the expected benefits from membership in the EU. The Commission has noted that the contribution of competition policy to economic efficiency early on. In the First Report on Competition Policy you can find passage: Competition is the best stimulant of economic activity since it guarantees the widest possible freedom of action to all. An active competition policy pursued in accordance with the provisions of the Treaties establishing the Communities makes it easier for the supply and demand structures continually to adjust to technological development. Through the interplay of decentralised decisionmaking machinery, competition enables enterprises continuously to improve their efficiency, which is the sine qua non for a steady improvement in living standards and employment prospects of the Community.

Economic efficiency is recognised implicitly in Article 81(3) – agreements which restricts economic freedom may be exempted when there are evidence that these restraints improve the production or distribution of goods or promote technical or economic progress.Typology of efficiencies:1) Allocative efficiency: existing goods and services are allocated to those who value them most, in terms of their willingness to pay. Resources thus allocated are used in the best way possible; no other distribution would increase aggregate welfare more. 2) Productive efficiency: this concept focuses on a particular undertaking or industry and considers whether a undertaking organizes its resources in such a way that it exploits all economies of scale, exploits existing technology effectively, and cuts all superfluous costs, so that production is at minimum cost. 3) Dynamic efficiency: this is a measure of whether undertakings have the ability and the incentives to increase productivity and innovate, developing new products or reducing production costs which can yield greater benefits to consumers.

Allocative and productive efficiencies are static measures: a snapshot of the current market position. Dynamic efficiency instead looks at the potential that the economy has to develop further. A high degree of dynamic efficiency yields an increase in allocative and productive efficiencies: as new products are developed, more goods that consumers value are produced and resources are used better than they would be without the technological development.

It must be remembered that while economic efficiency is relevant, it is not an end in itself. First, an exemption is only granted if the agreement and the individual restrictions in the agreement can’t be avoided to achieve the efficiencies. So if the efficiencies can be gained by causing less distortion, then the less harmful way must be chosen. Second, the consumer must benefit from these efficiencies. What the consumer could lose as a result of the restriction must be compensated by what the consumer could have as a result of the agreement. Efficiencies cannot justify an agreement that eliminates competition.

1.4.1 Efficiency’s place in the EU

The primacy of economic efficiency is noted even further by Regulation 1/2003, which gives that Article 81(3) has direct effect. Productive and dynamic efficiencies are key to EU overall economic policy. The flipside, however, is that the increased recognition of the significance of economic efficiency leads to a unitary assessment of Articles 81(1) and (3), whereby the restriction of economic freedom under Article 81(1) today operates as a primary filter to indicate that there could be an infringement of the competition rules, and the analysis of if there is economic efficiency in Article 81(3) is thorough assessment of the economic realities of the agreement’s impact.

Under Article 81(3) the agreement will not be tolerated if it eliminates competition, even if there are efficiency gains: Ultimately the protection of rivalry and the competitive process is given priority over potentially pro-competitive efficiency gains which could result from restrictive agreements. Rivalry between undertakings is an essential driver of economic efficiency, including dynamic efficiencies. To sum up, the main aim of Article 81 is to protect the competitive process in the EU. It means that the Commission thinks that, in the long term, rivalry gives greater economic benefits than efficiencies generated by the elimination of rivalry.

Today efficiency is an increasingly important value. 2002 Report on Competition Policy: Competition policy then serves to: 1) addressing market failures resulting from anti competitive behaviour by market participants and from certain market structures; 2) and contributing to an overall economic policy framework across economic sectors that is conducive to effective competition, on the other.

Chapter 2: Economics as part of competition law

2.1 Introduction

Competition law in the EU is now presented, by DG Competition, as a set of rules dominated by an economic paradigm that focuses on consumer welfare. Three economic models that have been present in the evolution of antitrust law in the US (the SCP, the Chicago and the post-Chicago paradigms). The note to be taken from this chapter is that the economic models and the underlying policies influence each other.

Differences among economic models should not be exaggerated. The differences are how to apply economics to regulate markets. Need for competition law intervention rises when there is a market failure. The market system is thought to be the best mechanism through which the fundamental economic questions are answered.

When there is competition, economic benefits accrue without the aid of government intervention. Markets fail when the way for undertakings to maximize profits is no longer to reduce price and maximize output, but to limit production and raise prices. This leads to an inefficient use of existing resources. Competition laws are then needed to overtake that market failure and restore competitive markets. There is also view that competition is also a means to achieve productive efficiency, in three ways: 1) competitive pressures give undertakings greater incentives to reduce costs; 2) competition is a means of selecting the more efficient undertakings, which will thrive; 3) competition to innovate is a source of productive efficiency. This means that competition may also be a means to achieve dynamic efficiency.

Inefficiencies is the thing to be penalized and inefficiencies will be found where there is market power.

2.2 Three economic paradigms

The Structure–Conduct–Performance (SCP) paradigm was prominent in the 1950s and 1960s. It says that by observing the structure of a market, inferences can be drawn as to how undertakings conduct themselves, and this allows to evaluate the markets economic performance. Market structure is a term used to define the characteristics within which undertakings operate in a product market. The main characteristics are the number of undertakings and their size, but also relevant is the market power of the undertakings customers, the ease with which new undertakings can enter the market and old ones can exit it. Performance is the key by which the conduct of undertakings is measured. The standard measure for this is whether the undertakings enhance economic welfare.

The theory says that the more closely the market in question approaches conditions of monopoly the worse its performance, and the more closely the market approaches the conditions of perfect competition the better its performance.

Perfect competition - many sellers producing homogeneous goods where the profits are primarily determined by consumer demand and consumer choice is based uniquely on price. Each seller’s market share is too small to have any impact on the price of the goods in question.

Monopolistic competition - many sellers producing differentiated products, whereby consumer choice is dictated by considerations other than price.

Oligopoly - a small number of sellers where each seller’s profits are affected significantly by the actions of the others in the market.

Monopoly - one seller in the market.

SCP paradigm for competition policy is that it is possible to identify which market structures lead to anticompetitive results and devise a competition law that is designed to modify or prevent market structures that are linked with poor economic performance. The policy recommendation that comes from the SCP paradigm is that special attention should be given to those market structures most likely to reduce consumer welfare, those are monopoly and oligopoly. SCP paradigm sees the structure of an industry as the cause of market failure, remedies are designed primarily to alter the structure of the market, and to prevent incumbents from raising entry barriers.

Chicago scholars launched three central challenges to the SCP approach. First, the connection between industry concentration and anticompetitive effects was refuted. Second, market structure could be affected by performance in the market. The main idea of this argument is observation that if a undertaking with a large market share chooses to behave anticompetitively by cutting output, this activity will invite new entrants to the market, which will improve competition. Third, the Chicago School questioned the SCP’s definition of barriers to entry. Barrier to entry for Chicagoans is a cost which must be borne by a undertaking which seeks to enter an industry but is not borne by undertakings already in the industry. For Chicago School only entry barriers are property rights conferred by government.

To sum up, these three ideas led to the basic principle of the Chicago School. Basic principle is that markets normally cure themselves and competitive outcomes are likely without significant government intervention. Chicago School also argued that courts or government agencies are often unable to effect any improvement in welfare.

Competition enforcement is much less intrusive under the Chicago School model. The orthodox Chicago position of the 1960s identified price fixing and large horizontal mergers as the only competition problems.

The post-Chicago paradigm is based on complex tools to determine whether there is a market failure, and the specific features of an industry are important to determining market failures. The main indicator of market failure is presence of market power, defined by the ability to set prices above marginal cost. This paradigm shows the wide generalisations that characterise both the SCP concern that structurally imperfect market structures are risks, and the Chicago view that markets are enough robust to smash anticompetitive strategies, in favour of fact investigations to determine what market characteristics create risk of market failure.

One of the key innovations of this paradigm is the rising importance of empirical fact-finding. (Check Matsushita Electric Industrial Co. v. Zenith Radio Corp and Federal Trade Commission v. Staples merger decision). The post-Chicago paradigm says that in addition to inefficient behaviour, undertakings with market power can cause market failures by taking steps to exclude entry.

2.3 European competition policy and economics

EU competition law has not been affected by economic paradigms as US antitrust has. There is no single reason why economic analysis had a less important role in the early years of EU competition policy than it did during the same period in the US. But, it is possible to note a different reasons why use of economics has increased over the years. Commission has gained more experience of the market it regulates, greater experience has allowed the Commission to identify practices that don’t have anticompetitive effects. A lot of factors have enabled Commission’s shift to an economics-based approach. However, there is no thing as the economic analysis of competition law, but there are different paradigms that compete to be the main one.

The Commission in 1996 published the Green Paper on Vertical Restraints, which marks the beginning of a legal and cultural change in EC competition policy – the Green Paper is the first serious attempt to develop economic analysis to competition law. Passage from the Green Paper: “economic theory cannot be the only factor in the design of policy. Firstlystrict economic theory is just one of the sources of policy. In practice the application of economic theory must take place in the context of the existing legal texts and jurisprudence. Secondly, economic theories are necessarily based on simplifying assumptions often obtained in the context of stylised theoretical models that cannot take into account all the complexities of real life cases.”

There are three themes in the EU economic approach. First, competition policy should be enforced to benefit consumers, second, competitive risks grow with increased market power, third, that economic freedom, and a plurality of market players, guarantees economic benefits.

2.4 Economics in competition law

It is important to notice that any legal standard will simplify the economic theory upon which it is based, and lead to risks of error. If standard is too strict so that the risk of Type 1 errors (prohibiting efficient behaviour is increased or the standard is too loose, because of that the risk of Type 2 errors inefficient behaviour is not prohibited arises. Community prefers pluralism, and because of that it is more willing to make Type 1 errors, rather than Type 2.

Chapter 3: Competition law and public policy

3.1 Introduction

Commission has move towards using consumer welfare as the benchmark for testing the legality of agreements. Commission has claimed that the purpose of Article 81(3) is: “to provide legal framework for the economic assessment of restrictive practices and not to allow the application of the competition rules to be set aside because of political considerations”. Commission is repositioning competition policy, saying that other non-economic values, except the single market, have no role in decision-making process. Everyday enforcement of competition law should not be preoccupied with achieving these political benefits directly, public interest goals should be translated into criteria for law enforcement, but they should be achieved by preventing distortions of competition in the single market. Basic aim of competition policy is the indirect pursuit of public policy gains, and if competition policy were used to contribute directly to wider public policy aims, the pursuit of the public interest might fail, and it would be a move away from free market principles that under run the EU.

3.2 Policies

Environmental policy represents most clearly the role played by other Community policies in shaping competition law. In pursuing environmental policy, Commission has encouraged voluntary agreements which pursue environmental objectives while respecting the competition. These agreements are supplementing EU environmental regulation. Agreements have been exempted under Article 81(3) when they reduced the volume of plastic wastes and reduced environmental risks, or led to lower consumption of energy and to better prospects for realising energy reduction programmes (Exxon – Shell; Philips - Osram). Environmental gains were part of the reason for exemption. Also agreement contribution to Community environmental policy was given greater prominence in DSD, where the Commission granted an individual exemption on the basis of that the agreement gave effect to environmental objectives.

The Commission’s commitment to achieving environmental objectives through competition law is reflected in part 7 of the Guidelines on Horizontal Agreements who describes the Commission’s policy on environmental agreements among competitors. The guidelines gives that the Commission exempts agreements which reduce environmental pressure so long as the net contribution to the improvement of the environment overall outweighs increased costs of production. This approach is clearly expressed in the CECED decision where the Commission exempted an agreement among washing machine manufacturers to phase out machines with high electricity consumption. Commission found that cost savings for consumers were greater than the price increase in the electrical goods subject to the agreements. In two similar notifications to CECED, the Commission issued a comfort letter where it sad that when reviewing an environmental agreement Commission will consider the cost savings for consumers of the goods in question, and then estimate the gains which the society derives from improved environmental situation. Environmental benefits to society are relevant, and environmental benefits to society are enough to grant an exemption.

Commission has been criticised, by acknowledging that environmental considerations are a factor that influences the Commission to exempt an agreement, the recent decisions have bin against the interpretation of Article 81(3) given by the Commission in its Guidelines on the Application of Article 81(3), which provide that efficiencies are the sole benefit that counts when exemptions are granted.

A second way to interpret the role of environmental protection in competition cases is to say that the duty imposed by Article 6 EC to integrate environmental protection in the Community policies and activities referred to in Article 3 EC means that environmental protection is normatively superior to competition law. Cleaner environment may be the ground for an exemption, regardless of any efficiency considerations.

In industrial policy there is relationship between two potentially conflicting fields, competition and competitiveness. These fields have been mediated by Article 157 EC: “while this provision empowers the Commission to ensure that European industry is competitive, it also provides that this shall not serve as a basis for the introduction by the Community of any measure which could lead to a distortion of competition”. The reason is that Community believes that competition in the internal market is the best source for industrial strength in world market. Competitiveness cannot be achieved through means that reduce competition.

Commission has said that agreements designed to reduce overproduction when demand is low yield three benefits: 1) efficiency; 2) less impact on employment; 3) facilitating and speeding up restructuring which would lead to enhanced competitiveness of EU industry. These considerations, cumulatively, provide a enough degree of technical or economic progress to outweigh the restriction of competition. These decisions show that promotion of industrial policy is not a enough condition for exempting an agreement. But anticipated industrial policy gains are noted when explaining the reason to exempt agreement. Compared to the environmental agreements, industrial policy arguments have less weight.

Employment policy. The relevance of an agreement’s effects on employment in the application of EC competition law has increased. Commission must take employment into account because Article 127 EC requires that the impact of employment will be considered for all Community measures. Exemptions have been granted to agreements which in addition to efficiencies led to stability in the labour market (Metro). The ambiguity of the importance of employment considerations is even more profound than that which surrounds the consideration of environmental or industrial policy.

Interpretation of the provisions of the Treaty is both effective and consistent means that agreements concluded in the context of collective negotiations between management and labour in pursuit of such objectives must be seen as falling out of the scope of Article 81(1). Court didn't say that any collective agreement falls outside Article 81, but sad that agreement in which was designed to guarantee a certain level of pension for all workers is improving their working conditions.

Consumer policy. Recently the attention of the Commission has been on the final consumer, and the Commission reflects more carefully about how the final consumer may or may not gain from the behaviour of undertakings. This is consistent with an increased Europeanisation of consumer law and policy following the Treaty of European Union. In the context of competition law, the increased interest in the final consumer is evidenced by the creation of the position of Consumer Liaison Officer in 2003, whose task is to act as a contact point for consumer organisations, to alert consumer groups to cases where their input might be useful, and to intensify contacts between DG Competition and other Directorates General whose portfolio includes consumer interests. Rhetoric from the European Competition Days. Yearly event organized by DG Competition to explain the role of competition law to European citizens. Emphasises the benefits that the final consumer obtains from the enforcement of competition law. The focus on the final consumer is such that some have suggested that the standard by which EC competition law operates is not efficiency, or economic freedom, but consumer welfare. The speeches of the former Commissioner indicate that under his stewardship competition policy focused on consumer interests narrowly defined to include only the final consumer. The interests of the final consumer are varied: they include low prices, high-quality products, a wide selection of goods and services, and innovation. At times it will be fairly easy to see how the enforcement of competition law can safeguard the consumer interest: blocking a cartel will lower prices; allowing an agreement which reduces production costs that yield lower prices also benefits consumers. In most cases prohibiting inefficient agreements and allowing efficient ones will lead to consumers benefiting. This need not always be the case: for instance, an agreement between two undertakings may lead to cost savings that are achieved by reducing certain services that some consumers value, simply to cut costs (shutting down local branches in favour of large out of town shops). This may reduce prices but inconvenience certain consumers who have further to travel to obtain the goods. Fact that an agreement can lead to some consumers gaining and others losing requires further consideration if one is to take the consumer welfare objective seriously.

While competition law and consumer law are designed to remedy market failures, they do so in slightly different ways. Competition law safeguards.

Agreement which is efficient but which may harm consumer interests may not be exempted unless modified to guarantee consumer benefits. This approach where exemption is available only if the parties undertake to take certain steps designed to accomplish other Community policies is found also in decisions affecting cultural policy.

Culture policy. In the field of book pricing, the legal issue for Commission is whether a nationwide agreement to fix the resale prices of books can be tolerated or not. The price-fixing agreement among publishers serves a public interest, while restricting competition. The solution reached by the Commission and the Courts is formalistic. Commission and Courts have held that in so far as a price-fixing system has no effect on trade between Member States, then Article 81 does not apply, because it forbids agreements whose anticompetitive effects have a cross-border element, but that as soon as price fixing affects trade between Member States, then it is prohibited.

TV broadcasting. The non-economic factor of sport is that it performs social function bringing people together. From the way Commission regulates this market we can see how competition, industrial policy and cultural policy interact to form the Commission’s approach. Commission intervenes in this sector by the liberalization of the broadcasting sector combined with developments in technology mean that new broadcasting media are on the market. The Commission thinks that popular sports and Hollywood movies are golden content what is necessary to attract consumers to new broadcasting services. The competition problems start when football clubs in a league allow their football associations to sell the broadcasting rights of each league collectively. The Commission’s biggest concern is that the way in which broadcasting rights are sold means that broadcasters who are unable to obtain rights to sports events may exit the market. From an economic perspective there are worries about the lack of competition in broadcasting and about the failure of new broadcasting technologies. The Commission got legal support from the Bosman decision when the Court stated that national restrictions on the right of workers to move freely could be justified having regard to the social importance of sporting activities and the need to maintain a balance between teams to preserve equality and uncertainty as to the results. Statement was used to support the view that the application of Article 81 could be affected by the social and cultural values of sport. However there is little reference to solidarity among the football clubs, but considerable reference to economic efficiency instead. Commission is using Article 81 to make markets more competitive, this is not necessarily within its powers.

The flipside of the Commission’s regulatory approach to the football broadcasting

sector is that it attempts to create a situation where a variety of policy interests are catered for: 1) by allowing leagues to sell all games collectively it retains a redistributive element; 2) by forcing packages of broadcasting rights to be sold separately it guarantees competition among broadcasters; 3) by creating mechanisms whereby internet and 3G broadcasters have opportunities to buy some rights, the industrial policy goal of introducing this technology is achieved. This process shows how competition law being used to achieve other public policy goals. Process is defective because by regarding industrial policy objective the Commission forces sellers to behave in specific ways, which looks like it is beyond what is allowed under Article 81.

National interests. Commissions interventions in the cultural sector predate a significant ruling of the Court of Justice, Wouters. Case provides a method by which anticompetitive practice could be tolerated if practice provides a necessary means to support a legitimate national policy. The ratio decidendi of Wouters judgment is that if a restriction of competition is necessary to safeguard the proper practice of the legal profession, then competition law does not apply to the measure, and the restriction of competition is tolerated. Public policy is more important that the application of competition law. Basis of the Court’s approach is established in the case law of the single market. Court has for long time recognised that in applying the rules on free movement MS could restrict the free movement of services if they impose certain restrictions that apply uniformly to all economic actors which are designed to protect a national interest. In Court’s view, MS can implement mandatory requirements to justify restriction on the free movement of services. In Reiseburo Broede v. Gerd Sandker the Court was faced with a German law that prevented a person from carrying out judicial debt collection work without the assistance of a lawyer. This affected foreign providers adversely and prima facie infringed their freedom to provide services. Court held that the restriction can be justified if the rule is necessary to ensure that clients are enough advised of their legal position. Court had recognised that the integrity of the legal system is a justification for restricting the application of the EU’ s internal market law, and Wouters extends this approach so application of EU competition law is suspended when necessary to keep the integrity of a national legal system.

Interpretation of Wouters means that any restriction of competition necessary to safeguard a national public policy consideration could be allowed to stand. Against this interpretation, it has been argued that Wouters stands for the proposition that a restriction of competition is tolerated when necessary for the proper performance of an economic service. The Commission applied Wouters to rule that prohibition was necessary for the proper functioning of the market in question. This approach is similar with the doctrine of ancillary restraints.

There are two main differences between the two views. One is that the doctrine of ancillary restraints applies in circumstances where the agreement is economically beneficial. To contrast, the rule on mandatory requirements allows for market rules to be suspended for a greater good even if the agreement is anticompetitive. The second is that the rule on mandatory requirements applies to actions which sustain state policies, whereas the ancillary restraints rule applies to agreements that sustain markets.

3. 3 Competition policy among other EU policies

Six methods are used to integrate public policy considerations in competition decisions. 1st method is exclusionary: the Community resolves the tension between competition and another policy by deciding that the other policy is more important, and because of that competition law does not apply. In the competition v. employees’ rights in Brentjens, the second in the resolution between competition policy and national public policy in Wouters. In Wouters, the Court deployed a proportionality test, so that a restriction of competition is justified when necessary to achieve the public policy objective. In Brentjens, the fact that the agreement is designed to safeguard the interests of employees is enough to find that competition law does not apply. The Brentjens approach can be deployed when other EU policies have greater weight than competition law. Article 81(3) exemptions have been made conditional upon certain contractual modifications (obligation to keep staff) while the non-application of Article 81(1) does not allow for any post-assessment changes.

2nd method is to redefine economic efficiency to include other public policy considerations. This occurred in environmental agreements where reduced pollution was characterised as an economic benefit.

3^rd method is to use non-economic benefits as factors that gives the balance in favour of granting an exemption. This approach is seen in some early environmental agreements and in cases where industrial policy considerations arise. It is the most opaque approach.

4^th method is to grant conditional exemptions and to use remedies to achieve the public policy goal. In this scenario an agreement is exempted only if the parties agree to take steps that help in securing certain Community objectives. Example is the UEFA decision.

5^th method is to rule that an agreement does not fall within Article 81 because trade between MS is not affected. This is how the Commission avoided confronting the culture questioning the books sector.

6^th method is to find that the non-competition consequences of the agreement are of such importance that if an agreement is inefficient, but contributes to another Community policy, it is exempt. But if an agreement is efficient but harms a particularly important policy, then it is not exempt. If the environmental effects are more relevant than the competition effects, then the analysis under Article 81(3) will be towards the ecological effects. This approach is also present in exemption decisions where impose additional commitments on the parties in the decision to exempt. The other interest (employment or industrial policy) has such weight that it enlarges the scope of considerations relevant for exemption.

Anti-doping and antitrust. The breadth of Wouters is illustrated by the recent judgment in Meca-Medina. Two athletes were given bans for doping offences. They alleged that the doping rules, fixed by the International Olympic Committee, were anticompetitive, in breach of Article 81. According to the Court of First Instance, the Commission was entitled to reject this allegation because the rule was about sport, not about trade, so the competition rules did not apply. But ECJ took a different line and subjected the anti-doping rules to the Wouters test. The key passage: “Not every agreement, which restricts the freedom of action of the parties or of one of them necessarily falls within the prohibition laid down in Article 81(1) EC. For the purposes of application of that provision to a particular case, account must first of all be taken of the overall context in which the decision of the association of undertakings was taken or produces its effects and, more specifically, of its objectives. It has then to be considered whether the consequential effects restrictive of competition are inherent in the pursuit of those objectives and are proportionate to them.

As regards the overall context in which the rules at issue were adopted, the Commission could rightly take the view that the general objective of the rules was, as none of the parties disputes, to combat doping in order for competitive sport to be conducted fairly and that it included the need to safeguard equal chances for athletes, athletes health, the integrity and objectivity of competitive sport and ethical values in sport.

Also given that penalties are necessary to ensure enforcement of the doping ban, their effect on athletes freedom of action must be considered to be, in principle, inherent itself in the anti-doping rules. That’s why, even if the anti-doping rules at issue are to be regarded as a decision of an association of undertakings limiting the appellants freedom of action, they do not, for all that, necessarily constitute a restriction of competition incompatible with the common market, within the meaning of Article 81 EC, since they are justified by a legitimate objective. Such a limitation is inherent in the organization and proper conduct of competitive sport and its very purpose is to ensure healthy rivalry between athletes. The parties argued that disqualification damaged their earnings as they would be unable to win prize money or obtain sponsorship. The Court held that the restriction of their economic freedom was justified because the doping rules pursued a variety of other public interest goals. It extends the rule that certain public policies justify the non application of Article 81 (namely athletes health and the integrity and ethical values of sport). On the other hand, more restrictively, the ruling may stand for the proposition that the proper conduct of an activity (in this case, long-distance swimming) may justify a restriction of competition.

The Commission has recognized the need to identify main priorities in its policy, consistency in its enforcement and transparency in its decision-making process. Essential step to achieve these objectives is to narrow down the meaning of Article 81(3). The recently published Guidelines on the Application of Article 81(3) EC achieve the narrowing down in two ways. 1) they interpret the phrase “technical or economic progress” so that it only includes economic efficiency, saying that the purpose of the first condition of Article 81(3) is: “to define the types of efficiency gains that can be taken into account”. 2) they marginalise the role of non-efficiency considerations: “Goals pursued by other Treaty provisions can be taken into account to the extent that they can be subsumed under the four conditions of Article 81(3). If this is read as if it were a statutory provision, the key is the verb “subsume”. That means to place an idea or a term in a wider category. Under this reading, employment, cultural and industrial policy considerations become irrelevant because they cannot be put under the efficiency gains of Article 81(3). Under this approach, most of the policy aims could not be integrated into an “efficiency” analysis. This is the intention of the Commission’s Guidelines. The anticipated benefit that the Commission envisages is that the degree of discretion in the hands of the national courts and National Competition Authorities (NCA) is reduced.

Commissions approach takes a wide conception of the economic effects of an agreement, which is not restricted to allocative, productive and dynamic efficiency. The Guidelines separate between cost efficiencies of an agreement and qualitative efficiencies. The Commissions decision in CECED suggests that qualitative environmental benefits are relevant. Any gain can be taken into consideration provided that it can translate into a benefit for consumers. Than we can subsume consumer safety and environmental improvements under Article 81(3), but not all industrial policy considerations.

This approach suffers from three weaknesses: 1) it risks being insensitive to certain interests which we may wish to preserve even if there is a reduction in economic efficiency or consumer welfare. It might be said that if there are policies of a noneconomic nature that should be safeguarded. This might make for a neater methodology. Agreements that have a consumer benefit are exempted under Article 81(3) and agreements with benefits of a non-economic nature are excluded from the application of competition law; 2) this approach raises the problem of characterisation. Question, is an agreement by certain manufacturers not to use a particularly harmful chemical in the workplace justifiable on the basis of the employees safety, or is agreement is a way of responding to a market failure, is needed to be asked. Public policy considerations retain a role in spite of the Commission’s guidelines to the contrary.

The approach doesn’t take into account that the Commission will remain competent to apply Article 81, and in this institutional setting it is less easy to escape the relevance of the impact of competition law on other Community policies.

While the EU remains as a model of disciplined pluralism, where intense concentrations of economic power are viewed with big suspicion, increasing importance is given to exempting agreements on the basis of economic efficiency as a means to obtain the tangible economic benefits upon which the EU project was built. This means that the future relevance of non-economic public policy considerations is vain. These may justify the non-application of competition law under the Wouters and Brentjens formulae, but exemptions under Article 81(3) are to be based upon proof of economic benefits.

Chapter 4: Meaning of market power in EU competition law

4.1 Introduction

There are four ways to think about market power. The first equates market power with the ability to increase price (the neoclassical approach), the second equates market power with commercial power, the third sees market power as the ability to exclude rivals so as to gain the power to increase price, and the fourth sees market power as a formal jurisdictional test.

The first way shows upon economic considerations. Market power is a worry when it is both significant and durable. This approach has been accepted by the US courts. However, market power is also a relative concept: the greater the power, the more harm the undertaking can inflict. Certain infringements, like excessive or predatory pricing, are penalised only when the undertaking in question has significant market power, while other kinds of infringement, like anticompetitive distribution agreements, may be penalised even if the undertaking has less market power, provided it is able to cause damage to competitors or consumers. Different thresholds of market power apply depending on the infringement in question, and, as a rule of thumb, the number and the degree of anticompetitive risks posed increases with higher levels of market power.

A second way of defining market power is to inquire whether the undertaking has greater commercial power than others on the market. This approach sometimes finds its way into the discussion of certain contract law doctrines, like economic duress, where one party has a “situational monopoly” over the other contracting party.

A third definition of market power, post-Chicago, provides that a undertaking has market power when it is able to devise strategies that can harm rivals and so give it, in the future, the power to raise prices and reduce output. This approach is wider than the neoclassical definition, but it has the same aim: to penalize undertakings whose strategies can have undesirable economic effects. On the other hand, this definition is also related to the definition of market power taken by those wishing to safeguard economic freedom, because it focuses on the power to harm competitors.

A fourth approach is to interpret market power as a jurisdictional concept. This has been applied in Article 81.Commission has sad that certain types of agreement are lawful provided the parties market shares are below a given threshold. Using market shares as a means to establish jurisdiction creates safe harbours so that undertakings below the relevant threshold know that certain prohibitions do not apply to them. The jurisdictional approach acknowledges that there may be market power below the level set by the safe harbour, but trades off under-enforcement for cheaper and more effective application of the law.

4.2 Dominance

In applying Article 82, and for most merger cases, the Commission must identify dominance. The way that the Courts have interpreted this notion allows us to understand what concept of market power prevails in EC competition law. Thinking about market power in the context of the meaning of dominance is also helpful because this concept is currently in transition: the Commission wishes to move away from the current emphasis upon commercial power and towards identifying dominance with substantial market power based on economic theories. In Michelin case, the Court held that Article 82 “prohibits any abuse of a position of economic strength enjoyed by an undertaking which enables it to hinder the maintenance of effective competition on the relevant market by allowing it to behave to an appreciable extent independently of its competitors and customers and ultimately of consumers”. This passage shows the standard test for dominance. The Courts reference to the ability to behave independently to an appreciable extent is relevant for two reasons: 1) because it means that Article 82 does not apply to the sort of market power that most undertakings have due to markets not being perfectly competitive; 2) because total control of the market is unnecessary to identify dominance. As the Court in Hoffmann La Roche stated, dominance does not preclude some competition, which it does where there is a monopoly or a quasi-monopoly, but enables the undertaking which profits by it, if not to determine, at least to have an appreciable influence on the conditions under which that competition will develop, and in any case to act largely in disregard of it so long as such conduct does not operate to its detriment. Dominance indicates a degree of market power which is considerably greater than that of other undertakings on the market and which can be used to harm the economic freedom of other market players by excluding them from the market.

As matters stand, it seems more plausible to suggest that commercial power is the basis upon which market power is identified in the case law. This can be justified in three ways. 1) EC competition law has been premised upon the protection of economic freedom. undertakings that have the power to undermine the economic freedom of others require control. 2) the aim of some of the abuse doctrines is to protect the economic freedom of other market participants, as such a definition of dominance which emphasises the commercial power of certain undertakings to harm the economic freedom of others is necessary. 3) the cases demonstrate an emphasis upon commercial power. The Court’s approach in United Brands is inconsistent with a neoclassical conception of market power and also with a post-Chicago perspective: dominance is found in the undertakings commercial strength irrespective of market outcomes.

4.3 Market power

Theoretically there is a precise, direct way of measuring market power as understood in the neoclassical sense: the Lerner index. This measures the difference between the price at which a undertaking sells its goods and its marginal cost. The greater the difference between the two, the more market power. But, this approach has little practical value because marginal costs are difficult to calculate. Moreover, undertakings with market power may have high costs and their prices will be just slightly above their inefficiently high costs, so the index underestimates their power. Instead, an indirect method is used to measure market power, based upon a calculation of the undertakings market shares and of barriers to entry. According to this approach, if a undertaking has very high market shares and entry for new competitors is blocked, then it holds market power because it is free to raise prices without fearing that its position may be undermined by new entrants.

Identification of market power is intimately connected with how we define the market. Market is defined by reference to demand substitutability and supply substitutability. Having established an agreed market and a share of that market, the next question is whether that market share is stable, or whether the market is so contestable that even with a large market share the defendant is unable to monopolise – this is a question of entry barriers.

Hypothetical monopolist test. The leading market definition test was first incorporated in the US Merger Guidelines in 1982, and now forms part of the approach deployed by the EC Commission. Example, the test asks if the defendant has a hypothetical monopoly in the strawberry flavoured ice cream market in Hyde Park, would he be likely to impose at least a small but significant and non transitory increase in price, assuming the terms of sale of all other products are held constant? If the answer is yes, than the relevant market is that of strawberry flavoured ice cream sold in Hyde Park because a price rise would be profitable. If the price rise would not be profitable, because most consumers would switch to other fruit flavours or leave the park and buy strawberry flavoured ice cream in the vicinity, the test is repeated by widening the market, say all fruit flavoured ice cream sold in Hyde Park and its vicinity, and considering whether a hypothetical monopolist of fruit flavoured ice cream would be able to increase the price; if not, the market is widened again, by adding the next best substitute, say all ice cream flavours, and asking the same question. This process continues until we find a market where a hypothetical monopolist would be able to raise the price of the product profitably. As a general simplification, the question the hypothetical monopolist test asks is this: is this a market worth monopolising? It is important to bear in mind that the hypothetical monopolist test will not usually be applied directly, rather it is a conceptual tool that allows one to assess the empirical evidence that is available. The hypothetical monopolist test is not designed to give the absolutely correct answer to the market definition question. Overall, the advantage of the hypothetical monopolist test is that it provides for a consistent and predictable framework – although each market definition analysis remains intimately fact specific. The Commission had begun to use this methodology under the Merger Regulation, and in 1997 it published a Notice on the Definition of the Relevant Market for the Purposes of Community Competition Law which indicated its increased willingness to deploy the hypothetical monopolist test in all competition cases.

4.3.1 Market definition

The early case law placed little reliance on economic assessment of the relevant market, as evidenced by the notorious United Brands decision, where the banana market was identified as a relevant market distinct from other fresh fruits. Problem with the United Brands approach was that the Court did not establish a clear methodology and at its centre rested a number of subjective considerations, with the risk of inconsistent decisions.

In the 1997 Market Definition Notice places the quantitative, hypothetical monopolist test at the helm of the assessment of the relevant product market, relegating the qualitative criteria to an initial stage to limit the field of investigation. Test: “The question to be answered is whether the parties customers would switch to readily available substitutes or to suppliers located elsewhere in response to a hypothetical small (in the range 5% to 10%) but permanent relative price increase in the products and areas being considered. If substitution were enough to make the price increase unprofitable because of the resulting loss of sales, additional substitutes and areas are included in the relevant market. This would be done until the set of products and geographical areas is such that small, permanent increases in relative prices would be profitable.”

In addition to a new methodology, one element of the United Brands test has been de facto overruled –the reference to the fact that certain customers prefer one product to another. According to the Market Definition Notice, if a product has no substitutes for a group of individuals the Commission will consider that there is a separate market only when such a group can be subjected to price discrimination. This novel approach is visible in Danish Crown case where sales of meat to caterers (restaurants, canteens and government offices) were treated as a different market from that for sales of meat to retail outlets (supermarkets and butchers). This distinction was justified on a number of grounds: for caterers the origin of the meat is less important (there have been more imports of meat from other EU countries by caterers). No arbitrage is possible between the two buyers because distribution and packaging is different (meat sold to supermarkets is already cut and packaged with all the information the consumer needs). Retailers need to buy all kinds of meat so as to offer clients diverse cuts, while caterers tend to buy larger quantities of one cut of meat. Market and the retail market are both worth monopolising separately, as the sales to one kind of customer are distinct from those to another.

The Commissions focus, both in its case law and in the notice, is on demand

substitution. Less attention is paid to the reaction of other undertakings. The Commission distinguishes between supply substitution whose effects are equivalent to those of demand substitution in terms of effectiveness and immediacy and supply substitution that takes longer to materialize.

A geographical market comprises an area where the conditions of competition are sufficiently homogeneous and which can be distinguished from neighbouring areas because the conditions of competition are appreciably different in those areas.

One general criticism of the geographical market definition is that often the focus is on the regions where the infringement of competition law takes place, when Michelin was accused of abusing a dominant position in the Belgian market, little analysis was carried out to determine competitive pressures from outside the border. But, when a joint venture agreement between Michelin and Continental was reviewed, the Commission had no qualms in identifying a European market, and at times even referred to the tyre market as a global one. Some see nothing problematic with this and suggest that in cases of restrictive practices the focus is on the market where the undertaking may be able to engage in abuses which hinder effective competition. While in cases of mergers or joint ventures the area is that where the undertakings are involved in the supply and demand of the products or services concerned. The definition of the relevant geographic market must take into account the geographic scope of the conduct in question. While doctrinally accurate, this approach can be challenged – if Michelin chooses to abuse its considerable market power in Belgium, it is legitimate to ask whether it is possible for competitors in other markets to enter the Belgian market, or for buyers of its tyres to exit Belgium and purchase elsewhere. Merely because under Articles 81 and 82 one is often judging the legality of past behaviour, this does not mean that an analysis of the consumers’ options is irrelevant. A static approach that defines the geographical market merely by identifying the area where the practice takes place can lead to unnecessarily narrow market definitions.

Market shares are a proxy for market power, not a precise measure. In EU law, when it comes to determining whether a undertaking holds a dominant position for the purposes of applying Article 82, the Court has indicated that a market share of 50 per cent gives rise to a presumption of dominance, and the case law suggests that a undertaking may dominate a market with market shares as low as 40 per cent. Lower dominance threshold in the EU is compatible with a competition policy designed to protect economic freedom because even undertaking that do not dominate the market absolutely are able to disrupt its functioning because of their commercial power. A lower threshold is also compatible with a strategic, post- Chicago, approach to market power because undertaking holding a large share of the market are more likely to be able to devise strategies that harm smaller competitors.

For completeness, two technical matters should be clarified. The first is that the market share of the dominant undertaking is insufficiently informative, and should be compared to the shares of competitors. The significance of a 40 per cent market share wanes if the second largest undertaking has a share of 39 per cent. In contrast, a large undertaking with a number of small fringe undertakings is more likely to have market power. A related issue is that the stability of market shares over time suggests dominance, while fluctuating market shares indicate lively competition.The second matter is about how to calculate market shares. For fungible goods market shares are calculated by the volume of sales. For other branded goods there is a preference to compute market shares by value, because this gives more prominence to the seller of expensive goods and reflects market power more accurately. In each case the aim is to identify the market share measurement that is the most accurate reflection of the undertaking future market power.

Market shares tell us little about market power if entry into the relevant product market is easy. In contestable markets (characterised by opportunities to enter and exit markets with ease), the incumbent has no market power, for it is continuously threatened by undertakings ready to enter the relevant market. There is controversy on the correct economic approach to the identification of barriers to entry. This is mostly because the economics scholarship is not useful when applied to competition law. Economists are divided into two camps: some, following Bain, define entry barriers as those factors which allow existing undertakings to raise prices without inducing entry. Others, following Stigler, define entry barriers as those costs that a new entrant must face which were not incurred by incumbents when they entered.

But the economic debate between Stigler and Bain over what is an entry barrier is not particularly helpful in the application of competition law. 1) it confuses the question of determining market power with the question of whether the person holding that market power should be penalised.Whether a monopoly should be condemned is a secondary question, depending for example upon whether its behaviour lowers economic efficiency, so the criticism that Bain classifies certain efficiencies as entry barriers misses the mark. 2) the debate is also unhelpful because the concern of competition law is not how the market will behave in the long run, but on whether in the short run the undertakings are able to raise prices and reduce output.

In both US and EU competition law, less effort is spent defining entry barriers and more is devoted to asking whether new entry is timely, likely and sufficient to counter the reduction in output and increase in price. 1) entry must be likely to occur in time to quell anticompetitive behaviour; 2) entry should be profitable, taking into account the sunk costs that entry entails; third, entry must be of a sufficient scale to reduce the market power of the existing undertaking. In the EU factors like a big technological lead over others, an effective sales network, or a successful advertising campaign that brings customer loyalty are entry barriers because they give the undertaking technical and commercial advantages over rivals. In the EU something like advertising is an entry barrier per se, but under the US Guidelines the question is whether the sunk costs of advertising by a new competitor can be recouped via successful entry, or whether the risk of failure is so great as to dissuade entrants.

Striking a balance between a too wide and a too narrow conception of entry barriers is not easy. The Community's wider conception of entry barriers might be motivated by a policy designed to make it as easy as possible for undertakings to participate in the marketplace, guaranteeing disciplined pluralism. The Community's approach may also be justified in light of economic evidence that entry in the EU market is in fact difficult, where entry has been by small-scale undertaking that occupy niche markets without challenging the larger undertaking, suggesting that the differences in the way markets work in the US and the EU may require different regulatory norms.

Certain goods need spare parts or maintenance. Consumers buy one printer, but make several purchases of ink cartridges throughout the lifetime of the printer; consumers buy a photocopier and require servicing when it breaks down. Market for the additional goods and services consumers need after the initial purchase is a separate product market for the purposes of competition law. Post-Chicago economic theory suggests that there may be a separate market in spare parts if certain conditions obtain: 1) the primary product (the cash register) is expensive as compared to the spare parts; 2) the primary product has a long life time; 3) the costs of the secondary product (the after sales repair service) are uncertain. In these situations it is profitable for the undertaking selling the primary product to exploit the fact that the consumer lacks information about the total cost of spare parts over the lifetime of the product, or is ‘locked in’ because switching to another primary product is more expensive than buying spare parts for the original. In these scenarios there could be market power in aftermarkets, justifying the Court’s narrow market definition in Hugin. The aftermarket is worth monopolising because a small increase in the price of undertaking parts sold to their current owners is tolerated so long as the extra costs are less than the cost of switching to a different machine where the spare parts are cheaper. The Commissions Market Definition Notice acknowledges this kind of analysis. Again the Notice de facto overrules the Court’s earlier approach in favour of a standard which is in line with contemporary economics, but on the basis of a policy designed to maximise consumer welfare, not a policy designed to safeguard the economic freedom of independent repairers. This more economically inspired standard was applied in Pelikan/Kyocera.

The post-Chicago aftermarket theory is controversial. It bases on specific market circumstances that appear unrealistic for some. For example, it assumes that the buyer is uninterested in working out the total cost of the goods he buys, or unable to work this out. In particular, in a competitive market for primary products, sellers are often eager to compete by offering inexpensive secondary products. The exercise of market power in aftermarkets is limited if the manufacturer is concerned about adverse effects on its reputation. If its aftermarket prices are too high, consumers may cease purchasing the primary product. In the long run then, exercising power in an aftermarket is counterproductive.

In recent cases the Commission has begun to move away from defining a market in order to identify market power, and towards testing directly whether the undertaking in question has market power. This can happen in the context of goods characterised by product differentiation because the Commission is able to obtain the data necessary to identify consumer preferences. Two merger cases exemplify the application of this emerging method to determine market power by the Commission. In Bayer Healthcare/Roche the parties to the merger sold anti-acid over-the-counter medicines. Bayer had the Talcid brand and Roche the Rennie brand. In Germany their combined market share was 30–35 per cent and the Commission considered whether the two medicines were particularly close substitutes so that they might raise prices. It considered replies from customers, competitors and pharmacists to conclude that the two products were poor substitutes: Rennie was considered as a relatively ‘simple’ drug, marketed through mass consumer advertising, while Talcid was in closer competition with a range of other medicines marketed through pharmacist endorsement.

In the Austrian market Rennie had a market share of 40–50 per cent and was the clear market leader, while Talcid had a market share of 10–15 per cent. However, the combined market share of 55 60 per cent overstated the effect of the merger because in this country too Rennie was positioned at the more casual end of the market, and Talcid was a distant substitute, so an increase in the price of Talcid would not result in more sales of Rennie. consumers would substitute for other high-end drugs.

In Johnson & Johnson/Guidant the parties were active in the market for endovascular stents. The merger would have reduced the number of undertakings from three to two. Competition was largely based upon brand reputation. Considering the three major undertaking, the Commission, relying upon a survey of doctors, discovered that the products of the merging undertaking were each others closest competitors. Doctors would substitute Johnson & Johnson most readily with Guidants product and vice versa - 62 per cent said Guidant was the best substitute for J&J and in 78 per cent of the responses it was the first or second best. There was no comparable closeness of substitution with the other brands. This evidence indicated that a merger would remove J&J’s strongest and closest competitor. The effect would be to eliminate head-to-head competition, causing higher prices and a slower rate of innovation.

In the first case closeness of competition was considered even when the market shares did not suggest there would be a dominant position in order to test whether the branded goods might be close competitors, while in the second case the close substitutability was helpful in predicting that the merger would lead to an impediment of effective competition.

The method deployed in these two cases made it unnecessary to prove dominance indirectly by reference to market shares. This method does not remove the need to consider entry barriers, or the ability of competitors to rebrand their goods to try and compete against the undertaking that has market power. There is nothing in the method just described that suggests it should be limited to merger cases, where the competition authority is predicting the future effects on the market should the merger go ahead. It is equally possible to apply this approach in situations where the competition authority is reviewing the previous conduct of undertaking if anticompetitive effects are shown to exist, then it can be presumed that the undertaking in question has market power, without the need to identify a product and a geographical market. This method allows a competition authority to identify undertakings that hold significant market power and are able to exploit it by monopolization, reducing output and increasing price. Foreseeable consumer harm is proven directly without the need to identify a relevant market.

It cannot be denied that sellers of branded goods are able to set prices that are higher than the competitive price Armani jeans are more expensive than unbranded jeans even if the only difference is the label. Economists use the notion of monopolistic competition to describe markets with differentiated products, for each manufacturer has a degree of market power, but not to the extent that a monopolist holds market power. In these markets, which represent most of the goods that final consumers buy, competition is often not based uniquely on price but also on other attributes of the product, be it brand image or quality. The question for competition policy is whether a undertaking selling branded goods can ever have enough market power to behave anti competitively and to cause damage to the values that competition laws seek to protect.

In some decisions, the relevance of brand image was noted. Commission noted that there was low substitutability in the minds of consumers between luxury cosmetics and other ranges of cosmetics, so that the former were a separate market, and in United Brands, the undertakings strong brand was evidence of dominance, although in other cases the facts showed that brands were irrelevant to consumer choice, but risk of using a strong brand as the only source for identifying market power for the application of competition law is that we could find market power in every imperfectly competitive market. There seem to be certain circumstances where market failures may result in the context of branded goods primarily because of product differentiation.

Considering product differentiation as the source of market power arises in the relatively novel notion of unilateral effects in merger analysis. US Merger Guidelines suggest that in markets with differentiated products, competition between some brands may be keener than among others. While the relevant market may include all similar products, the degree of competition between brands A and B is greater than that between brands A and C or B and C. In these circumstances, a merger between brands A and B may allow the merged entity to raise the price of one or both products, knowing that consumers are unwilling to switch to other brands. This approach was applied in the FTCs case against the merger between Heinz and Beech-Nut, the second and third largest manufacturers of jarred baby foods. The leading undertaking, Gerber, had a 65 per cent share of the market and was the undisputed leader, while Heinz and Beech-Nut trailed with 17.4 per cent and 15.4 per cent of the market respectively. However, the FTC observed that there was strong competition between the Heinz and Beech-Nut brands, and that a merger would eliminate this, leading to higher prices by the merged entity. While there was no market power in the traditional sense of high market shares, the Court identified a risk to consumer welfare, without needing to define a market. Market power was identified directly.

The kind of market power found in the Heinz case could be identified using the traditional tools we explained above. The evidence suggested that supermarkets bought Gerber no matter what because it was a must-stock brand. Then the supermarket chose either Heinz or Beech-Nut as a second choice. Accordingly, the relevant market from the perspective of the supermarket was the second brand of baby food. If one applied the SSNIP test one would find that if Gerber were to raise prices, it would not lose sales, but that if Heinz did, it would lose sales to Beech-Nut (and vice versa), and if, after the merger, the Heinz/Beech-Nut entity raised prices, it would be profitable as the supermarket had no other product to offer as a second choice brand. Though, this exercise is unnecessary: we do not need to define the market or to compute market shares when the data tell us that anticompetitive effects are foreseeable if Heinz and Beech-Nut join forces. In recent cases the Commission has begun to move away from defining a market in order to identify market power, and towards testing directly whether the undertaking in question has market power. This can occur in the context of goods characterised by product differentiation because the Commission is able to obtain the data necessary to identify consumer preferences. Two merger cases exemplify the application of this emerging method to determine market power by the Commission.

In Bayer Healthcare/Roche the parties to the merger sold anti acid over the counter medicines (medicines that treat heartburn and acid-related gastric disorders). Bayer had the Talcid brand and Roche the Rennie brand. In Germany their combined market share was 30–35 per cent and the Commission considered whether the two medicines were particularly close substitutes so that they might raise prices. It considered replies from customers, competitors and pharmacists to conclude that the two products were poor substitutes: Rennie was considered as a relatively ‘simple’ drug, marketed through mass consumer advertising, while Talcid was in closer competition with a range of other medicines marketed through pharmacist endorsement. In the Austrian market Rennie had a market share of 40–50 per cent and was the clear market leader, while Talcid had a market share of 10–15 per cent. However, the combined market share of 55–60 per cent overstated the effect of the merger because in this country too Rennie was positioned at the more casual end of the market, and Talcid was a distant substitute, so an increase in the price of Talcid would not result in more sales of Rennie: consumers would substitute for other high-end drugs. In Johnson & Johnson/Guidant the parties were active in the market for endovascular stents (specialised medical equipment). The merger would have reduced the number of undertakings from three to two. Competition was largely based upon brand reputation.

Considering the three major undertakings, the Commission, relying upon a survey of doctors, discovered that the products of the merging undertakings were each others closest competitors. That is, doctors would substitute Johnson & Johnson most readily with Guidants product and vice versa: 62 per cent said Guidant was the best substitute for J&J and in 78 per cent of the responses it was the first or second best. There was no comparable closeness of substitution with the other brands. This evidence indicated that a merger would remove J&J’s strongest and closest competitor. The effect would be to eliminate head to head competition, causing higher prices and a slower rate of innovation.

In the first case closeness of competition was considered even when the market shares did not suggest there would be a dominant position in order to test whether the branded goods might be close competitors, while in the second case the close substitutability was helpful in predicting that the merger would lead to an impediment of effective competition.

The method deployed in these two cases made it unnecessary to prove dominance indirectly by reference to market shares. This method does not remove the need to consider entry barriers, or the ability of competitors to rebrand their goods to try and compete against the undertaking that has market power.

The application of this novel method requires caution. In US v. Oracle the Department of Justice argued that in a market for a certain type of business software, Oracle and PeopleSoft (the two entities seeking to merge) were each others closest competitors. The evidence showed that when Oracle competed against PeopleSoft, consumers would get discounts that were 9–14 per cent greater than when Oracle sold the good with no competitors vying for the same customer. Court held that this evidence was insufficient on its own. It merely indicated that Oracle and PeopleSoft often meet on the battlefield and fight aggressively but it did not show whether the same aggressive competition also obtained when Oracle was facing other competitors.

The evidence must show that in markets where one of the two closest competitors is absent, the other is able to raise prices because it feels no competitive pressure from anyone else. The evidence must also show that when Oracle and PeopleSoft go head to head, there is not also a third supplier who is forcing the prices down.

There is nothing in the method just described that suggests it should be limited to merger cases, where the competition authority is predicting the future effects on the market should the merger go ahead. It is equally possible to apply this approach in situations where the competition authority is reviewing the previous conduct of a undertaking – if anticompetitive effects are shown to exist, then it can be presumed that the undertaking in question has market power, without the need to identify a product and a geographical market. This method allows a competition authority to identify undertakings that hold significant market power and are able to exploit it by monopolization - reducing output and increasing price. Foreseeable consumer harm is proven directly without the need to identify a relevant market.

When applying Article 81, there has been little substantive analysis of market power. But three themes comes from the case law and Commission documents, which show that different policy concerns have. Commission has applied a jurisdictional concept of market power, and devised safe harbours using market shares. The lowest safe harbour is found in the Notice on Agreements of Minor Importance. In the Notice the Commission begins by stating that agreements between undertakings where the market share is below a certain threshold do not appreciably restrict competition. In agreements between competitors the threshold is an aggregate market share of 10 per cent (joint venture agreement where one undertaking has 4 per cent and the other 3 per cent of the relevant market). In agreements between undertakings who are not competitors (distribution agreement) then provided neither party has a market share exceeding 15 per cent the agreement benefits from the Notice. But, the Notice goes on to say that there are certain types of agreement that do not benefit from this exclusionary rule. In the context of agreements between competitors, the following agreements are forbidden even if the joint market shares are below the 10 per cent threshold: 1) agreements fixing prices; 2) agreements limiting output or sales; 3) agreements which allocate markets or customers. The Notice fails to take account of the case law of the Court of Justice, which suggests that some agreements are between undertaking of such little economic significance that competition law should not apply at all.

The next thresholds are found in Block Exemption Regulations, and the format is the same as the Notice. The Regulations provide that undertakings with market shares below a given amount, and which enter into agreements that do not include certain terms that are expressly forbidden, have their agreements exempted automatically.

The relevance of the discussion in this chapter is captured by this assessment by he US Supreme Court. Because market power is often inferred from market share, market definition generally determines the result of the case. As the EU moves towards an enforcement policy based upon an economic approach, the identification of market power and the tools necessary for that task are of growing importance. However, there is a tension: on the one hand, the Commission is increasingly using market definition tools that rely on economic analysis and which are premised upon discovering whether a undertaking has market power in the neoclassical sense. On the other, the Court’s case law in Article 82 suggests that the meaning of market power is associated with commercial power. The means to define markets that the Commission is developing seem to be inconsistent with the notion of market power being used by the Courts and the Commission itself. This apparent tension can be resolved by indicating that Community competition law is in transition. There is a political commitment to embrace economic methods, but this is developing against a background of decades of enforcement premised upon the economic freedom model. But until a transition is made to a particular economic theory of competitive harm, there will be tensions and contradictions in the Community's application of competition law.

There are two more specific lessons that emerge from this chapter. The first is about the role of economic analysis in identifying market power. There is no doubt that since the coming into force of the Merger Regulation, market definition has become more sophisticated: it is in the merger cases before the Market Definition Notice that the Commission began to use the hypothetical monopoly test, and where it is now pioneering novel methods that allow the direct identification of market power, dispensing with market definition and market shares. The merger decisions tend to embody a more economics based approach than decisions about past infringements. Though, as we have seen, the Commission does not apply the hypothetical monopolist test regularly. More generally, the market definition plus market share approach is an interesting illustration of the resilience of certain economic paradigms.

Judging market power by market shares is an approach closely associated with the SCP paradigm, which was challenged by the Chicago School. Market share analysis remains at the heart of competition law inquiry because it provides a relatively simple rule of thumb to identify markets where competition is at risk. The chapter also suggests that the Commission has been less than enthusiastic in applying Chicagoan approaches, but has accepted some of the insights of the post Chicago paradigm, for instance aftermarkets and direct proof of market power. The second lesson from this chapter is about the role that policy plays in the definition of the relevant market. Even technical legal issues may be resolved on the basis of the underlying policies that animate the law. In this chapter we have seen some decisions where the Commissions definition of the relevant market can be characterized as strategic. market is identified in order to achieve a specific regulatory objective. Pay TV is seen as a separate market, because the Commission wishes to apply competition law to promote the development of this industry, or to safeguard pluralism.

Chapter 5: Abuse of a dominant position

5.1 Introduction

Dominance is not unlawful, but an undertaking is in a dominant position it is in obliged, where appropriate, to modify its conduct so as not to impair effective competition on the market regardless of whether the Commission has adopted a decision to that effect. Dual obligation on dominant undertakings, to avoid acts that harm competition and to modify their practices if they are likely to harm competition, forces them to observe the markets they operate in and to monitor the effects of their commercial practices, which may become illegal if market circumstances change. Every act of a dominant undertaking is laden with risk, in particular when even commercial behaviour regarded as normal may constitute abuse within the meaning of Article 82. Article 82 states that any abuse by one or more undertakings of a dominant position within the common market or in a substantial part of it shall be prohibited as incompatible with the common market insofar as it may affect trade between Member States. Such abuse may, in particular, consist in: 1) directly or indirectly imposing unfair purchase or selling prices or other unfair trading conditions; 2) limiting production, markets or technical development to the prejudice of consumers; 3) applying dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage; 4) making the conclusion of contracts subject to acceptance by the other parties of supplementary obligations which, by their nature or according to commercial usage, have no connection with the subject of such contracts. Article 82 contains a general section, which prohibits the abuse of a dominant position, followed by a non-exhaustive list of examples of behaviour that may constitute abuse. In contrast with Article 81, there is no stated purpose in the language of Article 82.

The abuse doctrine has four roles. 1^st is to protect the market from dominant undertakings when these reduce output and raise prices. 2^nd is to protect the market from dominant undertakings when they harm competitors so as to obtain the power to reduce output and increase prices. These two roles are based upon economic theories. This represents a neoclassical approach, while the second is representative of post-Chicago theories that undertakings behave strategically to gain market power. 3^rd role for the abuse doctrine is to protect other market participants from the acts of dominant undertaking. Under an economic freedom model, dominant undertakings are the major reason for competition policy. They have the commercial power to harm others, like, competitors, customers or consumers. 4^th role for the abuse doctrine is to protect the internal market. These last two roles are explicitly political, and correspond to two of the core values of EU competition law.

The reform process is designed to affect the concept of dominance and the concept of abuse. This suggests that protective scope of Article 82 should also be refocused so as to apply only to practices that harm consumers. Both the concept of dominance and the protective scope of the abuse doctrine should be moulded by considerations of consumer welfare. The effect is to restrict the protective scope of Article 82.

5.2 Abuse of a dominant position (BA/Virgin case study)

The role of the abuse doctrine can be examined by considering a recent decision - British Airways/Virgin. The dispute centered on rebate schemes that British Airways provided for travel agents. Travel agents buy tickets from the airlines and sell these to travelers. They make profits by a commission, which BA pays on the basis of the number of tickets they sell. BA offered travel agents additional financial incentives in the form of rebates if they sold more of its tickets. In 1993 Virgin complained to the Commission about BAs marketing schemes whose effect, it claimed, was to reduce the incentives for travel agents to sell tickets of competing airlines. The Commission defined the relevant market as that for air travel agency services in the United Kingdom. It found that BA was the dominant purchaser of these services. BAs market share in the total of air ticket sales handled by travel agents was between 39 and 46 per cent, while its competitors had market shares below 10 per cent. Combined with the fact that BA offered flights to many more destinations compared to its competitors, BA was an obligatory business partner of travel agents. Travel agents could not operate without selling BA flights as part of their portfolio.

The strategies are anticompetitive because they dissuade rivals from entering a new market or force rivals to exit, and in the long run allow dominant undertakings to raise prices, having removed all significant competitors. Often the strategies in question are exactly the kind of procompetitive response that we would expect from an undertaking whose market position is challenged. Thus a distinction needs to be drawn between responses by competitors that increase welfare (competition on the merits) and those that reduce it (anticompetitive exclusion). A simple example of anticompetitive exclusion is Ordover and Willig’s theory of predatory product innovation: an incumbent may design a new product aimed at diverting sales away from the rival. If successful, and provided re-entry is costly, the incumbent is then able to raise prices to an anticompetitive level, for he now dominates the market. The new product’s introduction is deemed predatory when the recovery of the costs incurred in developing the new product can only occur if the competitor exits the market, and not from the sales of the product.

If we analyse BAs strategy of offering incremental rebates to travel agents that sell more BA tickets under this paradigm, it may be described as a mechanism for raising rivals costs. BAs rivals have to offer similar, if not more generous, financial incentives to travel agents in order to remain in the market. Cost of competing against BA is increased through Bas marketing scheme. This strategy can be successful provided three conditions are satisfied: 1) the incumbent must be willing to pay to exclude the entrant more than the entrant is willing to pay to stay in the market; 2) new entrant must be unable to find alternative means of entering the market; 3) excluding undertaking must have market power so that it can exploit the disadvantage suffered by its rival. According to this legal test for determining whether BA’s tactics are in breach of competition law is following: 1) whether the dominant undertakings conduct raises rivals marginal costs; 2) if the undertaking that has adopted the cost-raising strategy can subsequently increase prices after rivals costs have been increased; 3) whether rivals have any effective counterstrategies to render the incumbents tactics worthless; 4) whether the incumbent can justify its strategy as increasing economic welfare.

From perspective of economic freedom, the concept of abuse can be extended to protect market participants from abusive tactics of dominant undertakings. This protection is justified by economic and political means. From an economic perspective, competition law should protect all undertakings that are threatened by an dominant undertakings activities, not only undertakings that are as efficient as the dominant undertakings. 1) it is not easy to determine whether a undertakings threatened with elimination is more efficient than the incumbent dominant undertakings: unless other undertakings are given an opportunity to establish themselves on the market, new undertakings will find it hard to enter; 2) when a new undertakings enters it is likely that the dominant undertakings will be more efficient, it will have an established distribution network, experience of the market, and generally lower costs.

Safeguarding pluralism is an important means of guaranteeing healthy markets, and the discipline of Article 82 is necessary for this reason. The standard of proof in the post-Chicago paradigm is too high, from this perspective. From a political perspective, a tougher line against dominant undertakings can be justified in another way. Economic power of dominant undertakings lower political power of the state.

According to liberal conception, the ability to participate in the market free from unacceptable constraints is a right to be safeguarded regardless of utilitarian considerations about the welfare effects of such protection.

Post-Chicago and the economic freedom paradigms is offering an explanation for the Commissions finding. The Commissions decision shows an uneasy tension, sometimes favouring a consumer welfare view premised upon some post-Chicago ideas, sometimes supporting the economic freedom paradigm. The reasons why the loyalty rebates BA offered to travel agents were condemned were following: 1) they removed the travel agents freedom to select its customers, and this regardless of any possibility for the travel agents or competing airlines to minimise or avoid effects of the loyalty schemes; 2) by offering discriminatory commissions to travel agents they distorted competition among them; 3) they harmed all BAs actual and potential competitors, and therefore harm competition in general and so consumers, rather than only harming certain competitors who cannot compete with BA on merit. The first two grounds are based on economic freedom conceptions, penalising a dominant undertakings for interfering with the freedom to trade of other market participants, while the third ground is premised upon a conception of abuse much closer to the post-Chicago theories by noting that consumer welfare is reduced when the dominant undertakings exclusionary tactics harm competitors who are as efficient as BA, and implicitly tolerating the exclusion of less efficient competitors. There is a similar ambiguity in the CFIs ruling that upholds the Commissions decision. The CFI held that as a result of BAs tactics: agents were thereby deterred from offering their travel agency services to airlines in competition with BA whose entry into or progress in the UK market for travel agency services was thereby necessarily hindered. Some aspects of the CFIs reasoning suggest that the Court was concerned about the exclusion of as efficient competitors. CFI added that BA was unable to offer any economic efficiency justification for the rebate schemes, leaving it to infer that the reward schemes were designed to oust rivals.

But CFI retains also a focus on the economic freedom model, by considering the losses of travel agents independence, and also in deciding that the actual effects of Bas practices are irrelevant. Having proven that BA had market power and that its competitors lacked the resources to compete against it, then Court follows that Bas practices are able to have adverse effects. This can be criticised for penalising a undertakings that has financial resources to devise effective marketing strategies. Court sad that there is an infringement of Article 82 when a undertakings attempts to oust rivals, because the fact that the hoped for result is not achieved is not sufficient to prevent a finding of abuse. Courts second response was that the growth of competitors was modest and that without the rebates it may legitimately be considered that the market shares for those competitors would have been able to grow more significantly.

Commissions action against BA needs to be seen in the wider context of the liberalisation of air transport. The aim of the Commissions policy is to allow the entry of new airlines and to offer greater choice for consumers.

Another point is that while the decision might be joyed upon in an economic freedom paradigm, there is also some attempt to justify the adverse finding using economic analysis. The roots for the reform of Article 82 are inherent in the current abuse doctrine. However, one major limitation in redefining the protective scope of Article 82 is the Courts general statement on the nature of abuse in the seminal cases, which has been regularly repeated by the judiciary. Article 82 covers practices which are likely to affect the structure of a market where, as a direct result of the presence of the undertaking in question, competition has already been weakened, and which, through recourse to methods different from those governing normal competition in products or services based on traders performance, have the effect of hindering the maintenance or development of the level of competition still existing on the market.

The most common concern in EU competition law has been practices that force rivals out of markets and/or prevent the entry of new competitors. These are at the centre of the Commission’s reform process, where the tension between the economic freedom and post-Chicago models of abuse is most acute.

5.3 Exclusion of competitors

In certain markets competitors are in a situation of economic dependence vis a vis the dominant undertaking. Deutsche Bahn case: The dominant undertaking held a statutory monopoly over the supply of rail transport services in Germany. These services were used by transport undertakings to move containers from ports in Northern Germany, Belgium and the Netherlands. Deutsche Bahn sold its services in a discriminatory manner. DB charged less to Transfrecht, transport undertaking that operated from the German port which was 80 per cent owned by Deutsche Bahn, and more to operators transporting containers from Belgium and the Netherlands. The impact of this was an increase in container traffic to Hamburg and a decrease of traffic through the Dutch and Belgian ports. Deutsche Bahn promoted its services at the expense of those of the Dutch and Belgian rail operators and also promoted the interests of its transport subsidiary. The Commission considered this a particularly serious abuse because it negated the EU aim to develop international combined transport services in the EU. Rivals costs may also be raised when there is no economic dependence. Rivals costs can also be raised indirectly through contractual agreements with customers that make it more expensive for competitors to lure customers away from their contracts with the dominant undertaking. Example: long-term contract with a penalty clause can raise rivals costs, because in order for the rival to induce the customer to switch, his price has to make it worthwhile for the customer to break the current contract, pay the penalty clause and buy the product from the new entrant. Dominant undertaking will often use both direct and indirect means to eliminate competitor.

From an economic perspective, these practices can be challenged because they raise the cost to the dominant undertakings rivals, reducing their output, allowing the dominant undertakings to act like a monopolist once competitors exit. Professors Krattenmaker and Salop have proposed a two-step analysis for exclusionary practices: 1) whether the conduct in question unavoidably and significantly increases the costs of its competitors; 2) whether raising rivals costs enables the excluding undertakings to exercise monopoly power, to raise price above the competitive level. The anticompetitive conduct takes place in two stages: 1) in the first stage rivals costs are raised; 2) in the second the newly acquired power is exploited. Significant difficulty in translating this method into competition law enforcement is that competition authorities intervene at the first stage. If intervention is delayed until the second then any remedy is unlikely to restore the market to the status quo ante, as competitors have already exited and new entrants may take time to have an impact. It is imperative for the enforcement action to take place at the time when the dominant undertaking is excluding rivals. At this stage one should establish whether the practices of the dominant undertakings are likely to eliminate rivals and give the aggressor the power to raise prices. This creates the risk of Type 1 errors in that the competition authority might punish efficiency enhancing conduct by dominant undertakings.

The EU penalises the dominant undertaking upon a showing that the costs to rivals have been raised, and often it does so without even quantifying expressly by how much those costs have increased. This is not because the EU presumes that the dominant undertaking will exploit its dominant position upon proof of anticompetitive exclusion, rather that exclusion in itself is an abuse of dominance, suffocating the economic freedom of market players.

Below-cost pricing can exclude or discipline competitors, increasing the market power of the dominant undertakings. Below-cost pricing harms rivals by encouraging consumers to switch away from the products offered by the rival. This technique is known as predatory pricing and defined as a response to a rival that sacrifices part of the profit that could be earned under competitive circumstances were the rival to remain viable, in order to induce exit and gain consequent additional monopoly profit (definition of predation).Main challenge in penalising these practices is that it is hard to distinguish between fair, aggressive pricing and unfair, predatory pricing.

US Supreme Court has expressed doubts about the likelihood that undertakings would engage in below-cost pricing, by considering the Chicago School critiques. Decision to engage in below-cost pricing is very costly, as it is unclear how long the prices have to be set below cost in order to drive out competitors. Also last undertaking standing must be able to raise prices to anticompetitive level so as to recoup the losses it has suffered.

In EU however is scepticism against this view. Because when the dominant undertakings has vast resources with which to finance an aggressive pricing campaign, and the prey only enough resources to fight the predator for a short time, then below cost pricing is rational. Also a bout of below-cost pricing may suffice to establish a reputation for below cost pricing, thereby deterring entry. Dominant undertaking needs to reduce price in only one market, making the victim believe that it is capable of a more widespread pricing tactic. Also a cost signaling strategy can be designed to make the prey believe that the dominant undertakings has lower production costs than the prey. Also belowcost pricing might be a tactic to soften up a potential takeover target, lowering its value before launching a takeover bid. These strategies can lead to the competitor exiting the market, or to it refusing to make new investment and innovate, and at the same time may deter other potential entrants, effects that benefit the dominant undertakings and reduce economic welfare.

But these economic ideas have had small impact upon legal standards. Courts have suggested that below cost pricing is abusive when the prices of the dominant undertaking fall below a certain measure of cost. In EU, prices below average variable cost (AVC) are always predatory, while prices below average total cost (ATC) (variable costs plus fixed costs, costs that do not vary according to quantities produced) but above AVC may be predatory when there is evidence of a plan to drive out a competitor. The justification for a cost based standard is that a undertaking setting a price which does not cover its variable costs is acting irrationally because it fails to recover any of its production costs.

Commission condemns distribution agreements by dominant undertakings when their effect is to foreclose entry by potential competitors. Most stringent attack has been on discount and rebate schemes offered to distributors. Commissions investigations are characterised by an extremely detailed examination of the kind of discount and rebate scheme, and the cases defy easy classification. But Commissions approach is basically a two-step: 1) detailed analysis of the distribution contracts is designed to answer one question - do the rebates induce loyalty? The answer is inevitably in the affirmative, and is explained this way – the distributor, before the discount, is already buying a large amount of goods from the dominant undertaking. Rebate offered if the distributor buys even more of its supplies from the dominant undertaking must induce loyalty; 2) loyalty inducing rebates may be justified only if the discount reflects genuine cost savings. If rebates are loyalty inducing, the rebate is abusive from the day the contract is signed.

The Commissions policy has been criticised for deploying a per se prohibition against many discounting practices which are normal business behaviour. Granted the dominant undertakings is able to offer discounts, but only if these reflect the dominant undertakings cost savings, or because the purchaser confers additional benefits on the dominant undertakings. Dominant undertakings may not use price strategically. This is unnecessarily restrictive because it ignores the possibility that a rebate scheme could be the most effective means of giving incentives to distributors to market goods aggressively, and therefore it is a cost-saving strategy.

The Community is interested to see the entry of new competitors, on the assumption that their entry enhances consumer interests. The economic approach embraced by the Community is not a total welfare approach based upon maximising efficiency, but a standard premised upon facilitating market access by a plurality of undertakings. Requirement to measure foreclosure is sufficient. Accordingly, in Deutsche Post the relevant finding which should become compulsory for all the commercial contracts is the following: “Successful entry into the mail order parcel services market requires a certain critical mass of activity and hence the parcel volumes of at least two cooperation partners in this field. By granting fidelity rebates to its biggest partners, DPAG has deliberately prevented competitors from reaching the critical mass of some 100 million in annual turnover.

Here Commission measures the degree of foreclosure, finding that the dominant undertakings prevented any entry into the relevant market by offering rebates to all major customers. Conversely, had the rebates been offered to only one customer, leaving the remainder free to utilise other parcel delivery undertakings, then the rebate should not be deemed unlawful. Under the economic freedom paradigm, modified to ensure consumer welfare via the presence of a plurality of sellers, it is unnecessary to go further and speculate whether the market would be more efficient with more participants.

Dominant undertakings may extend its power into other markets, thereby excluding competitors from a market it does not dominate. Leveraging is a general term that encompasses a variety of strategies that a undertakings might use to extend its market power from one market to another, for instance by tying, rebates or predatory pricing. Two issues come up: 1) whether one should be concerned about leverage; 2) whether Article 82 is an adequate legal tool to address this practice. One of leveraging forms is tying.

Hilti manufactures a popular nail gun used in the building trade. To operate it the user has to purchase a cartridge and nails. The cartridge explodes when the nail is shot. Once the nails are used up, a new cartridge and a new set of nails is needed. Cartridges and nails are used in fixed proportions. Hilti was dominant in three separate markets: nail guns, cartridges and nails. The Commission accused Hilti of abuse because it tied the sale of cartridges to that of its nails. That is, consumers wishing to buy a cartridge from Hilti had to buy Hilti nails. The effect of this tactic was to make entry of competing nail manufacturers more difficult. An economist looking at this factual matrix would say that Hilti is not guilty of anticompetitive practices because there is no evidence that, having eliminated all competing nail manufacturers, prices will increase: Hilti, the monopolist in the cartridges market which seeks to create a monopoly in the market for nails, is unable to obtain any extra profits through leverage because it cannot earn any extra profits on the sale of nails. This conclusion can be explained with a this example. Lets believe that consumers are willing to buy the Hilti gun if the price for the spare parts (nails plus cartridge) is E 6. At a higher price, the consumer switches to another nail gun, so E 6 is the monopoly price for the two components.

Say the market for cartridges is monopolised by Hilti and the market for nails is competitive, and the competitive price for nails is E 1 and nails and cartridges are sold separately. In these circumstances, the monopolist will sell the cartridge at E 5. If the cartridge price is set higher, consumers switch to another nail gun, making E 5 the monopoly price for the cartridge. Tying cannot raise price, because the monopoly price for the tied goods is E 6, whether or not there are competing nail manufacturers. Whether the monopolist ties or not, his monopoly profits are the same. Now a new, efficient, nail producer enters the market and sells nails at E 0.50. Should this worry Hilti to such an extent that it will want to tie the sale of its cartridge to that of its nails? The answer is no, in fact Hilti should welcome this new, low-cost entrant because it allows Hilti to increase the price of the cartridge to E 5.50 (the total price for the consumer is still E 6, now Hilti pockets an extra E 0.50). If an efficient nail manufacturer comes along, Hilti should shut down its inefficient nail production plant and benefit from the efficient suppliers existence. The upshot of this example is that Hilti has no economic reason to tie the sale of the cartridge to its nails. The maximum price that Hilti can set for nails and cartridges is E 6, so it cannot make greater profits by selling both goods. In fact, as the example shows, its profits increase when a more efficient nail seller arrives. Leverage may injure competing nail manufacturers, but has no adverse welfare consequences. This is the Chicago School position on tying. This view suggests that tying can only be explained on grounds of efficiency. If it allows the manufacturer to ensure that consumers have the most suitable nails for the gun they buy, or it is actually cheaper for the manufacturer to make the two together than for the buyer to assemble the goods. It has been argued that the justification given by Hilti, that tying was designed to ensure that the nail gun operated safely was probably plausible. Unless the manufacturer has no safety concerns, it will prefer not to tie the cartridge to the nails, because in a more competitive nail market each nail manufacturer will try to make nails that work best with the Hilti gun, and this will increase demand for the monopolised product. This analysis should not be taken to mean that tying is always procompetitive and that leverage is an unrealistic strategy.

So far there is suggested that leverage is a profit-maximising strategy but, leveraging can also give the dominant undertakings power to increase prices towards its own customers when the dominant undertakings leverages into an aftermarket. In these contexts there is a market for the primary product (the nail gun in Hilti) and a separate market for the secondary products (the cartridge and the nails). Hilti was not analysed in these terms, but we can use the facts to think about why tying would be profitable if the purchaser of the nail gun did not think about the costs of the spare parts when he bought the gun. Having acquired the gun, this person is locked into the market for spare parts. Now it is profitable to raise the price of nails. Here, in contrast to the Chicago School model, there are two monopoly prices. Example: there is a monopoly producer of guns and spare parts and the price for the gun is E 100, the cartridges and nails cost E 6 and the lifespan of the gun is 100 cartridges, and the consumer notes that the total price for this system is E 700 (if the consumer thinks in these terms, then there is no aftermarket). Assume a new entrant comes in selling the cartridge and nails for E 3.A Chicago economist would suggest that the monopoly should not tie the gun and the nails but just raise the price of the gun and maintain the same amount of profit. But, this suggestion only works if the consumer thinks about the gun and the spare parts as one system, one single product. If the consumers choice of gun depends only on the price of the gun and the consumer does not take into account the price of the parts, then the price of the gun cannot go up in response to entry in the spare parts market. Guns monopoly price is set independently from the monopoly price of the spare parts. W when the new, more efficient entrant comes into the parts market, the incumbent loses profits on that market and cannot compensate by increasing the price of the gun. Leveraging to gain monopoly in the parts market is profitable because it is a new, distinct market which is worth monopolising. Leveraging can help extend a dominant position from one market to another. But the effect this has on consumers can be more than just a higher price. In the Commissions decisions two additional harmful effects are identified: the consumers reduced choice and the harm to innovation.

Microsoft was found to have abused Article 82 in tying its operating system with Windows Media Player (WMP), making market access for competing media player software more difficult. The Commission reasoned that because consumers would find WMP pre-installed, they would use it and not install other competing media player software. The effect is that content providers would wish to format their music so it could play on WMP and fail to format music to be played on other media player formats, thereby giving Microsoft a competitive advantage that had nothing to do with the quality of WMP but everything to do with Microsoft's dominance of the operating systems market. The Commissions concern was that this would stifle competition in innovation in this market. According to the Commission, the normal competitive processes for several undertakings to participate to invent better and better media player software, and tying would foreclose market access because nobody would be interested in investing resources in competing software which cannot be sold as a result of the tie. More generally, the Commission also feared that tying in this market would reduce investment in other types of software because the profitability of new software developments would be stifled if Microsoft were to design a competing product and tie it to the operating system.

Tying shields Microsoft from effective competition from potentially more efficient media player vendors that could challenge its position. Microsoft thus reduces the talent and capital invested in innovation of media players, not least its own, and anti competitively raises barriers to market entry. Microsoft’s conduct affects a market which could be a hotbed for new and exciting products springing forth in a climate of undistorted competition.

Most significant aspect of the decision is the Commissions choice to explore the ways in which tying would cause harm to consumer welfare. The methodology is much closer to the economic models of leveraging than that displayed by the Commission in its earlier tying case law. In contrast to the earlier cases where it was not clear what specific harm the consumer or society suffered when cartridges and nails were sold jointly, here the economic welfare losses of tying are set out in detail.

To sum up, there are plausible economic reasons for being worried about leveraging. But it must be remembered that from an EU perspective, the anticompetitive nature of tying is not based only upon concerns over efficiency. In Tetra Pak 2 for example, tying is characterized as an abuse because it deprives the customer of the ability to choose its sources of supply and denies other producers access to the market. It is important to note that this passage embodies two distinct concerns: 1) the harm to consumers is associated with the exploitation of a dominant position. But the Commission, until Microsoft, never investigated in detail what the nature of this harm entailed, so that it seems as if consumer choice is beneficial in itself; 2) the passage refers to the exclusionary potential of tying but there is no attempt to test how far the exclusion of rivals harms consumers.

5.4 Damage to other market participants

Dominant undertakings also abuses its dominant position when it restricts the freedom of non-competitors. This occurs in distribution agreements where the harm is the distributors inability to exercise their commercial freedom to look elsewhere for goods, to diversify their business and to look for other suppliers. Primary basis for penalising the dominant undertakings is the foreclosure of competitors, but regular mention is also made of the injury which distributors suffer. In Michelin the ECJ held that abusive discounts tend to remove or restrict the buyers freedom to choose his sources of supply and as a result bar competitors from access to the market. The general principle animating the protection of other market participants was explained by Arved Deringer, who sad that the purpose of the competition rules is to preserve the freedom of choice of those who transact business in the market as well as the free interplay of supply and demand in competition. The exploitation is therefore an abuse where the dominant position is used to restrain or eliminate the freedom of decision in competition either of competitors or the consumers.

An dominant undertakings may be penalized for causing harm to distributors independently of any effect on consumer welfare. As the ECJ stated it in United Brands, it is important to preserve the independence of small and medium sized undertakings in their commercial relations with the undertaking in a dominant position.

The most remarkable exemplification of this approach can be found in the Michelin 2 decision. The Commission found that specialized tyre dealers were placed despite themselves in a situation of economic dependence that makes Michelin an unavoidable partner. Having established such power over retailers, the Commission went on to condemn a raft of rebate schemes as unfair – some rebates were awarded so late that dealers were forced to sell at a loss, placing them in a precarious situation also with respect to subsequent negotiations with Michelin. Commission also found that dealers experience of Michelins bonus scheme was arbitrary. The contracts with dealers included a service bonus linked to the quality of service offered by the dealer to his customers, and the level of quality was something that Michelin would measure. This was condemned because it allowed the manufacturers representative to put strong pressure on the dealer as regards future commitments and allowed him, if necessary, to use the arrangement in a discriminatory manner. In these passages the Commission is condemning unfair contractual practices that harm distributors, not anticompetitive practices.

Article 82 (c) provides a specific example of abuse that is designed to protect customers of the dominant undertakings, where abuse may consist in applying dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage. This provision is often applied in cases where anticompetitive rebates are found, because the rebates discriminate among dealers, thereby placing them at a competitive disadvantage vis a vis each other.

5.5 Other types of abuses

Market-partitioning. Market integration is a core value for EU competition policy and the abuse doctrine extends to these practices. Infringements that jeopardize the proper functioning of the single market, such as the partitioning of national markets are very serious and deserve the highest fines.

In Deutsche Post, the basis of the abuse was a rebate scheme that foreclosed access to the German parcel delivery market, but because the excluded competitors were from outside Germany, this aggravated the abuse: “this walling off of a national market affects the development of trade to an extent highly inimical to Community's interests”.

EU is keen of taking action in excessive pricing cases. Of the few decisions targeting excessive pricing by dominant undertakings, the real motivation seems to have been the concern that prices contribute to the partitioning of the single market. Some decisions targeted the car industry where, until 1993, national approval systems were in place to ensure that imported vehicles conformed to domestic safety standards.

More recently, the policy of market integration has appeared in decisions where the Commission investigates sectors that have been liberalised by Community law. In a Directive on access to the groundhandling market at Community airports (that is, the provision of services like cleaning, refueling and baggage handling), the EU had sought to allow free access to competing groundhandling services, in accordance with the Treaty's provisions on the free movement of services (Article 49 EC). In France, access for groundhandlers at Orly Airport is regulated by Ae´roports de Paris and it charged different fees to different operators, whereby the highest fees were paid by AFS, a company based in the UK.

This benefited the French operator (OAT, a subsidiary of Air France), whose running costs were considerably lower. The Commission found that the price discrimination was an abuse which in the first instance harmed OATs competitors, and in the second instance caused harm between competing air transport services because the higher groundhandling costs would be passed on to them. The price discrimination impaired the smooth functioning of the single air transport market.

Two aspects of this ruling are needed to remember. 1st is that competition law is used to ensure the functioning of the internal market which the Community sought to create by legislative means. 2nd the dominant undertaking was not present in the markets where the abuses took place.

5.6 Different types of defences

Court has said that certain practices are abusive unless objectively justified. In technical terms, this seems to reverse the burden of proof from the Commission to the defendant. But there is little to support this conclusion. It is best to say that there is a space within the context of the definition of abuse for the dominant undertaking to bring evidence to show that the acts in question do not constitute an abuse.

The abuse doctrine applies to distribution agreements in two main circumstances, first, when these are loyalty inducing and/or when they discriminate among dealers. The Court has regularly suggested that price incentives may be objectively justified on economic grounds. In Michelin 2 the CFI stated that: “a rebate system in which the rate of the discount increases according to the volume purchased will not infringe Article 82 EC unless the criteria and rules for granting the rebate reveal that the system is not based on an economically justified countervailing advantage but tends, following the example of a loyalty and target rebate, to prevent customers from obtaining their supplies from competitors”. The basis of the Courts approach is to consider whether the rebate was linked to any cost savings that the dominant undertaking made, so that if increasing the quantity supplied results in lower costs for the supplier, the latter is entitled to pass on that reduction to the customer in the form of a more favourable tariff.

The discussion paper on Article 82 suggests that there can be an efficiency defence in Article 82, modelled upon Article 81(3). The Commissions suggestion is to apply the same four conditions as under Article 81(3). This requires proof of the following: efficiency; that there is no other way of achieving the efficiency so that the abuse is indispensable; that the efficiencies benefit consumers; that the abuse does not eliminate competition. The scope of application of the efficiency defence is

limited to rebate and tying abuses.

Earlier there was noted how in certain circumstances the ECJ has ruled that Article 81 was inapplicable even if the agreement was a patent restriction of competition because the agreement was necessary to promote or safeguard a Community or a national public interest. There is no comparable case law on Article 82. But the same principles that the Court recognised for Article 81 should apply here.

EC Treaty provides that if a undertaking is entrusted with the provision of a service of general economic interest, then it is exempt from the Treaty obligations in so far as this is necessary to provide the service in question, on the basis of Article 86(2). This provision has led to a statutory exemption for several state-created monopolies. Also when the Court in Wouters decided that Article 81 was inapplicable, it did so by reference to its case law on the internal market, suggesting that there is a general principle of Community law whereby the Treaty obligations are inapplicable when their enforcement would undermine legitimate national concerns. Wouters also is about the relationship between competition law as a whole and other public interest considerations, and is authority for the proposition that there are circumstances where other public interest considerations trump the application of competition law.

5.7 New policy directions

The key policy that has underpinned the Community’s regulation of dominant undertakings is to protect the economic freedom of other market participants. The early case law focus was upon considerations of fairness to those who traded with the dominant undertakings, while the case law since the 1980s has placed more emphasis upon the elimination of rivals. In ensuring that dominant undertakings do not thwart the freedom of rivals or other undertakings, the Court of Justice has expanded the meaning of abuse to ease the burden of proving an infringement, and some recent decisions of the CFI have continued this trend by indicating that an abuse may be found even before the exclusion of rivals can be felt. In DSD the Commission summed up its perception of Article 82 in this way, The Court has stated in this matter that a system of undistorted competition, as laid down in the Treaty, can be guaranteed only if equality of opportunity is secured as between the various economic operators. Such equality of opportunity is particularly important for new market entrants on a market in which competition is already weakened by the presence of a dominant undertaking and other circumstances. Small competitors should not be the victims of behaviour by a dominant undertakings, facilitated by that undertakings market power, which is designed to exclude those competitors from the market or which has such exclusionary effect.

Equal access to the market is a right valuable in itself to, without the need to demonstrate that economic gains flow from its guarantee. But economic benefits are perceived to flow on the basis that greater numbers of market participants create the disciplined capitalism which lies at the heart of the Community's economic constitution. The case law suggests that other policies are in play in the application of the abuse doctrine, in particular the protection of a single market. Economic efficiency and consumer benefits might be an indirect result of the application of Article 82, but they are not the central goals of the abuse doctrine.

Chapter 6: The Abuse of a dominant position

6.1 Introduction

Case Merci Convenzionali Porto di Genova. Contractual dispute arose between a hipper and dock workers in the Italian port of Genoa. The shipper claimed that the dock workers had delayed unloading his goods, causing loss. During the litigation it transpired that the dock workers had a statutory monopoly. The Court of Justice referred to Article 82a), b) and c) and suggested that the dock workers failures might constitute an infringement of these provisions either because the undertaking asks for payment for services that have not been requested, or because the prices are excessive, or because the undertaking refused to have recourse to modern technology, which involves an increase in the cost of the operations and a prolongation of the time required for their performance. Commission has avoided developing the abuse doctrine to penalise lazy monopolists. Instead, inefficiency can be the basis upon which the states decision to confer a monopoly on an undertaking is challenged.

6.2 Abuse

Many say that the best prize for a monopolist is the quiet life that the undertaking enjoys after having achieved dominance. Surrounded by entry barriers, it has no need to innovate or to cut costs. However, in some circumstances sloth can be an abuse of a dominant position. This arises in particular where the dominant undertaking enjoys its dominance by virtue of a state-granted monopoly.

In Merci Convenzionali Porto di Genova, a contractual dispute arose between a shipper and dock workers in the Italian port of Genoa. The shipper claimed that the dock workers had delayed unloading his goods, causing loss. During the litigation it transpired that the dock workers had a statutory monopoly. The Court of Justice referred to Article 82(a), (b) and (c) and suggested that the dock workers failures might constitute an infringement of these provisions either because the undertaking asks for payment for services that have not been requested, or because the prices are excessive, or because the undertaking refused to have recourse to modern technology, which involves an increase in the cost of the operations and a prolongation of the time required for their performance.

In spite of these findings, which are derived from a literal interpretation of Article 82, the Commission has avoided developing the abuse doctrine to penalize lazy monopolists. Instead inefficiency can be the basis upon which the states decision to confer a monopoly on a undertaking is challenged. A little more effort has been devoted to regulating excessive pricing, and we explore the Commissions strategy below.

6.2.1 Excessive prices

Commission has little enforced against excessive pricing, but instead the Commission has announced to apply Article 82 to exclusionary abuses. Commission has been active in regulating prices in newly liberalized sectors, in particular in the field of telecommunications. The reason for this is strategic, in two ways. 1^st it is a means of obtaining support for its deregulatory agenda. As one senior Commission official noted, the Commissions actions against excessive prices are aimed particularly at passing on rapidly the advantages of liberalization in terms of price reductions and service developments to consumers is a major objective in order to show as fast as possible the effective consumer benefits and to secure sustained public support for liberalization. 2^nd Commissions strategy has not been to reach decisions under Article 82, but to launch investigations and use these to put pressure on undertakings and National Regulatory Authorities in charge of telecommunications to achieve price reductions. In 1998 it commenced fifteen cases, one against each MS, alleging that interconnection charges for calls from mobile to fixed phones were excessive, but these were dropped as undertakings agreed to reduce charges or the Commission agreed that NRA would address the price concerns. It reached a similar result when it challenged certain charges of Deutsche Telekom to competitors, the Commission dropped the case once satisfied that the NRA would regulate prices.

The role of price control by bespoke regulators is different from that achieved under Article 82. Three examples of price control drawn from the field of telecommunications can help us to see the differences. First, we can consider two important features of telecommunications: access to the local loop and interconnection. The local loop is an (upstream) infrastructure that connects each house to the telecommunications network. This is normally the property of the former state monopoly telephone provider who is now in competition with several other providers of (downstream) telecommunication services. New competitors must have access to the local loop to provide services because it is uneconomical to duplicate this part of the network. For example, a new entrant wishing to provide internet services in competition with the former state monopoly must be able to connect his services to the local loop that is still owned by the incumbent. The second characteristic of telecommunications is that it is a network market, where interconnection between users of different networks is essential for the success of a new entrant – the entrant’s users must be able to communicate with users of other networks. Interconnection prices must be competitive or else consumers will not buy the services of new entrants.

In both of these scenarios, when the owner of the local loop is also the largest provider of telephone services, it has no incentive to facilitate entry of competitors, and so the law may impose access remedies and regulate the prices of access to the local loop and interconnections. Note that in these two contexts (markets exhibiting natural monopoly characteristics and network effects) the regulatory ambition of the Community is not merely to prevent anticompetitive behaviour, but to inject more competition by facilitating the entry of new undertakings by regulating access and price.

The Community imposes duties on National Regulatory Authorities (NRAs) to regulate price, subject to guidance established by the Commission. In regulating prices for access to the local loop the NRA must take into account a variety of policy objectives: 1) that prices foster fair and effective competition; 2) to ensure economic efficiency and maximum benefit of users; 3) to monitor that prices are cost-oriented. Cost-orientation means that the local loop provider can cover costs plus a reasonable return, so that he has the incentive to develop and upgrade the local loop. Furthermore, the price should give incentives to the requiring undertakings to develop competing infrastructure so as to eliminate their dependence on the incumbent’s local loop. The effect of these rules is that prices should be low enough to encourage new entrants, but high enough to encourage the incumbent to maintain the network and stimulate the construction of competing networks. This is a tricky mix of policies, and NRAs have yet to devise the most effective pricing structure. Similar guidance applies when NRAs regulate interconnection prices where account must be taken of the operators investments so as to allow a reasonable rate of return. This achieves two aims: to ensure that the undertaking that grants interconnection has incentives to develop and maintain the infrastructure, and to inject competition by facilitating entry.

These two examples show that there are similarities with Article 82, but also divergences. The reason for intervention (market power) is comparable, but the differences lie in the timing of intervention (which may be ex ante, to prevent excessive prices), in the institution (a specialized body, not a competition authority or a court), and in the variety of policy goals (not just protecting buyers from high prices) that must be taken into consideration when determining an appropriate price.

A second example can be found in Article 30 of the Universal Service Directive. This provides that consumers should be able to switch from one telecommunications provider to another and retain their telephone numbers. Consumers are more likely to switch if they can hold on to their telephone numbers so this facility is very important to encourage new telephone service providers, as it lowers entry barriers. However, switching is not costless. If a subscriber wishes to change from, say, Vodafone (the donor operator) to Orange (the recipient operator), Vodafone incurs costs to ensure that the number is portable. It will charge Orange to recover these costs when the consumer switches. The consumer may well pay nothing directly, but if the charges set by the donor operator are too high, then some of the costs result in higher bills for users of the recipient operator, and consumers will not switch. Article 30(2) empowers the NRA to ensure that the prices are cost-oriented and that if the consumer has to pay for this, the level is not such as to act as a disincentive for the use of number portability. The Belgian regulator recently made use of this provision to regulate the wholesale charges of donor operators. In Belgium, Belgacom, the major provider of mobile phone services, had a market share greater than 50 per cent and faced competition from two relatively new entrants (Mobilstar and Base). Given Belgacoms market position, it had an incentive to hamper number portability. The Court of Justice, in approving the Belgian NRA’s approach to price regulation, noted especially that when the donor is a large undertaking, then effective price controls are crucial to guarantee a competitive market. This example is a little different from the previous one. While the regulator is concerned about injecting more competition in the market, the power to intervene does not depend on finding a undertaking with a dominant position.

On the facts of the Belgian case, the former state monopoly did have a dominant position but this is not a precondition for price regulation under the directive. Moreover, the prices controls applied to all undertakings, as any of the three could be donor operators.

A third example is price controls that apply in relation to goods and services the state decides should be available to all at affordable rates. In Community law, these are labelled universal services. The Universal Service Directive imposes an obligation on NRAs to monitor retail rates to ensure that they are affordable for those on low incomes or with special needs. In this context, the link with Article 82 is non-existent. The market is not regulated to achieve efficiency or consumer welfare: retail prices are controlled to achieve fairness.

Price controls in electronic communications indicates a doctrinal and policy overlap between Article 82 and sector-specific regulation carried out at national level on the basis of a common EU-wide framework. Both share an interest in economic efficiency and consumer welfare. Regulation is wider in its ambit, however: not only is it designed to promote competition rather than merely prevent its deterioration, but other goals (development of the industry and fair access to basic services) are also pursued.

6.3 Refusal to assist to rivals

Sometimes dominant undertaking have an obligation to facilitate the entry of other undertaking, but there is no certainty about when this duty arises. There are four related doctrines: 1) is that a dominant undertaking cannot rescind a contract with a undertaking that is not in competition with it unless there is justification; 2) say that a dominant undertaking must facilitate the existence of competitors; 3) this doctrine applies when the dominant undertaking refuses to share intellectual property rights with competitors. Obligation to share is imposed only if a stricter test is met, which is consonant with the trend towards a consumer welfare model. This is because the person wishing to obtain access to information protected by intellectual property rights must show that this information will be used to provide a new product and benefit consumers; 4) Commission imposes obligations to cooperate with rivals with scant regard to the limitations set out by the ECJ in the case law. 4th doctrine has led to legislation to ensure market access as a means of securing the injection of greater competition into newly liberalized economic sectors.

The Courts first important judgment on refusals to supply illustrates the prominence of the economic freedom approach. In Commercial Solvents a undertaking dominant in the market for a raw material chose to extend its activities in the downstream market for a derivative of that raw material. The Court and Commission focused on Zojas potential elimination as the evidence of competitive harm, to the extent that a number of commentators have suggested that the decision sought to protect a small

European undertaking against the might of its American supplier. This criticism is given further credence by the fact that other undertakings were present in the ethambutol market. There was nothing to suggest that price of ethambutol would rise. It was sad although not considered, that aminobutanol was available from other sources and that ethambutol could be manufactured with other chemicals. That's why Zoja was not necessarily going to leave the market, although it would have to make some adjustments to stay.

The Court returned to consider refusals to contract in Telemarketing. The Court said that there would be an abuse when the dominant undertaking holds a dominant position in a service that is indispensable for the other undertaking, and where the refusal would eliminate all competition from the competitor.

The Court returned to consider this matter in Oscar Bronner. The Court held that there would be an abuse where: 1) a undertaking in a dominant position in one market (the market for newspaper deliveries) refuses to supply a undertaking with which it is in competition in a neighbouring market with raw materials or services which are indispensable to carrying on the rivals business; 2) where the refusal is likely to eliminate all competition on the part of that undertaking; 3) where there is no objective justification for the refusal. In setting out this three-stage test the Court seems concerned about injury to a rival (note the singular in the italicised passages above) not about injury to consumer welfare. In this light, all three judgments (Telemarketing, Commercial Solvents and Bronner) reflect a concern about ensuring that market players remain on the market, rather than about ensuring that markets are efficient or that consumers benefit.

The concern remains with the risk of foreclosing market access, but the Court now only finds an abuse when the undertaking that risk foreclosure are as efficient as the dominant undertaking and still unable to enter the market. Foreclosure of efficient rivals harms consumer welfare by depriving the market of a worthy participant. This suggests that a dominant undertaking has an obligation to continue to cooperate with actual or potential competitors.

In the main cases where the IP right conferred a dominant position on the owner, the Court has required, as with the doctrine examined above, evidence that the refusal to license the IP right eliminates all competition on the part of the undertaking requiring access and that the IP right should be indispensable to carrying on that business in that there are no actual or potential substitutes, so that competition in the downstream market where the IP rights are exploited is eliminated.

The reason why Courts impose an obligation to license only in exceptional circumstances is because there is a presumption that a refusal to license IP rights benefits the economy. The grant of an IP right is made to motivate the creative activity of authors and inventors by the provision of special awards and the imposition of a duty to license would weaken this incentive system, customers are left without new products. Reason for this presumption is that IP rights sacrifice a degree of allocative efficiency in favour of promoting dynamic efficiency by creating a financial incentive to invest. Here IP law and competition law pursue the same objective. Use of competition law to impose obligations upon holders of IP rights to license these would undermine the economic rationale for IP rights and harm consumers. Presumption that a refusal to license is economically efficient may be rebutted, and the ECJ case law suggests that in exceptional cases where the consumer surplus outweighs the dynamic efficiency considerations, then the refusal to license is abusive.

Commission also imposes obligation to license when exceptional circumstances suggest that there is a public interest in remedying the harm suffered by consumers. In reaching this conclusion the Commission makes a prima facie case that the obligation to license causes benefits that outweigh the potential reduction in incentives to innovate. It is then up to the defendant to suggest that this presumption is inaccurate.

Result is when a refusal to license excludes the undertakings that are responsible for innovation, there is a loss in dynamic efficiencies. This is main position to the Commissions understanding of how markets work, for them competition is the key to efficiency.

Once the Commission decides that a refusal to supply is abusive, it orders that he dominant undertaking cooperate with the competitor, but this gives rise to a number of risks. First, the dominant undertaking and the new entrant may collude so that the market conditions are no different. Second, the dominant undertaking may enter into a contract with the competitor at a high price, leading the new entrant to raise its prices to customers, allowing the dominant undertaking to keep its prices high also. Third, the dominant undertaking can operate a price squeeze so that the rival cannot afford to remain in the market. In other words, the dominant undertaking that is forced to cooperate with a rival has no incentive to create a more competitive market. Forcing the dominant firm to allow market access without more is pointless, and the declaration that there is a duty to collaborate does not end the Commissions involvement. The upshot is that the regulatory burden on the Commission might be so heavy that the cost of administering remedies far exceeds the benefits to economic welfare. It has been suggested that when imposing an obligation to cooperate with competitors, the Commission should carry out a cost–benefit analysis that takes into account the following four factors: 1) the benefits to the requiring undertaking; 2) the costs on the dominant undertaking; 3) the costs for the competition authority in monitoring the remedy; 4) the benefits to consumers. It is worth considering how far these considerations have been taken into account when the obligation to supply is imposed: first when the Commission imposes it, and second when enforcement is in the hands of NRAs.

In the case of Commission enforcement most of the costs and benefits have been borne in mind: the benefits in (1) and (4) are necessary for the obligation to be imposed in the first place and the costs on innovation are taken into account in the IP cases. The cost of making the facility available is not taken into account expressly. Court has held that a refusal to grant access to a facility may be justified when it is impractical to grant access. The cost to the competition authority is not mentioned expressly. In both IMS Health and Microsoft the Commission indicated that a third party should be appointed to monitor the supply obligations imposed by the Commission. This is a recognition that the cost of monitoring can be excessive and unpredictable and, as a result, enforcement is privatised. In most cases this privatisation means that a dispute about the grant or cost of access is referred to an expert for resolution. In the Microsoft case monitoring is carried out by a trustee who may also take measures to ensure that Microsoft is complying with the obligation to grant access. This is tantamount to creating an ad hoc regulator for the undertaking. In newly liberalised sectors instead, the EC has been able to delegate the task of regulating access to bespoke National Regulatory Authorities who are better resourced and in a better position to carry out the sort of day-to-day monitoring which the imposition of a duty to allow access requires. Some of the sector-specific legislation requires that the NRA carry out a cost–benefit analysis. However, there is no obligation on the NRA to take into account the costs that it would incur in enforcing the remedy in question, but it might be argued that this consideration is for the NRA to carry out in order to decide which cases to investigate.

6.4 Regulation policy in competition law

The nature of competition law enforcement has certain unique features: 1) Community is using competition law as a tool to benefit consumers in newly liberalised markets. Its efforts in regulating prices in telecommunications are designed to put pressure on NRAs to act under sector-specific regulation. But it is not only in newly liberalised sectors that the case law reviewed in this chapter advances an industrial policy. The early refusal-to supply doctrine was designed to safeguard the economic freedom of small players, and the more recent explanations of the doctrine suggest that the aim of imposing a duty to cooperate with competitors is designed to allow the entry of undertakings that are able to provide better products and services and more choice for consumers. Victim s market share had been increasing during the alleged abuse. Also access remedies are granted with scant attention to trading off benefit to consumers from the remedy with the deterrent effect on innovation.

Chapter 7: Merger policy

7.1 Introduction

Regulation of mergers is the principal instrument by which competition authorities control the structure of an industry. Merger law is necessary for the following reasons: 1) once the market is dominated by one or a few strong undertakings, the competition authorities may not have the resources or the information to pursue every anticompetitive action so that preventing the merger is a more cost effective way of maintaining competition. 2) merger policy may be to the benefit of the merging undertaking, it might be problematic if two undertakings are allowed to merge and become dominant to be assiduously regulated after the merger, or even broken up by the competition authorities some years after the two businesses have been integrated; 3) if fully resourced, the competition authorities might not be able to remedy all the anticompetitive effects of a change in market structure; 4) anticompetitive effects of a merger extend beyond the parties, when a merger creates a dominant undertaking and it raises prices, it creates a price umbrella and other competitors raise prices to the same level.

Combination of factors led to the adoption of the EC Merger Regulation (ECMR) in 1989. 1^st was that the Court of Justice found that some mergers fell under the jurisdiction of Articles 81 and 82. 1973 it held that if a dominant undertaking were to merge, strengthening its dominance, this could constitute an abuse under Article 82.

The basic principle of the ECMR is that mergers are scrutinized before they are implemented. The parties announce their plans to merge, but then put these on hold pending the Commissions decision. Upon notification, merger is reviewed under phase 1 procedures, which are carried out within four weeks of notification. During this period interested parties are able to comment on likely impact of the merger, and the Commission verifies the information it has received and consults MS. At the end of phase 1, five outcomes are possible: 1) transaction is found not to be a merger, so scrutiny under ECMR is unnecessary; 2) merger does not have a Community dimension, so it falls outside of the Commissions jurisdiction; 3) merger is compatible with the Community and may go ahead; 4) merger is compatible with the Community, but amendments are required before it may go ahead; 5) merger raises doubts as to whether it is compatible with the Community, and a phase 2 investigation must be carried out to review transaction in more depth. Phase 2 investigation lasts for four months, and in addition to the consultation processes in phase 1, there is an oral hearing where the Commission, parties to the merger and interested parties participate. Besides formal procedure, there are regular contacts between parties to the merger and Commission. Decisions in phase 2 cases are those of the Commission acting as collegiate body and are discussed among all Commissioners, who also have to give attention to the opinion of the Advisory Committee. Three outcomes are possible at the end of phase 2: 1) merger is compatible with Community and thus may proceed; 2) merger is compatible with Community but may only proceed if modifications are made; 3) merger cannot go ahead because it is incompatible with Community. Majority of mergers have been cleared unconditionally in phase 1, and only nineteen mergers have been blocked, the last one in 2004.

From 1 April 2004 ECMR was rewritten, and new test for the substantive assessment of mergers is the following: “Concentration which would significantly impede effective competition, in the common market or in a substantial part of it, in particular as a result of the creation or strengthening of a dominant position, shall be declared incompatible with the common market.”

7.2 Horizontal mergers

Biggest risk posed by mergers arises when the merging undertakings are competitors and the merger creates or strengthens a dominant position. Analysis starts with considering new undertakings market share, the size and strength of its rivals, potential competition faced by the merged entity and any customer power. In merger cases the countervailing strength of other market actors is taken into account more fully. This is so for two reasons: 1) is that the analysis in Article 82 cases is retrospective while in merger cases the analysis is prospective. In Article 82 cases what other market players might have done is not helpful because one is analyzing historical facts; 2) reason is that to date there has been a different approach by Commission in merger cases, where there is greater reliance on economic analysis, which has led to a more comprehensive appraisal of market power.

Market shares allow for a quick look to see the possibilities of anticompetitive risk, and market shares above 50 per cent are prima facie evidence of dominance. But added refinement in merger cases is that the market share analysis is dynamic. Adding up the market shares of the merging parties today may not yield an accurate estimate of the parties market power in the future. Finding that post merger market shares of the merged entity are low will normally result in the merger being cleared, while if market shares are high, the inquiry extends to considering the competitive strength of competitors, the likely entry of new competitors, and the countervailing strength of consumers. These findings help to measure whether the merged entity is able to act independently of competitors, customers and consumers.

Competitors. In determining is entry likely, the test is if a hypothetical market player would find entry profitable, taking into account the risks and costs of entry, the effects of entry on price, and the anticipated response of the other undertaking in the market. Principal consideration is if new entrants face insuperable entry barriers. EU has favoured a broad conception of entry barriers. In the Commissions practice entry is said to be timely if it occurs within two to three years. Entry is enough if it is able to defeat or deter the merged entity anticompetitive conduct. This approach suggests that the Commission does not trade off short term anticompetitive price increases for long term competitive effects, as a total welfare economic model might recommend. On other side, the threat of new entry must prevent all anticipated anticompetitive behaviour by the merged entity. This means that although entry of the potential competitor may take two years, the merged entity should have no incentive to attract such entry by exploiting its temporary dominant position. This approach is consistent with the priority placed upon consumer welfare. Commission rejects the view that welfare losses for today's consumers can be accepted if the market is more competitive in the future.

Behaviour of the merged entity. In some cases the Commissions analysis of a mergers effects is informed by considerations of whether the merged entity will raise rivals costs after the merger, thereby considering whether the merged entity is itself likely to put up new entry barriers. Extensive analysis of the anticipated exclusionary conduct of the dominant undertaking is only raised infrequently in horizontal merger cases, but there is a pattern. These are cases of network industries where the concern is that the holder of the networks access points has an essential facility which the Commission wishes to ensure is available to all competitors. Concern also arises in other instances where the merged entity has control of an upstream or a downstream market through which it can deny access to rivals.

Consumer power. Powerful buyers can eliminate the risk of dominance if two conditions are met: 1) buyers can obtain the goods elsewhere; 2) all buyers are able to exercise power so that the dominant undertaking is unable to price discriminate, offering low prices to the powerful buyers and high prices to the weak.

Buyer power is better seen as an additional consideration to determine the strength of existing rivals and the viability of potential competition and not as a standalone device that automatically eliminates the risk of anticompetitive effects.

7.3 Vertical mergers

Recent economic analysis has revealed that some vertical mergers can reduce competition, in particular because a vertical merger may foreclose rivals. The foreclosure analysis is comparable to the leveraging. Foreclosure can be two way: 1) foreclosure can occur in the non-dominated market. Take for example a undertaking dominant in the upstream market acquiring a undertaking in the downstream market. The merged entity could foreclose competition on the downstream market by limiting supplies, or raising the price in market. Effect of this is that the merged entity can use its dominance in the one market to leverage its position in the other market; 2) foreclosure can occur in the dominated market. Dominant manufacturer in one market merging with a non-dominant but significant manufacturer in other market. If merged entity refuses to buy supplies from competitors, this could lead to the exit of competing manufacturers in first market, consolidating the dominant position upstream. Foreclosure can also be achieved when a vertically integrated dominant undertaking refuses to sell to competitors on the downstream market.

7.3.1 ECMR

The anticipated anticompetitive effects in a vertical merger depend upon the merged entity behaving in a specific manner. Unlike horizontal mergers where proof that the merger creates a dominant position or substantial market power is normally sufficient to establish that the merger will substantially impede effective competition, the Commission must show that the merged entity will have, post-merger, an incentive to foreclose and the capacity to foreclose, and that foreclosure can have significant anticompetitive effects. The Commission can satisfy the first element by showing that foreclosure of rivals would be in the merged entity's commercial interests.

The way the CFI has interpreted this requirement appears to suggest that if leverage would take the form of an abuse of dominance, it seems very unlikely that the merged entity will have the incentive to engage in leverage. The second step is to evaluate the capacity to leverage. To prove this, the Commission should consider whether the market context makes leveraging profitable. The final step is to prove that the merger will have anticompetitive effects. Assessment of vertical mergers requires a consideration of two sources of competitive harm: anticompetitive foreclosure or the creation or strengthening of a dominant position. It follows that a vertical merger can be blocked even if the parties are not dominant pre-merger in the up- or downstream markets and even if there is no post-merger dominance, provided there is foreclosure.

7.4 Conglomerate mergers

Mergers are conglomerate when they are neither horizontal nor vertical, the products of the parties may be complementary or completely unrelated. Conglomerate mergers are now rare. Commission is increasingly interested in the potential anticompetitive effects of conglomerate mergers. According to the Commission’s practice, these mergers raise no anticompetitive concerns when the products of the parties are unrelated, a merger between a car manufacturer and a wine producer cannot raise competition concerns. Of more worry for the Commission are mergers where the undertakings produce two complementary products, or products that consumers tend to buy together. The Commission’s principal concern is that the merged entity may engage in leverage, by tying the two products and thereby harming undertakings that sell only one. In contrast, economists normally see ties as efficiency enhancing because they result in lower prices to consumers. In addition to tying concerns, the Commission fears a conglomerate merger because it may create undertaking with an incentive to behave strategically and harm competitors. A third concern arises when a undertaking acquires a potential competitor, thereby safeguarding its market power. This last scenario is best seen as a horizontal merger. We consider each of these three theories of competitive harm in turn, paying particular attention to leveraging, which is the most controversial theory of harm.

7.4.1 Guinness/Grand Met case study

Commission observed how a merger combining the activities of the two leading suppliers of spirits in the world would confer upon the new undertaking (Diageo) the ability to offer a wide portfolio of drinks to its customers. The anticompetitive risks of this were noted on the Greek market, where Guinness had strong sales of whisky, rum and gin while Grand Met was strong in the market for brandy, ouzo, tequila and liqueurs. The Commissions concern was not that the parties would come to dominate one of these spirit sectors, but that by holding a strong position in so many types of spirits (and a dominant position in gin, rum and brandy) the portfolio held by the new company would create a position of dominance in the sales of all spirits. The Commission identified four effects of the merger: 1) customers would prefer purchasing from the merged entity because it offered a wider range of goods, and would be able to offer appealing discounts; 2) the merged entity would have lower costs because of the economies of scale and scope that result from integrating two distribution networks. Neither of these advantages are a good reason for blocking the merger: they show instead that the merger leads to efficiencies;

3) concern was more rational from an economic perspective, undertaking with a powerful portfolio of brands can offer bundles of drinks or tie the sale of its popular drinks with the sales of other spirits, thereby increasing its sales. Alternatively, the undertaking could demand that bars promote the spirits sold by it, which the buyers would have to accept since they could not afford not to stock the famous brands sold by the new undertaking; 4) undertaking could threaten not to supply the retailer with certain must stock drinks if he did not buy all goods from it. These strategies would strengthen the position of the new undertaking in all product ranges and reduce the sales of other competitors who lack comparable portfolio power.

The Commission allowed the merger by requiring the parties to eliminate the source of the anticompetitive risk, terminating Guinness right to distribute Bacardi rum in Greece, and by the divestiture of some whisky brands. The Commissions concerns about leveraging in this decision can be interpreted in three different ways: 1) judged from an economic perspective, leveraging is anticompetitive only if the effects are the consolidation of monopoly power, either in the markets that the undertaking already dominates or in new markets; 2) perspective is to suggest that the Commission was interested in consumer welfare. Commission interest in competition cases is not about generating total welfare, but about consumer welfare; 3) way of interpreting this case is to suggest that the decision protects rivals. On this perspective, safeguarding competitors economic freedom is the basis upon which the decision is reached.

Rather it is that the Commission focuses on the immediate negative impact of the merger and has a pessimistic attitude towards the ability of the market to restore competition when competitors exit the market. In sum, the Commission believes in competitors, the DOJ believes in markets.

7.4.2 Tetra Leval case study

The ECJ has accepted that mergers may be blocked when leveraging concerns arise, but has placed some limits to the Commission’s powers in regulating these mergers in Tetra Laval. This case concerned a merger between Tetra Laval, dominant in the market for carton packaging, and Sidel, which held a strong position in a related market (plastic PET bottles). The Commission opined that post-merger the undertaking would leverage its dominance in the cartons market to create a dominant position in the PET market. As in the earlier cases, leverage would take the form of tie-ins (whereby bottlers would have to buy both carton and PET packaging systems from the merged entity), or predatory pricing or rebates to induce customers to purchase exclusively from it, thereby weakening other competitors on the PET market. On appeal, the CFI and ECJ acknowledged that conglomerate mergers may be blocked, but installed a more rigorous standard for determining the anticompetitive risks from leveraging. The CFI suggested that, in contrast to horizontal mergers, it could not be presumed that conglomerate mergers would lead to anticompetitive effects; in fact these mergers might be beneficial to consumers. And the ECJ complemented this sentiment by indicating that the Commission had to prove its theory convincingly. The analysis of a ‘conglomerate-type’ concentration is a prospective analysis in which, first, the consideration of a lengthy period of time in the future and, secondly, the leveraging necessary to give rise to a significant impediment to effective competition mean that the chains of cause and effect are dimly discernible, uncertain and difficult to establish.

That being so, the quality of the evidence produced by the Commission in order to establish that it is necessary to adopt a decision declaring the concentration incompatible with the common market is particularly important, since that evidence must support the Commission’s conclusion that, if such a decision were not adopted, the economic development envisaged by it would be plausible.

The standard of proof set out by the Court is not different from that in other types of merger, but because the analysis is more complex than merely deciding if a merger creates a dominant position, every step leading to a conclusion that the merger can lead to leveraging risks must be proven. The Court of Justice’s judgment suggests a four-step inquiry: 1) determining whether leveraging is possible, if the economics of the markets make it feasible in theory for one undertaking to use its dominance in one market to enhance its market position in another, related market; 2) determining whether the undertaking has an incentive to leverage; 3) proving that leveraging would result in the anticompetitive effects predicted; 4) considering the possibility that behavioural commitments would remedy the anticompetitive risks created by the merger. This is the same kind of analysis that is carried out in vertical merger cases.

Courts test makes it difficult to bring a successful leveraging case as the Commission has to show convincingly that the merged entity has the power to foreclose, that competitors are likely to exit, that buyers are unlikely to thwart the leveraging strategy, and that new entry is unlikely.

7.5 Different merger remedies

Structural remedies are preferred because they are an immediate solution to the anticompetitive risk: they remove any threat posed by dominance by creating a new competitor or strengthening an existing undertaking. No long-term monitoring by the Commission is required. Commission has started to put greater controls in place to monitor divestitures to ensure that the seller does not act opportunistically and to ensure that the buyer is a viable competitor. There are four types of control. 1) Commission considers whether the assets the parties propose to divest are likely to allow the purchaser to act as a competitive force. To achieve this the Commission has occasionally required that the merged entity divest a wider range of assets, including products where no competition concerns arise if their sale is necessary to create a viable competitor; 2) it ensures that before the assets are divested the merged entity does not cause them to lose competitiveness pending the identification of a buyer and their sale. This is achieved by the so-called hold separate commitment by which the parties agree to keep the assets viable and which is monitored by a trustee who may have the powers to impose any measure necessary to ensure the business remains viable; 3) Commission can reject the proposed buyers; 4) is to monitor the merged entity's behaviour after divestiture when there is a risk that it could act to thwart the purchaser. Process of divestiture has become complex and costly, in particular if ex post behavioural remedies are necessary to protect the new entrant.

Behavioural remedies. The most common behavioural remedy is to require the merged entity to cooperate with competitors, so as to ensure the continued presence of other market players.

Access remedies are also granted under Article 82 EC, and that these have been agreed in the vertical mergers reviewed above. But when access remedies are provided in merger cases Commission does not always assess whether the facilities to which access is granted are essential. The normal Article 82 safeguards are not applied in all cases. In particular license of intellectual property rights is normally only compelled if the undertaking obtaining the license is seeking to develop a new product. This safeguard is absent in merger remedies.

The behavioural remedies may be divided into two categories: 1) merged entity had considerable buying power, which allowed it to acquire exclusive rights to attractive content to broadcast via its channels. The remedy was for the merged entity to give away the exclusive rights so as to allow other competitors to obtain some content, either by starting a competing satellite pay-TV network or through other forms of broadcasting; 2) merged entity was able to foreclose access of competing digital TV operators because it controlled the set-top box technology. A new entrant would require access to that set-top box in order to market his products, and the parties undertook to grant such access. The effect of these remedies was that the Commission allowed a merger, which created a monopoly pay-TV supplier in Italy, and subjected it to a range of behavioural commitments designed to facilitate the entry of competitors.

These two types of behavioural remedy have in common that they are designed to facilitate the presence of one or more competitors: in other words, they achieve the same objective as structural remedies. But the Commission prefers structural remedies because behavioural ones require continuous monitoring to ensure compliance, e.g. by requiring parties to submit regular reports to describe how the remedy is being complied

with. The Commission has reduced monitoring costs in two ways: 1) by asking that the parties establish an arbitration agreement whereby performance by the merged entity is monitored by the beneficiary; 2) by enlisting national regulatory authorities to monitor post-merger commitments on its behalf.

Key to merger remedies has been to guarantee the presence of one or preferably more actual or potential competitors to the merged entity. A promise by the merged entity not to cause the exit of rivals may thus be an acceptable merger remedy, in line with the Commissions policy.

When merger remedies are considered, the Commission’s work is of a different nature from when it enforces EC competition law in more traditional scenarios like hard core cartels. The power to impose remedies allows it to regulate markets in a detailed manner, a process achieved through negotiation with the parties to the merger. The parties may accept that the merger can only go ahead subject to the remedies required by the Commission, or simply withdraw the merger.

In exercising its powers the Commission states that it does not assess the compatibility of business strategies with the common market and restructure an individual undertaking business strategy. Structural remedies make parties restructure the industry in the way the Commission considers works in the best interest of consumers. Behavioural remedies are not limited to the parties avoiding committing acts incompatible with EU competition law. Behavioural remedies are designed to achieve a well-functioning market after the merger.

The Commissions merger policy, consistently with Articles 81 and 82, is grounded in two concerns: 1) that a merger may harm consumer welfare; 2) that the merged entity might have the power to exclude rivals. It may thus be said that the aim of merger policy is to promote or ensure the maintenance of rivalry in the market. But in some cases one can detect the wish to safeguard other interests.

The broadcasting cases suggest a wish to guarantee media pluralism, and some decisions seem to be designed to protect the continued existence of particular undertakings. These decisions safeguard rivalry plus other interests.

Commissions decisions suggest that for the vast majority of cases the governing principle is that mergers are prohibited when they stifle economic opportunities, whether of consumers or of competitors. While there are resonances of the economic freedom concept, the tenor of the decisions taken as a whole is to support such freedom as a means to guarantee consumer welfare, not as an end in itself. The ECMR in this respect did not undergo the same paradigm shift we noted in the context of Article 81; instead the Commission pioneered an economics-based approach in this field of activity. You can see that Commissions decisions do more than protect consumer welfare: mergers are blocked or modified, or indeed even allowed in spite of their anticompetitive potential, in accordance with wider Community interests. Commissions jurisdiction is limited in two formal ways: 1) only mergers above a certain turnover threshold may be reviewed at Community level; 2) national interests may be invoked to refer a merger back to a Member State. Protectionism is on the increase and the Community is struggling to halt this, creating a 3) informal means by which national interests can be safeguarded.

Thresholds are complex, requiring a) a combined aggregate turnover of 5 billion ecus and b) aggregate Community turnover of each of at least two of the undertakings of more than 250 million ecus. And if each of the undertakings achieves more than two thirds of its aggregate Community wide turnover within one and the same MS, Commission does not have jurisdiction. The thresholds are politically motivated attempts to safeguard national merger policy.

7.6 National interests in merger cases

Commission has also articulated more clearly the grounds upon which it will decide to refer the merger to a Member State: in four cases it has based its decision on considering whether the NCA was in best place to examine the transaction, and the criteria to judge this included the existence of national markets as well as that authorities expertise in the economic sector in question. MS may also invoke public interest grounds for demanding that merger be referred to the national authorities so that the merger clearance respects national public policy considerations. Under Article 21(4) the range of public interests includes only three: public security, the plurality of the media and prudential rules. Member States may request that other public interests be taken into account.

The Commissions recent reforms, are a clever way of attempting to create a true European merger policy, by facilitating the review by the Commission of mergers that have an appreciable effect on trade between Member States.

Chapter 8: Oligopoly

8.1 Introduction

Economists are divided over the level of regulation required in oligopoly markets because while high degrees of concentration might make markets less competitive, it is also possible to find lively competition in an oligopoly market. The price leadership exercised by one undertaking in an oligopoly is a recognised mechanism by which undertakings can tacitly collude by sending signals to each other about their actual and anticipated conduct. Most economists see no difference between express collusion and tacit collusion, because the effects are the same: prices rise and output falls. But from a legal perspective, the distinction between express and tacit collusion is crucial: the former can be pursued under Article 81 EC, but it is not clear how far tacit collusion can be regulated – in fact, it seems to be immune from Article 81.

While the original version of the ECMR only prohibited mergers that create or strengthen a dominant position as a result of which effective competition would be significantly impeded, Commission was quick to extend its scope to oligopoly markets causing coordinated effects by stating that in such markets the undertakings would be collectively dominant. While merger law allows Commission to prevent the creation of market structures where express and tacit collusion are more likely, ex post regulation of anticompetitive behaviour in oligopoly markets can only tackle express collusion.

8.2 Control of mergers

Commission considers that in an oligopoly market, the undertakings hold a collective dominant position if it can be shown that together, in particular because of correlative factors which exist between them, undertakings are able to adopt a common policy on the market and act to a considerable extent independently of their competitors, their customers, and also of consumers. Coordinated effects in oligopoly markets have similar effects to the exploitation of a dominant position of a single undertaking, therefore, as with single undertaking dominance, entry barriers must be high, and competitors and consumers weak. The distinguishing feature that is necessary to establish collective dominance is the interdependence of the oligopoly, which establishes the likelihood that the undertakings will coordinate behaviour. Following on from the game theoretical insights, a test to determine collective dominance includes the following four elements: 1) a highly concentrated market; 2) economic conditions which make it easy for the undertakings to coordinate their behaviour; 3) conditions that deter each undertaking from deviating from the common policy; 4) lack of power of the other undertakings and lack of potential competition or buyer power.

This structure provides a mechanism for explaining the Commissions decisions reached to date as well as being broadly in line with economic theory. First two elements facilitate coordination so that prices are higher than under competitive conditions. But when prices are high, it would be profitable for one undertaking to cheat and give some customers secret discounts, as this would increase its sales and profits at the expense of rivals. Tacit collusion can only be stable if those who cheat in this way are identified and can be penalised by the other undertakings. Also tacit collusion can only be sustained if buyers are unable to find alternative supplies, and so are forced to pay the higher prices set by tacit collusion. Third and fourth conditions guarantee that the effects of tacit collusion are stable.

The degree of concentration among the undertakings is very high in all cases: this is an important condition as cooperation is possible only when the number of undertakings is small. Concentration provides a prima facie indication of possible coordinated effects, requiring further scrutiny. Majority of the Commissions decisions measure concentration by considering the markets concentration ratio: that is, what market share the leading undertakings in the oligopoly have.

Before determining that the market is sufficiently concentrated, the next crucial question is whether the market conditions facilitate tacit collusion. While this is an intimately fact-specific exercise, one feature is a necessary condition for a finding that anticompetitive effects may result: market transparency. Transparency is crucial for two reasons: 1) parties must be able to work out their strategies on the basis of the performance of others: there can be no coordination unless the parties can see what the strategies of the others are; 2) transparency is necessary to detect any deviation from the agreed practice by one undertaking, thereby allowing the others to react. Significance of transparency has evolved with the case law. Gencor: CFI noted the importance of transparency as a mechanism necessary to facilitate retaliation and in Airtours CFI confirmed the dual purpose of transparency. As a matter of evidence, market transparency may be proven in different ways. There may be close structural links between undertakings, or prices may be easily observable because the goods are sold on the open market, or there may be reliable data that the parties are able to obtain which allows them to monitor each others actions, for instance from customers, or the parties may all be members of a professional association where information is available. The parties can make the market transparent by their own efforts. In one decision concerning a merger of stainless steel tube pipes, the price lists were publicly available and all constructed by reference to one particular type of tube pipe, and both consumers and competitors knew that prices for other tube pipes were calculated by reference to that one type, facilitating price comparisons and price leadership by one supplier. Transparency can also be facilitated by the homogeneity of products, which allows undertakings to compare each other's market performance more easily than when products are highly differentiated.

While transparency is necessary, it is not sufficient to create the degree of interdependence that leads to collective dominance, and a range of other factors can be deployed to determine whether coordinated effects are likely. Commissions decisions can be criticised for giving inconsistent messages about what other factors are also necessary. In one decision it stated that symmetrical cost structures were a precondition for the existence of a stable duopoly. This point finds favour in some economic quarters. Symmetrical market power means that the parties have less incentive to compete as they do not have a decisive comparative advantage over each other, and if the profits resulting from tacit coordination are similar for each member of the oligopoly this will induce cooperative strategies more easily than otherwise.

Certain market characteristics are more likely to lead to coordinated effects, but this does not prove that in their absence there can be no coordinated effects a fortiori. But when entry barriers are high, coordinated effects may occur even in growing markets. Other regular finding in mergers that have been blocked is product homogeneity, and the economic literature confirms that coordination is more likely with homogeneous products than in cases of consumer goods.

Court has clarified that structural links are not necessary if other economic factors are sufficient to facilitate coordinated effects. From an economic perspective, this is the correct conclusion: the key question is whether the market facilitates tacit coordination and allows the parties to monitor each others behaviour. Commercial links might help establish these economic conditions but are not necessary.

Upon determining that the undertakings are able to present themselves on the market as a collectively dominant entity, the analysis then turns, as in all merger cases, to considering the countervailing power of other suppliers and buyers. Merger case law adopts a more perceptive assessment of actual and potential competition than the dominance case law under Article 82. Relevant question is if a sufficiently vast number of small entrants can thwart the behaviour of the dominant oligopoly by offering a number of short haul holidays to neutralise the anticompetitive effects of the oligopoly’s restriction in output.

Important consideration is that Commission must prove that merger is the likely cause of an impediment of competition. A causal link may be established when the merger eliminates a maverick competitor that had made coordination difficult, or the merger leads to more symmetrical cost structures, or greater transparency, or improved retaliation mechanisms, all of which enhance the capacity of the parties to coordinate their action.

In EU competition law it seems that, provided the Commission can show that the merger facilitates tacit collusion, it may be blocked, and this can occur even if there is already evidence of collusion on the market.

8.3 Express collusion

Economic analysis used to identify situations of collective dominance in merger cases is also helpful in examining the behaviour of cartels. Cartels are the major concern of competition authorities, described as ‘cancers’ in the economy. Most cartels have been uncovered in oligopoly markets where the possibilities to coordinate behaviour are significantly facilitated by the market structure. A relatively contained number of undertakings, fungible goods, a falling market, the lack of innovation, and weak buyers. And cartels can collapse when there is no deterrent effect, as happened for instance to the cartel between the four major providers of electrical equipment in the United States (General Electric, Westinghouse, Allis-Chalmers and Federal Pacific). The cartel self-destructed when in 1957 one major customer (the Florida Power and Light Company) pressurised Westinghouse to reduce its prices in exchange for a promise that it would buy all its requirements from Westinghouse. However, General Electric found out and matched the lower price offered to Florida Power. This led to a price war until discounts in the winter of 1957–8 reached 60 per cent. GE and Westinghouse learned the lesson of the 1950s when in the 1970s they coordinated again in the turbo-generator market. This time the parties included a mechanism of deterrence. They both offered a price protection plan to their customers by which a discount on new orders would also apply retrospectively to all orders for the past six months. The effect of this was to make cheating highly expensive. This says that if the cartel is poorly designed, it is likely to end by itself, without the intervention of a competition authority. But economic insights about the conditions for successful collusion are irrelevant for competition authorities in cartel cases. First, competition authorities distinguish between express and tacit collusion and indicate that they are merely empowered to catch the former.

Second, a finding of express collusion is based only upon whether there is evidence to show that the parties have communicated directly with each other. Express collusion is per se unlawful. The per se rule is justified because the risk of Type 1 errors is low. When competitors agree to fix prices or allocate territories, it is unlikely that this has a pro-competitive justification. Conversely, the harm from Type 2 errors is significant: if the cartel is successful, it can cause severe harm.

The Commissions policy in cases of express collusion is strengthened by three interrelated features: 1) a wide meaning to notions of the words agreement and concerted practice; 2) wide powers to obtain information and the possibility to infer an agreement from market behaviour; 3) and the availability of incentives on parties to cartels to inform the Commission of their complicity in an infringement of competition law.

Article 81 applies to three forms of coordination: agreements, decisions by associations and concerted practices. In the present context only the concepts of agreement and concerted practice are relevant. Agreement is found when parties express their joint intention to act on the market in a specific way. It is not limited to contractual agreements, nor to any form of communication.

A concerted practice differs from an agreement for two reasons: 1) it is less intense. It is a form of coordination between undertakings which, without having been taken to a stage where an agreement has been concluded, knowingly substitutes for the risks of competition practical cooperation between them; 2) a concerted practice requires conduct on the market after the coordination and a causal link between coordination and subsequent conduct. ECJ has sad, that the concepts of agreement and concerted practice are intended to catch forms of collusion having the same nature and are only distinguishable from each other by their intensity and the form in which they manifest themselves.

In long-running cartels the Commission has observed both forms of express collusion, with some rounds of coordination taking the form of agreements and others the form of concerted practices. This reflects economic studies, which suggest that cartels may be unstable and that differing levels of coordination might be expected throughout the lifetime of a collusive endeavour. In these situations, the Commission can characterise a long-running cartel as an agreement and/or concerted practice. That is, just because the parties cooperation varies, this does not mean that each round of cooperation is a distinct infringement. When the collusive scheme is a complex cartel, containing various forms of cooperation, it is a single infringement. This facilitates the Commissions investigation in two ways: 1) it allows the Commission to catch collusion which manifests itself sometimes in the form of an agreement and sometimes in the form of a concerted practice. Otherwise the Commission would be forced to take separate actions against the undertakings for each round of collusion; 2) this allows the Commission to hold each undertaking responsible for the whole agreement even if they did not take part in every aspect of it.

It is just in oligopoly markets that a concerted practice can serve to reduce uncertainty. This is the market structure that has been found in the majority of the cases where concerted practices have been uncovered, but the Courts definition fails to declare this to be an essential element. The second limb of the definition of concerted practice is that the efforts to coordinate behaviour have an effect on the undertaking future conduct. But the burden of proof for the second limb is very low. First, provided the undertaking stays in the market, the Commission is entitled to presume that its conduct was influenced by the coordination efforts.

Second, Article 81 prohibits collusion whose object is the restriction of competition, and so a concerted practice can be prohibited regardless of the effects. Third, the Commission does not have to prove that the subsequent conduct has an anticompetitive effect, but only that the undertakings subsequent conduct is caused by the cooperation, which can be presumed. Private dissent from collusion is insufficient to escape liability. This makes it impossible for undertakings to rely on the fact that they did not follow the recommendations of the meeting in their subsequent behaviour. If they lowered prices contrary to what had been negotiated. The test is so easy to satisfy by the Commission and so hard to rebut by the undertaking that it has little practical value.

In the majority of cases, the Commission proves agreements and concerted practices by obtaining documentary evidence of collusion, like, faxes, emails or recorded conversations. These documents must show either that the parties agreed to behave in the market in a specific way (so that the documents can prove the presence of an agreement), or that the parties have exchanged information about their future intentions, thereby facilitating coordination and showing the presence of a concerted practice. The Commissions standard of proof is low, because all that has to be shown to find an infringement is that a undertaking has participated in a meeting where anticompetitive activities were discussed.

The principal legal issue that has arisen when the Commission relies on documentary evidence has been about the powers that the competition authority has exercised to obtain this information and the degree to which these infringe the undertakings' fundamental rights. This has given rise to extensive case law that has circumscribed the Commissions powers somewhat. Commission has considerable powers to look for evidence, which are set out in Regulation 1/2003. These include the power to enter premises of undertakings that are suspected of engaging in anticompetitive activities and to seize relevant documents, as well as the power to interview employees.

In certain circumstances an agreement may be inferred because the behaviour witnessed on the market, plus other evidence, allows the competition authority to infer that there has been collusion. Neither the other evidence nor the behaviour is sufficient, but together they allow an inference of collusion.

While wide definitions of agreement and concerted practices and the power to infer agreement from economic evidence could facilitate the Commissions burden of proving express collusion, the most successful plank of the Commissions anti-cartel policy has been the use of leniency programmes that offer undertakings who confess to having participated in an anticompetitive agreement and provide the Commission with sufficient evidence to challenge the other parties to the cartel an immunity from fines, or at least a substantial reduction on the fine that would otherwise be set. Immunity from fines is available if a undertaking is the first to provide evidence that enables the Commission to adopt a decision to carry out an investigation or enables the Commission to find an infringement. Fines can be reduced for undertakings that provide evidence that strengthens the Commissions ability to prove the infringement, and the reduction is more generous for the first undertaking that provides this evidence. The benefits of immunity can be substantial given that the Commission has the power to impose a fine of up to 10 per cent of the undertakings turnover. Leniency programmes are based upon the same economic logic as collusion. According to game theory, collusion makes sense because it is the way for profit to be maximised. Leniency policy uses the same logic to suggest to undertakings that it is in their interests to confess.

8.4 Tacit collusion

There is a gap in competition law. Ex ante merger control prevents tacit and express collusion, but ex post competition law controls only express collusion. When the Commission uses economic evidence to infer the existence of an agreement, parties can rely on evidence that the market structure was conducive to tacit collusion to escape a finding of collusion. As the Court has repeated frequently, Article 81 does not deprive economic operators of the right to adapt themselves intelligently to the existing and anticipated conduct of their competitors.

Article 82 applies to the abuse of dominance by one or more undertakings, but the Commission has made little use of the notion of collective dominance to date. One manifestation of collective dominance is the presence of a tight oligopoly. The definition of collective dominance deployed in the context of the ECMR can be transposed to Article 82. Than it would seem to follow that the anticompetitive effects resulting from an oligopoly could be labelled as exploitative abuses of collective dominance. This would close the oligopoly gap, but while this approach is attractive because of its apparent simplicity, it should be rejected because it conceals several difficulties. The first is how to characterise oligopoly behaviour as abusive. Court has struggled to define excessive prices and the regulation of such behaviour is relegated to newly liberalized markets.

The second difficulty relates to the remedy. If we believe that tacit collusion is the result of the market’s structure, neither a fine nor a behavioural remedy is appropriate. It follows that a structural remedy is appropriate, such as in Nestle/Perrier where the divestiture of assets eliminated the risk posed by collective dominance. Today the Commission has the power to impose a structural remedy, but only in very limited circumstances. Article 7 of Regulation 1/2003 provides that the Commission may impose. Any behavioural or structural remedies which are proportionate to the infringement committed and necessary to bring the infringement effectively to an end. Structural remedies can only be imposed either where there is no equally effective behavioural remedy or where any equally effective behavioural remedy would be more burdensome for the undertaking concerned than the structural remedy.

Three conditions must be satisfied to impose a structural remedy: First it must be proportionate, necessary and effective. The recitals provide some guidance on when a structural remedy will be proportionate. Changes to the structure of an undertaking as it existed before the infringement was committed would only be proportionate where there is a substantial risk of a lasting or repeated infringement that derives from the very structure of the undertaking.

Structural remedy is needed when there is no other effective behavioural remedy. The Regulation implies that the structural remedy must be more effective, or as effective but cheaper to implement, than a behavioural remedy. On the one hand, it might be argued that a remedy is effective if it stops the infringement. Satisfying this requirement may be problematic and the Commission must be sure that the divestiture eliminates the risk of tacit collusion. There is also a procedural risk in developing a sound theory of abuse of collective dominance. That the Commission uses Article 82 to catch express collusion. Commission must obtain convincing documentary evidence to prove an agreement or concerted practice. It could avoid this difficult exercise by identifying an oligopoly market and then finding parallel conduct to be an abuse of collective dominance.

There are two risks here. The first is that the Commission uses collective dominance to overcome the higher burden of proof in Article 81.The second is that, having done so, the Commission can impose a structural remedy on a market where coordination had been achieved through express collusion that went undetected, imposing unnecessarily onerous remedies.

Chapter 9: Distribution agreements

9.1 Case study

Compared to collusion or abuse of dominance, a distribution agreement, where neither manufacturer nor distributor holds a dominant position, seems to be conduct that should not worry competition authorities. While a cartel is an alliance where parties’ shared aim is damage to the competitive process by reducing output and increasing price, a distribution agreement shared aim is to increase output and reduce price to the benefit of consumers.

Delimitis v. Henninger case. The dispute concerned a contract between a brewery and a publican. In exchange for the brewer offering the publican a lease of a pub on favourable terms as well as equipment and furniture necessary to operate the pub, the publican undertook to buy a certain quantity of beer exclusively from the brewery. The Court recognised that this agreement did not have as its object the restriction of competition because of the benefits to brewer, publican and consumer. Contract guaranteed that the brewer had certain outlets that would sell its beer, allowing it to plan sales and to organise production and distribution effectively. The publican gained access to the beer distribution market under favourable conditions and with the guarantee of beer supplies, and their shared interest in promoting sales of the contract goods likewise secures for the reseller the benefit of the supplier’s assistance in guaranteeing product quality and customer service. The consumer gains from joined efforts of supplier and distributor.

Even vertical agreements can have anticompetitive effects. The Court went on to say that these would occur if all brewers were to adopt a similar distribution system, locking all pubs into exclusive purchasing agreements, which would make it impossible for new brewers to enter the market, for there would be no outlets for them to sell beer. Vertical restraints might foreclose market access just as effectively as a dominant undertakings strategies to raise rivals costs. Degree of economic harm will be less. In a market that is foreclosed but where several breweries are able to find outlets there is competition among the existing brands which drives costs and prices down, to the benefit of consumers. While greater access to more breweries would allow for greater variety of goods, there are diminishing returns to variety. Competition intervention is warranted, especially if there is a risk that significant numbers of brewers are foreclosed and the remaining sellers are able to engage in tacit collusion. If a competition authority seeks to guarantee economic freedom of market participants in addition to consumer welfare, foreclosure is an evil to be remedied no matter how irrelevant the extra supply is to consumer satisfaction.

The facts also suggest that vertical restraints by large breweries that own several pubs may endanger the continued existence of small, local breweries. From a policy perspective then, distribution agreements may warrant regulation to safeguard certain Community interests, for instance the protection of small and medium sized undertakings or the protection of regional economies. Given the Commission’s interest in protecting consumers, vertical restraints that reduce the diversity of goods available on the market may raise competition law concerns.

There are various types of distribution agreement. The manufacturer may deploy an exclusive dealing contract, which forbids the retailer to sell competing goods, or he may select only distributors that agree to provide certain additional services to consumers. Alternatively a manufacturer may restrict the way the distributor does business by requiring the distributor to sell the goods only from one retail outlet, or by fixing resale prices. A distribution contract can restrict both whom the manufacturer deals with and what the distributor can do. Manufacturer may distribute via franchising contracts, by which an aspiring retailer is supplied with know how on the marketing of the manufacturers goods and sells only the manufacturer’s products using the instructions specified by the manufacturer. This contract guarantees uniformity at all points of sale so that the consumer can expect the same quality at all retail outlets. The reason for imposing these restraints is to give the distributor incentives to market the goods more aggressively. If the distributor sells only the manufacturers products, he will devote all energies to that task. If prices are fixed, the retailer must ensure the goods are sold by advertising or providing non-price benefits to consumers.

Vertical restraints are neither always harmful nor always beneficial, and that a determination of whether a vertical restraint reduces economic welfare requires an appraisal of all relevant facts. Beyond this broad consensus lie different shades of opinion about how to regulate vertical restraints, because the borderline between restraints that enhance and those that restrict competition is uncertain. In what follows we explore the economic arguments from the perspective of two groups that are affected by vertical restraints: third parties wishing to enter the markets of either the manufacturer or the distributor, and consumers.

Distribution agreements are necessary for manufacturers to reach consumers. the distributor has a comparative advantage over the marketing of the goods, and the manufacturer may be unable or unwilling to integrate vertically and acquire a distribution network, in which case distribution contracts are necessary to bring goods to the market. Vertical restraints can enhance efficiencies in distribution. Manufacturer may prefer to sell large quantities to a few dealers rather than small quantities to many dealers as a way of reducing transportation costs. For new entrants at the manufacturing stage, the freedom to enter into distribution agreements is the means to market access. On the other hand, vertical restraints can also be used to make entry of new competitors more difficult. This can occur intentionally. Leading manufacturer wishing to preclude entry by a new and more efficient competitor can enter into exclusive agreements with major distributors and prevent effective entry by others.

Vertical restraints can reduce the price of goods. It is stated that when both manufacturer and supplier have market power there is a risk of double marginalisation. Manufacturer sets a price above marginal cost and the distributor sets its price above marginal cost so two parties raise the price. Vertical integration can eliminate double

marginalisation but so can vertical restraints limiting the distributor’s ability to raise price. Other vertical restraints might induce higher prices. If a manufacturer uses a system of selective distribution to sell only to retail outlets that meet certain criteria, he can reduce competition among retailers selling the same product and where inter-brand competition is weak then distributors are able to raise prices.

It is also possible that allocating exclusive territories to distributors can become disadvantageous for consumers when the market is oligopolistic and distributors have no incentives to pass on price cuts that result from a reduction in the wholesale price of goods if these would be matched by dealers of competing products. This eliminates incentives for manufacturers to reduce prices, to the detriment of consumers. The impact of a vertical restraint on price depends upon the nature of the restraint, the market structure and arguably the kinds of products involved. In relation to the products involved, studies suggest that the removal of vertical restraints in markets like toys and jeans reduced prices and increased output. This is better explained by the fact that when consumer demand is not affected by pre sale promotion or services, then there is no need to create a distribution network to generate incentives for promotion, and that restrictions on intra-brand competition are more likely the result of powerful retailers wishing to increase profits.

It has been sad that vertical restraints are a means to support tacit collusion. Selective distribution or franchising agreements are a means by which intra brand competition is reduced, and if all manufacturers use this mechanism, this can be a way of softening price competition. It has been suggested that vertical restraints are the means to police an express agreement But plausible these arguments are, there is little empirical support for RPM being used to strengthen upstream cartels.

There are circumstances where consumers may receive more information and lower priced goods when vertical restraints are deployed, but there may also be circumstances where consumers receive unnecessary amounts of information and find that goods have higher prices because of vertical restraints.

Economic literature suggests that manufacturers may use vertical restraints as strategic devices to harm competitors. Example is when a manufacturer knows that because of economies of scale, competitors must sell a given number of units to make a profit. The manufacturer may enter into several exclusive dealing agreements so as to prevent the competitor from finding enough purchasers to sell his good profitably, thereby excluding him from the market.

In other way to the American application of Chicago economics, Communities early response to vertical restraints failed to take into consideration both of the policy implications identified above. Rather than considering the effects of vertical agreements, the Commission distinguished pro and anticompetitive agreements by their form, and rather than facilitating flexibility, the Commissions policy led to an approach whereby distributors operated under a regulatory scheme that was described as a straitjacket. Brief review of the original approach is helpful to explain and appreciate the magnitude of the reform carried out in 1999. ECJ had in some cases suggested that an analysis of whether agreements infringed competition had to be carried out by considering the economic context, but this case law had two limitations: 1) in the same year that the Court advocated a flexible, economics oriented approach to vertical restraints, it also held that when a vertical restraint threatened to disintegrate the common market, this would be deemed anticompetitive. Court sent mixed messages. On the one hand applying what came close to an economic cost benefit analysis of vertical restraints, while on the other applying a per se prohibition towards restraints segmenting the market. The core value of market integration took precedence over considerations of economic efficiency; 2) limitation of the Courts case law also was that it had little impact on the Commission, who refused to follow the Court.

Instead the Commission relied upon findings that vertical restraints were restrictions of competition because they restricted the distributors’ economic freedom. This meant that an exemption was necessary for the vast majority of vertical restraints. But under Regulation 17/62 the sole way of obtaining exemption was to notify the agreement to the Commission. Commissions resources meant that it was unable to cope with the number of notifications, so parties suffered from an administratively inefficient system. The response was that the Community drafted Block Exemption regulations for the more common types of vertical restraints. These provided a list of contractual restrictions which would benefit from exemption (a white list) and a list of restrictions that would not qualify for exemption (a black list). undertakings able to draft agreements to fall within the four corners of a Block Exemption benefited from automatic exemption.

While the Block Exemption resolved the administrative inefficiency for some, it created a new problem. The black lists were very extensive and they reduced the ability of parties to draft vertical restraints according to their commercial needs. For undertakings, the choice was between redrafting their contracts to try and fit within the Procrustean bed of a block exemption, or notifying the agreement to the Commission in the hope of obtaining an exemption, which would normally result in the Commission issuing a comfort letter that did not provide sufficient legal security. From the perspective of consumer welfare, the Commissions approach led to Type 1 errors for two reasons: 1) Block Exemptions were so prescriptive that they probably prohibited some efficient agreements; 2) uncertainty of the notification system discouraged undertakings from experimenting with more efficient forms of distribution. It also caused Type 2 errors because the Block Exemptions were form-based and not premised on economic analysis and could protect inefficient agreements.

Rather than considering the effects of vertical agreements, the Commission distinguished pro- and anticompetitive agreements by their form, and rather than facilitating flexibility, the Commissions policy stifled freedom of contract for no good reason. In 1996 the Commission signalled a change to this policy. It was with vertical restraints that the Community kicked off the modernisation of competition law by abandoning the emphasis on economic freedom in favour of an emphasis on economic efficiency. In what follows we trace how successfully the new law implements this policy direction.

The way this policy has been implemented is curious from a legal perspective. It might have been suggested that vertical agreements where undertakings lack market power do not restrict competition and that therefore many vertical agreements escape the application of Article 81(1). This point would have conferred the maximum amount of freedom to undertakings to plan their business strategies without restrictions. Instead, the Community drafted a Block Exemption Regulation which exempts from the application of Article 81 those agreements where the market share of the manufacturer does not exceed 30 per cent and where the contract does not contain certain black-listed clauses. This approach sidesteps the Courts ambiguous case law on the interpretation of Article 81(1), buries the Commissions overly aggressive past policy and offers a method that enhances planning opportunities and targets the use of vertical restraints only when there is market power, and hence less inter-brand competition. Theoretically it seems wrong and cumbersome because the Regulation exempts and a fortiori assumes that vertical restraints restrict competition when instead economic theory suggests there is no such harm if the undertaking lacks market power.

The Block Exemption applies to vertical agreements where the market share of the manufacturer does not exceed 30 per cent. Above this market share, the manufacturer is taken to have sufficient market power to damage the interests of consumers and competitors. In cases of exclusive purchasing agreements, the market share of the distributor is taken into account because the concern is about how much of the distribution market is foreclosed by the agreement. The market share level was a compromise – prior discussions mentioned thresholds of 20 to 40 per cent.

Contracts where the market share does not exceed 30 per cent can be designed as the parties wish, provided the clauses prohibited in Articles 4 and 5 of the Block Exemption are not included. Article 4 provides a ‘black list’ of clauses whose presence deprives the agreement of the benefit of the Block Exemption. The black list mostly prohibits various types of market segmentation. It also prohibits resale price maintenance (RPM) and clauses that restrict the sale of spare parts to repairers or other service providers. The exclusion of RPM can be disputed on the basis that, according to economists, it is just one method of creating incentives for the distributor to market goods aggressively. Another black listed restriction is that the manufacturer cannot be prevented from selling spare parts to independent repairers or end users. This reflects the Community's interest in preventing the exercise of power in after markets and in promoting the presence of independent repair outlets.

Article 5 of the Block Exemption provides that three types of non-compete clauses do not benefit from the Block Exemption although the remainder of the agreement may do so if it can be severed from the non-exempted clauses. The first are non-compete obligations lasting for more than five years, which require a distributor to buy more than 80 per cent of his requirements from the manufacturer. This is designed to prevent the market being foreclosed to new sellers. Two sectors are implicitly excluded: beer and petrol, where it is common for the manufacturer to lease premises to the distributor. Non compete obligations for the duration of the lease are exempted. The second are post contract non compete obligations, which are not exempted unless this is necessary to protect know-how that has been transferred to the distributor. In this case, which is likely to arise in franchise agreements, a one year non-compete obligation is exempted. Selective distribution agreements that prevent members from selling the brand of particular competitors do not benefit from the Block Exemption. The Guidelines on Vertical Restraints suggest that this provision is designed to prevent a group boycott by several manufacturers of the goods of a competitor.

To over win the risk that the block exemption is too permissive and results in Type 2 errors, the Commission or a National Competition Authority may withdraw the benefit of the block exemption in an individual contract where the effects of the agreement are incompatible with Article 81(3). Commission is also empowered to withdraw, by regulation, the benefit of the Block Exemption from an entire market where a series of unrelated vertical agreements cover more than 50 per cent of the relevant market and all contain specific restraints. The primary preoccupation that underpins the powers of withdrawal is that networks of similarly worded vertical agreements by many

different undertakings can foreclose the markets, both for suppliers and for distributors, because of their cumulative effects.

There has only been one withdrawal decision to date. Langanese - Iglo where withdrawal occurred for a set of exclusive purchasing obligations because the leading ice cream manufacturer’s contracts foreclosed market access in a duopoly market, preventing the entry of a third large manufacturer.

The Commission claims that even after five decades of the EU existence the single market is still not a reality. One oft cited piece of evidence for this proposition is that price differences among MS remain high. The desire to integrate the market has led to the Community experimenting with various forms of regulation addressed to MS. The aim of market integration has resulted in strict regulation of business practices that contribute to market segmentation. In the context of vertical agreement, this strict approach is in direct tension with the free rider argument, which suggests that market segmentation can be a useful technique to guarantee that distributors act to maximise sales. One may go so far as to say that the Commission’s policy is counterproductive to market integration. Territorial restraint designed to avoid the free-rider effect was banned with the consequence that the manufacturer integrated vertically, an option, which is anti competitive because it forecloses the market for other manufacturers as one distribution channel is now owned by a competing manufacturer. The Commissions new approach resiles from the per se condemnation of territorial segmentation and offers a more flexible framework which takes into account the free-rider arguments even though the manner by which this is achieved is somewhat untidy.

First, Guidelines on Vertical Restraints indicate that when the manufacturer wishes to penetrate a new market or introduce a new product, the distributor may be afforded absolute territorial protection for two years without this infringing Article 81(1). This means that the distributor may be prohibited from selling the contract goods outside its territory and other distributors can be prohibited from selling in that territory.

Second, less absolute forms of territorial protection benefit from an exemption, but only under strictly specified conditions. In an exclusive distribution contract, or a franchise agreement where the manufacturer has granted a distributor an exclusive territory, the manufacturer may prevent other distributors from making active sales into the territory reserved to another distributor. In order to qualify for exemption all the manufacturers distribution contracts must be assessed, so as to discover which territory or territories are allocated to an exclusive distributor and then to check that all other distributors are unable to make active sales into those territories. To benefit from this exemption, other distributors must be able to make passive sales into the protected territories. Active sales occur when a distributor approaches customers in the protected territory or establishes outlets in that territory, while passive sales are those made in response to requests from consumers in the protected territory to import the goods from a distributor outside the territory.

Third, the Commission tolerates agreements that divide the market when they are necessary to safeguard the integrity of the distribution system in question. First, the contract may prevent wholesalers from selling directly to end users. Second, in a selective distribution contract active and passive sales by distributors to unauthorised outlets may be prohibited.

The importance of market integration wanes as economic analysis shapes competition policy.

For undertakings whose market shares are above the safety zone of the Block Exemption, or which contain black listed clauses, the mechanics of the application of Article 81 are of crucial importance, in particular given that notification to the Commission is now impossible and parties need to be able to predict whether their planned distribution agreement is antitrust compliant.

Commission has published Guidelines on Vertical Restraints which set out the Commissions analytical framework for assessing the economic effects of distribution agreements. The Guidelines pay little attention to distinguishing finely between agreements which are lawful because they do not infringe Article 81(1) and those which are lawful because they benefit from Article 81(3), as the effect on the parties is the same, albeit the burden of proof is upon complainants to show an infringement of Article 81(1) and upon defendants to show that Article 81(3) applies.

Commission fears four types of negative effects from distribution agreements: 1) foreclosure of suppliers or distributors; 2) reduction in inter brand competition; 3) reduction in intra-brand competition; 4) the creation of obstacles to market integration. The first three effects are consistent with the economic analysis canvassed above. Focusing on economic effects, the starting point for determining whether these materialise is an evaluation of the market power of the undertaking as well as the position of its competitors, the market position of the distributor, and entry barriers. The result of this exercise is that if there is sufficient inter-brand competition then the undertaking in question lacks market power and no competition problems arise, while if the agreement restricts inter-brand competition this facilitates foreclosure and price increases. The significance of the Guidelines is that the Commission is increasingly willing to find that distribution agreements may not infringe Article 81(1). In instances where the market is highly concentrated it also keeps a close eye on whether the effects of the agreement harm consumers or foreclose entry. The type of consumer harm depends on the type of distribution agreement. In the context of single branding, consumer harm may result either from the fact that if these agreements are used by all suppliers tacit collusion is facilitated, or from a loss of inter-brand competition. To test the risk of tacit collusion the analysis focuses on the concentration in the market. The loss of inter brand competition is measured by the level of trade and the degree of product differentiation. If single branding is imposed on retailers in a market selling branded goods with high degrees of product differentiation, consumer harm is likely in a highly concentrated market. Applying this methodology to the market for the sale of beer in public houses, the Commission in Bass held that a single branding agreement reduces inter-brand competition when the person operating the pub is unable to sell competing beer brands, and the high level of concentration resulting from the cumulative use of single branding meant that the restriction harmed consumers. In exclusive distribution contracts, where the manufacturer appoints one distributor for a specific geographical region, the main risk is a reduction in intra-brand competition. The methodology adopted here is to gauge the market power of the manufacturer and to measure the level of inter brand competition. The higher the level of inter-brand competition, the less important is the reduction in intra-brand competition, while weaker inter-brand competition means that consumers would benefit from intra-brand competition.

The style of analysis in the Guidelines reflects the Commissions new approach. Rather than merely identifying a restriction of economic freedom of consumers or competitors or other economic actors, the Commission seeks to measure the restriction whereby only appreciable restrictions of economic freedom are found to infringe Article 81(1). The weakness of this approach in the context of distribution agreements is that proof of such restriction is not always enough to show harm to consumer welfare. All agreements that yield anticompetitive effects may also gain exemption if the four criteria in Article 81(3) are met. Even if the approach under Article 81(1) remains more aggressive than is necessary, Article 81(3) can be used to measure whether the efficiency gains outweigh the restrictions of economic freedom. Even agreements that are anticompetitive by object may benefit from exemption. Very severe forms of territorial restriction can be characterised as pro competitive. There is no per se rule against resale price maintenance. Court has indicated that RPM may be justified in the distribution of newspapers and periodicals. RPM might be the only means to support the financial burden of having to take back unsold copies, if taking back newspapers is the only way in which a wide selection of newspapers are made available to readers. The Court does not guarantee an exemption but suggests that the pluralism of the media is a consideration, which is relevant in determining whether the agreement is exempt. In this market, the interests of the purchaser as citizen in addition to the interests of the purchaser as consumer are taken into consideration. Efficiencies are the basis for the application of Article 81(3). The burden is placed on the parties to prove these. Speculative evidence will not satisfy the Commission and efficiencies must benefit the consumer. Proof of efficiencies is fact-specific. In the context of exclusive distribution, one usual claim is that exclusivity is necessary to facilitate investment by the distributor to build the image of the product. The Commission indicates that this efficiency claim is more easily accepted in the case of a new product, or for a complex product where giving the distributors incentives to promote the goods by providing pre-sale services is important.

For some types of vertical restraints the Commission drafted Block Exemptions in the 1970s and 1980s, and in the absence of a Block Exemption for selective distribution, the Court had developed criteria to test its legality. Selective distribution is a system often used to exclude discount stores from a distribution network. The contracts often provide that the distributors must ensure staff hold certain qualifications or undergo training, and that the location and appearance of the retail outlet are suitable for the goods in question. In Metro the Court held that selective distribution systems constitute: “an aspect of competition which accords with Article 81(1), provided that resellers are chosen on the basis of objective criteria of a qualitative nature relating to the technical qualifications of the reseller and his staff and the suitability of his trading premises and that such conditions are laid down uniformly for all potential resellers and are not applied in a discriminatory fashion”. This statement encapsulates a rule so called simple selective distribution networks that satisfy the criteria set out above do not restrict competition, provided there are no anticompetitive effects, which only materialise if there is a cumulative effect when all manufacturers use similar distribution networks. The rule applies only to certain goods where selective distribution benefits the consumer, and has been applied to luxury goods because the aura of prestige surrounding the brand is an essential factor in competition among brands, to technically complex products because trained staff would facilitate consumer choice, and to newspapers and periodicals because the consumer expects each outlet to offer a representative selection of publications. The advantage of falling within this rule was that an exemption was unnecessary.

Metro doctrine is not limited by a market share threshold. It may be advantageous for parties whose market share does not allow them to benefit from the Block Exemption to fit their contract under the Metro standard, thereby avoiding the requirement of proving that the agreement satisfies the four conditions of Article 81(3). Application of the Metro rule today is at odds with the economic philosophy underpinning the Block Exemption because the rule is not limited by a market power test. The rule should be quietly abandoned. This is likely to happen in practice, as parties will shape their agreements according to the Block Exemption and the Guidelines rather than the case law. But some aspects of the Court’s jurisprudence may be worth retaining, for instance the criterion that selective distribution is only appropriate for certain types of goods. undertakings should not be allowed to quell intra-brand competition when consumers gain no added benefit from the distribution system.

Dominant undertakings are subject to both Articles 81 and 82. The Guidelines on Vertical Restraints take harsh line if the Commission were to apply Article 81. Where an undertaking is dominant or becoming dominant as a consequence of the vertical agreement, a vertical restraint that has appreciable anti-competitive effects can in principle not be exempted. Unless the vertical restraints do not infringe Article 81(1) an exemption will not be available, irrespective of whether the practice infringes Article 82. The justification for this position is that the last criterion under Article 81(3) is that the agreement must not eliminate competition and that dominance means that competition is eliminated. A fortiori, dominant undertakings cannot benefit from Article 81(3). This means that dominant undertakings have an incentive to integrate vertically to avoid the application of Article 81, leading to a situation where form rather than substance dictates business plans.

In more recent statements, Commission has softened its stance on the application of Article 81(3) to dominant undertakings. This is because the CFI corrected the Commissions opinion on the role of the final condition of Article 81(3). As noted above, the Commissions view is that dominant undertakings cannot benefit from an exemption even if their activities may bring economic benefits because their dominance eliminates competition. But the CFI has held that the phrase elimination of competition in Article 81(3) has an autonomous meaning which is unrelated to the concept of dominance. Accordingly, while Article 81(3) cannot apply where the dominant undertaking abuses its dominant position, an exemption may be granted when a dominant undertakings agreement falls under Article 81(1). In the Guidelines on Article 81(3) the Commission rescinds its position in the Vertical Restraints Guidelines so as to be consistent with this judgment. This means that dominant undertaking are now treated like not dominant undertakings so that the Commission will adopt the same economic rationale weighing up the positive and negative effects of agreements entered into by dominant undertaking, allowing exemptions even for vertical restraints by dominant undertaking.

The regulation of vertical restraints discussed above applies to all goods and services covered by EU competition law, unless another Block Exemption exists, and the one sector to which special distribution rules apply is the motor vehicle market. In contrast to the process of reform that took place in the context of distribution agreements for other goods, where the Commissions motivations were to facilitate business planning and to introduce an economically sound basis for regulating vertical restraints, the reform of the rules regulating car distribution contracts was characterized by a tension between the Commissions desire to liberalize car distribution and lobbying by interested parties to retain a system of distribution operating outside the rules of competition.

It should come as no surprise that the car sector has enjoyed a privileged position under EU competition law. After all the automobile industry has been one of the most important in the EU both for its contribution to the EU economic wealth and for the vast number of persons employed in the sector, and it has also benefited from a range of protectionist measures designed to curb imports of Japanese cars, notably a voluntary export restraint that expired in 1999.

The power of the car lobby was evident with the first Block Exemption in the motor vehicle sector of 1985. This allowed manufacturers to appoint exclusive dealers and to prevent them from selling competing cars. Exclusivity could be linked with a system of selective distribution to ensure that retail outlets conformed to the manufacturers requirements. Manufacturers could tie up car sales and repair services, so that a distributor wishing to sell cars would be obliged to offer repair services as well. This link was justified as being more efficient and in the interest of consumers; Manufacturers reputation depended on providing effective repair services. The 1985 Block Exemption formalized the practices that existed in this sector at that time. It was premised on the belief that cars were a special product, which necessitated a different distribution network with close cooperation between the manufacturer and a small number of dealers. The Commission tolerated methods of distribution that restricted inter brand competition, prevented intra-brand competition, and thereby damaged the internal market.

The regulatory thinking under the current Motor Vehicle Block Exemption Regulation (MVBER), which came into force on 1 October 2002, is very different. The premise is that dealers have little bargaining power vis a vis car manufacturers and that by increasing the independence of dealers as well as their number, the market will become more competitive. Commission observed that car owners should also benefit from competitive conditions when seeking to repair their car, and the old argument that sales and repair were carried out most efficiently by the same undertaking was abandoned. The MVBER is designed to inject more competition at both levels: car sales and car repairs and maintenance. As a result, the new regulatory structure is stricter than that under the Regulation on Vertical Restraints because the Commission sees its task as creating a competitive market rather than merely preventing market failures. It was accompanied by increased action against manufacturers who attempted to prevent parallel trade in cars by their dealers.

In order to facilitate competition among dealers, manufacturers wishing to benefit from the Block Exemption must choose between three types of distribution system: 1) exclusive distribution; 2) quantitative selective distribution; 3) qualitative selective distribution. This differs from the earlier Regulation that allowed a combination of exclusive and selective distribution which insulated dealers from competition. Dealers who opt for exclusive distribution must have a market share below 30 per cent to benefit from exemption, while manufacturers who opt for quantitative selective distribution must have a market share of less than 40 per cent, and no maximum market share applies if manufacturers opt for qualitative selective distribution, which is the option that allows for the widest number of dealers to gain access to the market. Intra-brand competition is most difficult when there is one exclusive dealer in one territory. Exemption is only available when the manufacturer has little market power and he loss in intra-brand competition is compensated by the presence of inter brand competition.

When manufacturers opt for selective distribution instead, the possibilities of intra-brand competition are greater because there are more dealers per territory and dealers are free to make sales in other territories. In particular, a dealer appointed via a selective distribution network is able to establish additional outlets in other locations in the EU, something unavailable to selective distribution networks under the general Block Exemption.

MVBER tries to make entry into the retail market easier for new kinds of distributors by preventing the manufacturer from forcing dealers to have their own repair facilities. Instead, the manufacturer must allow them to sub contract the provision for repair and maintenance to repairers that are authorized by it. The anticipated effect of the MVBER is to lower the cost of entering the retail market, making it possible for supermarkets to sell cars since they do not need to provide after-sales services and for cars to be sold via the Internet. MVBER facilitates the role of intermediaries who shop for cars around the EU on behalf of customers by forbidding the imposition of any obligation on dealers not to sell to such persons.

Major innovation of the MVBER is that it disaggregates the retail and repair businesses. Retailers cannot be obliged to offer repair services, and to complement this, supplier may appoint authorized repairers and must allow these businesses to limit their activities to providing repair and maintenance.

MVBER facilitates market access for independent repairers. These are repairers that are not appointed by the supplier and that carry out a significant proportion of repairs. Under MVBER, no exemption is granted where the supplier refuses to give independent repairers access to any technical information, diagnostic and other equipment tools or training required for the repair and maintenance of these motor vehicles. The supplier loses the benefit of the Block Exemption if it refuses to deal with third parties. This leads us to the second special feature. Obligation imposed is to deal with strangers. Under the general Block Exemption for vertical restraints, the supplier selects its distributors and the Block Exemption places limits upon what terms may be inserted in the contract. Here supplier has a duty to deal with any independent repairer and if he refuses to do so he loses the benefit of exemption for the entire distribution network. Manufacturer may be able to justify a refusal to deal with a wholly incompetent independent repairer; nevertheless the commercial freedom of the supplier is severely restricted.

The Commission devised an aggressive regulatory scheme for the car sector. Sector specific competition laws empower the Commission to design rules to inject more competition and to redesign markets.

Block Exemption also regulates the contracts between suppliers and dealers, so that no exemption is available unless: 1) the supplier who wishes to terminate gives notice in writing and gives detailed, objective and transparent reasons for wishing to terminate the contract; 2) the contract has a minimum duration of five years or is for an indefinite period, and notice of termination is two years; 3) contract provides for an arbitration procedure in case of disputes. It remains to be seen whether the Commissions reform is successful. In the Commissions favour is the fact that the law is cast without evidence of regulatory capture. Commission consulted all stakeholders and designed rules which it believed would benefit consumers. This is unlike the earlier Block Exemptions, which favoured car manufacturers. On the other hand, the Commissions highly prescriptive approach may have adverse effects as it increases manufacturers costs.

By giving greater power to dealers, it makes it difficult for manufacturers to terminate ineffective dealers and to generate incentives for efficient dealers. It is too soon to determine whether the MVBER is successful. Preliminary indications suggest a bleak picture. To date the vast majority of manufacturers have opted for selective distribution systems because they allow them to control the dealers sales outlets. When manufacturers sell via exclusive distribution systems they can control the dealers sales outlet but cannot control the quality of other retail outlets to which the dealer may sell the goods, while with selective distribution, dealers may sell only to other authorized dealers. The effect of this choice is that alternative forms of distribution cannot emerge when selective distribution is used. This is because the criteria for joining a selective distribution network can be onerous. This explains why some, including the UK's Competition Commission, have called for a ban on selective distribution to allow for the creation of innovative distribution channels. While there has been some reduction in price differences between MS, it is not clear how far this can be attributed to the new distribution regime, nor whether the still significant price differences will be affected as new distribution networks emerge.

Judging the effectiveness of competition policy towards vertical restraints requires us to determine what policy objectives we have in mind. Judged from the perspective of economic efficiency, the new-style Block Exemption Regulation reflects economic thinking in that it provides a market power screen to filter out those undertakings whose behaviour has no effect on the market. The analytical framework for agreements that require individual exemption is also by and large consistent with economic theory. But there still remain a number of systemic barriers that prevent the complete economic conversion of EU competition law in this field. The first is that the role of market integration remains, although in a less dogmatic form. An economic approach would suggest that if a undertaking has no market power, efforts to segment the market are unlikely to reduce economic welfare. Central political imperative of EU competition law continues to stifle the use of economics. The second barrier is that the Commission retains an interest in consumer welfare as opposed to economic welfare. Aspects of distributive justice inform the decision making process. Commission reserves the right to withdraw exemption if it considers that consumer interests are not well served even if the distribution network is economically efficient. The third barrier is that the Commission retains an interest in the other core value, economic freedom. Accordingly, the analysis of foreclosure is premised upon the elimination of suppliers or retailers rather than upon the adverse economic effects that may arise as a result of excluding certain market players; similarly, the right of access to spare parts for independent repairers is premised upon granting them market access rather than upon any risk to competition that may result from their absence. These three barriers to a full economic analysis suggest that the Commission retains an allegiance to political values.

Whether or not the Commission identifies new markets where sector-specific competition law is necessary, one may query whether EU wide regulation of distribution agreements is beneficial, because national markets retain local particularities.

Chapter 10: Enforcing of competition law in EU

10.1 Institutions which enforce competition law in the EU

The growth of economic analysis and expertise is analogous to that which occurred in the United States in the early 1960s, where increasing numbers of economists in the DOJ and FTC affected the direction of antitrust law, facilitating the success of the Chicago School views in the 1970s. But, the increased reliance on economic theories by DG Competition is unlikely to have the same radical effects that a similar process had in the United States. This is because the Commission, not DG Competition, has the last word in controversial cases, and it has not embraced the economics-oriented approach of DG Competition, while US antitrust agencies have greater policy and operational independence. While DG Competition is clearly committed to a more economics based approach, this has not led to the complete exclusion of public policy considerations in competition cases.

Potential impact of Regulation 1/2003, the so called Modernisation regulation, on competition law. This Regulation makes three significant hangs to the enforcement of Article 81. The first is that Article 81(3) is deemed to have direct effect, and so can be applied by national institutions namely competition authorities and courts). The second is to Europeanise Competition law by requiring that National Competition Authorities apply EU competition law when reviewing business activities that affect trade between Member States. This means that from 1 May 2004, when Regulation 1/2003 came into force, the Community moved from having one competition authority (DG Competition) to twenty-six (DG Competition plus all National Competition Authorities). The third change is that the system of ex ante notification and exemption is abolished (undertaking cannot notify agreements to the Commission or to National Competition Authorities to obtain an individual exemption). It means that parties bear the burden of determining on their own whether the conduct they are planning complies with EC competition law, and risk fines if their assessment of their measures competitive impact is wrong. Taken together, this means that enforcement of competition law changes in two ways: 1) the identity of the enforcer, EU Commission, or National Competition Authorities, or national courts; 2) the nature of enforcement, ex post enforcement by the competition authorities, and claims for damages by parties injured by anticompetitive behaviour.

Until 1 May 2004, competition law enforcement was based on Regulation 17/62. The main rule that served to centralise enforcement in the hands of the Commission was in Article 9(1), which provided that the Commission was the only body able to grant exemptions under Article 81(3). It meant that while national courts and NCAs could apply Article 81(1), they had no competence once the undertaking had notified the agreement to the Commission. And once the Commission had granted an exemption, one could not apply stricter national competition laws to prohibit the agreement. The effect of this was to incapacitate national courts and NCAs because they were unable to apply Article 81 in full. Moreover, the Commissions wide interpretation of Article 81(1) contributed to centralising enforcement in the Commissions hands.

Nothing in the Treaty required the institutional makeup established in 1962. Centralisation was a conscious decision by the Member States. Today the work of DG Competition might be taken for granted by many, but one must bear in mind that the powers which the Commission obtained under Regulation 17/62 were unique. The Commission can implement competition policy largely independently of the Council and the Member States, and impose financial penalties on undertaking for breach of the rules. This contrasts with the traditional Community method whereby the EU legislates and leaves implementation and enforcement to Member States and national courts.

The practical challenge arose as early as one year after the introduction of Regulation 17/62: by then the Commission had received notifications of over 35,000 agreements. It did not have the staff to address all these notifications in an efficient manner, and in many cases there were significant delays between notification and decision. As the years went on the number of notifications increased but DG Competition’s resources did not. This had two consequences. First, competition enforcement was inefficient. In the period 1994–7 the Commission managed to reach a formal decision in only 95 cases, while 1,755 cases were closed informally, so only approximately 5 per cent of cases received full treatment. Furthermore, at the time the White Paper on Modernisation was published, only nine notified agreements had been subsequently prohibited by the Commission between 1962 and 1999. This small figure suggests that most agreements that were notified were largely innocuous and the Commissions resources were wasted. Second, the Commission was unable to develop its enforcement priorities because it had to react to notifications. Again taking the 1994–7 period, the Commission received 1,022 notifications and 620 complaints about anticompetitive behaviour but commenced only 251 cases on its own initiative. The Commission adopted a range of mechanisms to counter these problems, but none were deemed to be completely satisfactory. We can discern three phases in the Commissions attempts to reduce its workload while attempting to ensure the uniform application of EU competition law.

In the first phase, from the mid-1960s to the 1980s, the Commission deployed three strategies to reduce the time spent on notifications. First, it developed a de minimis rule whereby agreements of minor importance were deemed not to infringe Article 81. This removed some agreements from its reach, which also facilitated the Commission’s policy of favouring cooperation among small and medium sized undertakings. Second, it drafted Block Exemption Regulations (the first regulation was in 1967). These identified certain types of agreement and detailed which contract clauses were contrary to EU competition law and which were lawful. Parties whose agreements fell within the four corners of the Block Exemption were granted automatic exemption. The early Block Exemptions were highly prescriptive, so that undertakings wishing to benefit from these would have to rewrite their contract to ensure that it complied. Their commercial interests were compromised by the need for legal security. Third, it developed procedures for settling notifications informally. These took the form of comfort letters issued to undertakings that had notified their agreements. A comfort letter was designed to provide the undertakings with reassurance that their agreement did not infringe EU competition law or that it would probably benefit from an exemption. Though, this practice was criticised for offering undertakings little comfort. The letter did not bind national courts or competition authorities so the undertaking still faced the risk of its agreement being challenged under national competition law.

Undertakings faced a stark choice, modify their agreement so as to fit within a highly prescriptive Block Exemption, or notify to the Commission and face uncertainty either because of delays should the Commission decide to issue a decision granting exemption under Article 81, or because the response took the form of a comfort letter. It is little wonder that some advised undertakings not to notify and to hope that the Commission would not challenge the agreement.

These measures failed in two respects: 1) they did not reduce the Commissions workload; 2) they did not provide a workable system for undertakings. In the early 1990s the Commission attempted a new route to reduce its workload, trying to deflect complainants from contacting DG Competition by galvanising enforcement at national level by involving NCAs and national courts. It obtained support from the Court of First Instance, which ruled that the Commission did not have an obligation to investigate all complaints that it received, but could set its own enforcement agenda by taking up cases that had Community interest. The Commission thus indicated that it would focus its enforcement principally on cases that raised a new point of law and cases involving Article 86(1), while NCAs should consider cases where the effects are felt within their territories and those unlikely to qualify for exemption under Article 81(3). Though these moves were unsuccessful: complainants were reluctant to seek remedies in the national courts, and NCAs were not as active as the Commission desired. According to the German Federal Cartel Office and the Federal Ministry of Economics, the following reasons explain why. First, the NCA could not apply Article 81 to controversial agreements which might require appraisal under Article 81(3) because only the Commission could at that time grant exemptions. Second, in 1993, only a few Member States empowered the NCA to apply EC competition law, so decentralised application could not occur. And even in Germany, where the Federal Cartel Office had the power to apply Articles 81 and 82, the NCA preferred to apply German competition law. The third and final attempt to reduce workload occurred in the late 1990s and, in contrast to the two previous phases, the Commission engineered a substantive rather than a procedural change in policy: it reconsidered its system of Block Exemptions. As we noted above, the Block Exemptions that had been drafted so far were criticised for creating a straitjacket effect. Parties had to make significant modifications to their contracts to fit within the scope of a Block Exemption. Commission embraced a radically different approach with the Block Exemption on vertical restraints in 1999. First, the Block Exemption has a market power screen whereby its application is restricted to undertakings below a given threshold. Second, the Block Exemption is significantly more permissive in that it contains only a brief list of agreements that are forbidden and gives the parties considerable latitude in designing agreements according to their commercial necessities. A similar approach has been applied to all other Block Exemptions.

It had been suggested that this approach was likely to reduce the Commissions burden considerably as more undertakings would take advantage of the new Block Exemptions and so the number of notifications would fall. Though, it was impossible to judge the significance of this final effort on the Commissions workload because the Commission was eager to implement a more radical reform in the shape of Regulation 1/2003, which we consider more fully below. The number of new cases between 1999 (the year of the first newstyle Block Exemption) and 2004 (the final year when notifications were possible) shows a significant downward trend in the number of notifications when compared to the period 1989–98.

In the latter period, the Commission received over 200 notifications a year, peaking at 368 notifications in 1995. In 1999, the number of notifications fell to 162, and in the first years of the new century, notifications fell significantly: 101 (in 2000); 94 (in 2001); 101 (in 2002); 71 (in 2003); and 21 (in 2004). Commission had been working hard at reducing the backlog of cases: over 3,000 notifications were pending in 1980, but this figure had fallen to 1,204 in 1998 and 473 in 2004.

The Commission never clarified whether its limited resources would have remained insufficient even with this significant reduction in notifications that seems to have been caused, in part, by the new-style Block Exemptions. The need to reform Regulation 17/62 resulted not only from what the Commission diagnosed as the inadequacy of the system of notification in an enlarging European Union. There was also a political challenge that arose in the mid-1990s soon after the Commission gained powers to regulate mergers. Certain Member States, in particular Germany, expressed concern about the infusion of politics in competition decisions, and the lack of transparency in the Commissions decisions. German commentators began to demand a radical institutional reform: the creation of a European Cartel Office, operating independently of the EC Commission and able to deliver decisions based exclusively on legal principles. While the proposal for a European Cartel Office was never likely to be implemented, in particular because few Member States backed the project and because of the legal difficulties in creating independent regulatory agencies at Community level, Regulation 1/2003 can be read as a response to these criticisms. By surrendering enforcement to NCAs, the Commission was sending a signal that the political meddling by the Commissioners would wane.

In 1962, only Germany had a credible system of competition law. Though, this picture changed radically from the mid-1980s. At the same time that the Commission was attempting to decentralise enforcement, significant moves were afoot within the Member States: a number of them amended national laws, aligning them to the EU provisions. By 1999 all Member States except Germany had adopted national competition laws that were similar to Articles 81 and 82 and eight out of fifteen Member States (Belgium, France, Germany, Greece, Italy, the Netherlands, Portugal and Spain) conferred express powers

on National Competition Authorities to apply Articles 81 and 82. They were not compelled to take either of these measures by the Community and their reasons for reform are varied. Some Member States (e.g. Italy and Ireland) had no national competition laws; some (Spain, Greece and Sweden) adopted such laws in anticipation of joining the EU. Others had an unsatisfactory competition policy. Among this last camp was the United Kingdom, where reform of the rules had been raised several times but the law was changed only in 1998. The old rules were perceived to be too weak, and the role of ministers in competition decisions too prominent. While existing Member States Europeanised national competition laws without any obligations stemming from Community law, the countries seeking to gain access to the EU were required to put into place a system to enforce competition law and used the EU model to achieve this. The effect of all this legislative activity was that the norms of EU competition law were spreading into the national laws of the Member States even before Regulation 1/2003 was being discussed.

In several national laws, interpretive provisions were inserted to guarantee a high degree of uniformity in the application of the law.

Another significant development in the Member States is that NCAs grew in prestige. It has been suggested that the creation of independent National Competition Authorities was to a large extent a symbolic exercise, demonstrating commitment to free market values by the state, with the expectation that the agencies would not be very active. But governments expectations were confounded. Several national competition agencies have become powerful and highly regarded enforcement institutions. This is because the agencies were given enough political independence to be insulated from national politics and they developed technocratic expertise in law and economics, thereby narrowing the criteria they used to enforce the laws, further excluding political considerations. This development is significant because it served to embed the culture of competition in the Member States, and it made the Member States support an enhanced profile for NCAs. If we take these national developments together, the implementation of Regulation 1/2003 by the Council of Ministers becomes both possible and palatable. It is possible because the Community is able to trust NCAs. They have developed independently of government and are highly professionalised. They can be trusted to apply competition law in a non-political manner. Furthermore, the spontaneous convergence of national laws minimised the risks of the application of stricter national competition law, so creating a level playing field of decentralised competition law enforcement was feasible. Implementation of Regulation 1/2003 is palatable, to Member States, because the NCAs had already been applying rules with a view to ensuring that the application of national law was comparable to that of EU law, so the Regulation would not be seen as revolutionary by the NCAs or by the electorate. Regulation 1/2003 complements and strengthens the pre-existing Europeanisation of competition law. Some have seen these national developments more cynically, however, and argued that business support for the reforms was a tactical ploy designed to remove from the statute books strict national competition laws, and governmental acquiescence to business demands was a means to rein in the power and activism of NCA.

Declaring the direct effect of Article 81(3) was seen as necessary to galvanise NCAs, as this was the major stumbling block to decentralised enforcement. Their involvement would allow the Commission to increase its ability to take on cases of Community interest and become more proactive. The result is to multiply the number of agencies able to enforce EC competition law, leading to more rigorous enforcement. This reform by itself was insufficient because of the risk that NCAs would reach divergent results by applying this provision in different ways, so that the Commission has had to intervene to narrow down the interpretation of Article 81(3). But it remains to be seen whether all NCAs will apply the law in the same way or if divergences make competition law enforcement less predictable for undertakings, and thus less efficient. As NCAs were already applying national competition law moulded upon the EC norms, it is not clear why empowering them to apply EC competition law enhances the effectiveness of competition law enforcement. Abolishing the notification procedure was seen as essential for the Commission to redeploy its resources and develop a proactive enforcement policy. This argument seems overstated, for several reasons. First, the backlog of notifications which the Commission had received was falling in the years leading up to Regulation 1/2003, and more efficient management of the backlog could have eliminated the Commissions heavy workload. Second, the claim was not consistent. Furthermore Commissions work priorities could have been streamlined automatically with the coming into force of the new Block Exemptions.

Argument that the abolition of notification was necessary because of the Commission’s limited resources was probably not intended to be taken seriously, referring to a comment by a Commission official that the proposed modernisation would go ahead even if the personnel in DG Competition was doubled. Therefore it is wrong to say that the lack of direct effect of Article 81(3) and the notification procedure were jointly responsible for an ineffective and reactive competition policy. It was the Commissions inefficient management of notifications, combined with its unreasonably wide conceptualisation of what restricts competition under Article 81(1), that led to the systems ineffectiveness. This means that the reason why Regulation 1/2003 was implemented had little to do with abandoning a system that could not work, but rather the Commission was refusing to make the current system work well, and it wished to opt for a solution that brought EU competition law in line with a US style enforcement policy of ex post application of competition law coupled with deterrence elements. A substantive policy change is inherent in the procedural reform.

The third major board of Regulation 1/2003, the application of EC competition law at national level and the exclusion of divergent national competition rules, can be said to be crucial to ensure coherent enforcement across the Member States. To a certain extent, one might query whether Regulation 1/2003 needed to make express provision for this because the vast majority of Member States had already aligned their competition laws with those of the Community, so substantive divergence resulting from the application of national law might have been minimal. Moreover a degree of substantive divergence might well be beneficial. But, when the Commission originally proposed that EU competition law should apply exclusively, the larger Member States that had retained certain distinctive features in their national laws vetoed this, so a compromise was necessary. The first two paragraphs of Article 3 provide as follows: 1) where the competition authorities of the Member States or national courts apply national competition law to agreements, decisions by associations of undertakings or concerted practices within the meaning of Article 81(1) of the Treaty which may affect trade between Member States within the meaning of that provision, they shall also apply Article 81 of the Treaty to such agreements, decisions or concerted practices. Where the competition authorities of the Member States or national courts apply national competition law to any abuse prohibited by Article 82 of the Treaty, they shall also apply Article 82 of the Treaty; 2) the application of national competition law may not lead to the prohibition of agreements, decisions by associations of undertakings or concerted practices which may affect trade between Member States but which do not restrict competition within the meaning of Article 81(1) of the Treaty, or which fulfil the conditions of Article 81(3) of the Treaty or which are covered by a Regulation for the application of Article 81(3) of the Treaty. Member States shall not under this Regulation be precluded from adopting and applying on their territory stricter national laws which prohibit or sanction unilateral conduct engaged in by undertakings.

Article 3(1) contains an obligation to apply Articles 81 and 82 in parallel with national competition law. The first sentence of Article 3(2) is designed to ensure the supremacy of EU competition law in cases of parallel proceedings – thus stricter national law cannot be applied. So if an agreement does not infringe Article 81, stricter national competition law cannot apply to enjoin it. However, this was not enough to satisfy all Member States, and the French government in particular insisted on the second sentence of Article 3(2). This is because French competition law has two special rules that are stricter than Article 82.

Abuse of economic dependence, and the other is a rule that prohibits the sale of consumer goods at a price that is significantly below cost even when the undertaking has no dominance. As national competition laws are concerned then, Article 3(2) limits the possibility of divergence in so far as Article 81 is concerned, but tolerates stricter competition laws that apply to unilateral conduct.

The third paragraph of Article 3 goes further by allowing national laws that prohibit acts that constitute unfair trading practices whether they are unilateral or not: 3) without prejudice to general principles and other provisions of Community law, paragraphs 1 and 2 do not apply when the competition authorities and the courts of the Member States apply national merger control laws nor do they preclude the application of provisions of national law that predominantly pursue an objective different from that pursued by Articles 81 and 82 of the Treaty.

The difficulty in applying Article 3(3) is to determine which rules of national law are not to be considered EU competition law rules. The examples used here are borderline: economic duress could be compared to the abuse of a dominant position by a situational monopoly, so perhaps similar policies animate that doctrine. Borderline between what is competition law and what is not might be the subject of greater controversies in the future, especially as the current vogue is to see EC competition law as designed to promote consumer welfare.

It is debatable whether Regulation 1/2003 was necessary to achieve efficient enforcement and that it will actually lead to more effective enforcement. In the White Paper on Modernisation the Commission presented four other options for reform but none were given any serious consideration.

10.2 New structure of enforcement in the EU

The major implication of modernisation is that the Commission has freed itself from the burden of reviewing harmless agreements and is capable of setting its priorities. It intends to focus upon serious infringements (cartels) and enforce state aid rules with more rigour. This could be a significant change. In the period 1989–96 the Commission had begun own initiative enforcement action in only 13 per cent of the cases; the rest of its activity was reactive. The Commissions new policy priority is complemented by greater enforcement powers. First, Regulation 1/ 2003 empowers the Commission to carry out unannounced inspections in private homes as well as company headquarters; it may seal premises and offices to ensure evidence is not destroyed, ask for oral explanations and even, if the parties consent, carry out interviews. It has been suggested that the increase in investigatory powers that the Council granted to the Commission is a tacit endorsement of the Commission’s commitment to prioritise cartel enforcement. Second, the Commission has been increasing the level of fines set for cartel infringements, a policy which has been backed by the ECJ. Third it has imitated the United States in offering leniency to undertakings that confess to being party to an anticompetitive agreement. While the enforcement powers and the penalties are not as significant as those provided for in the United States and in some EC Member States they provide a coherent shape to the Commission’s new enforcement policy.

In addition to acting as a cartel buster, the Commission also has three major additional tasks to perform. The first is to dictate the development and direction of EU competition law. This is accomplished by the publication and renewal of soft law instruments and Block Exemptions.

The second task is to assist undertakings that are planning agreements but are uncertain about the competition law implications, and the third is to monitor the performance of NCAs. One major gap in the new system is that parties are unable to notify agreements ex ante. While the notification/exemption system was not perfect it offered parties some legal security, which they now lack.

The Council and the Commission has responded to the risk of legal uncertainty in two ways: 1) by attempting to clarify the substantive meaning of Article 81; 2) by creating procedures that allow for a substitute to ex ante notification. The substantive clarification of Article 81 can be witnessed by the fact that the Commission used the final years of Regulation 17/62 to publish a vast number of individual exemptions in a range of markets so that parties and NCAs are aware of how Article 81(3) operates. In addition, it sought to restrict the scope of Article 81(3) so that public policy considerations are excluded from its ambit. These are designed to make the application of Article 81 more predictable and to aid business in a system without ex ante notifications. Nevertheless these two measures are unlikely to be of help when parties engage in practices not foreseen by the guidelines. Moreover, because competition cases are intimately fact specific, it has been argued that general guidelines and precedents are unlikely to provide sufficient legal security to those planning an agreement.

At a procedural level, Regulation 1/2003 provides three additional substitutes for the now defunct notification system. The first is in Article 9 under which undertakings are able to offer commitments to the Commission whereby they promise to modify their behaviour when the Commission intends to take action against them. This allows the parties to negotiate a solution with the Commission after the agreement has been implemented and investigated by the Commission. Thus, there is still scope for some form of consultation with the Commission. Article 9 route seems to be the functional equivalent of a notification/ exemption system. The second substitute for ex ante notification is Article 10: “Where the Community public interest relating to the application of Articles 81 and 82 of the Treaty so requires, the Commission, acting on its own initiative, may by decision find that Article 81 of the Treaty is not applicable to an agreement, a decision by an association of undertakings or a concerted practice, either because the conditions of Article 81(1) of the Treaty are not fulfilled, or because the conditions of Article 81(3) of the Treaty are satisfied. The Commission may likewise make such a finding with reference to Article 82 of the Treaty.”

The preamble suggests that the intention behind this provision is to clarify the law, in particular when the parties engage in practices for which there is no precedent. Thus, the ‘public interest’ is to promote legal certainty and to ensure coordinated enforcement. Having attempted to seal off the use of Article 81(3) as a tool for public policy, the Commission might reintroduce this risk with Article 10 of Regulation 1/2003. The third substitute is a suggestion in the Regulation’s preamble that the Commission is still able to offer informal guidance to parties where a case gives rise to genuine uncertainty. Such informal guidance is reminiscent of the comfort letters that the Commission would issue, and while the Commission has emphasised that this guidance would be provided only when the legal issues are novel and unresolved and of Community interest, the guidance, like comfort letters, does not bind national courts or competition authorities. This procedure allows for continued dialogue between industry and the regulator but it is a further recognition that a shift to an ex post enforcement policy needs to be balanced by an effective ex ante notification system.

It remains to be seen whether these methods of granting some form of guidance are going to be workable. They present three challenges. The first is whether the guidance is sufficient for parties. The second is the extent to which they can be used to negotiate or impose upon parties obligations that have nothing to do with the anticompetitive effects but are designed to open markets. First Article 9 settlement on football broadcasting rights raised questions as to the relevance of cultural and industrial policy. The third risk is whether the Commission’s workload might be affected so that these procedures remove resources from its central activity, fighting hard core cartels.

National Competition Authorities (NCA) are expected to take on more cases than the Commission, and the UK government suggests this places NCAs in the driving seat for much competition law enforcement. In particular they will address local competition law infringements where they have a comparative advantage because of their familiarity with the local markets and are better placed to regulate national markets than the Commission. The degree to which this division of labour will provide effective enforcement depends on three variables: whether the Commission has managed to save resources with Regulation 1/2003; whether enforcement among the twenty-six competition authorities can be coordinated effectively; and whether NCAs enforce competition law with equal determination. Some NCAs are less politically independent than others, others have fewer resources and less expertise and also as a result, some will have more prestige than others. The flipside is that enforcement may be more intensive and sophisticated in states with stronger and more well-resourced competition authorities and less so in states where competition authorities lack the resources or expertise to enforce competition law actively. In less than two years since the operation of the network began differences were already beginning to appear. Between 1 May 2004 and 30 June 2006, the three busiest competition authorities were the French (ninety-four cases), the German (sixty-four cases) and the Dutch (forty-four cases); while twelve Member States’ NCAs initiated fewer than ten cases. Diversity in the composition of NCAs is acknowledged under Regulation 1/2003 so long as the NCA is able to carry out the tasks under the Regulation.

Involvement of NCAs is subject to one further uncertainty even in the easy cases of a flagrant breach of competition law. In order for the application of EC competition law to be engaged, the practice must affect trade between Member States. It follows that the anticompetitive effects will occur in the Member State where the NCA is located and in other Member States as well. However, the penalties that the NCA can impose seem to be restricted to effects felt in its territory, and the NCA has powers to enforce the law only against undertakings located in its territory. If so, this would risk undermining the rigorous enforcement of EC competition law because the NCA would not be able to impose a fine reflecting the entire harm of the infringement. The alternative, that once the first NCA reaches a decision, the other NCAs where the agreement causes harmful effects will institute their own proceedings, may violate the undertakings rights, but moreover seems a highly inefficient use of resources. It remains to be seen whether the European Competition Network will elaborate solutions to this issue.

The European Competition Network (ECN) was established in 2002, when Regulation 1/2003 was agreed. The ECN is not an administrative body, but a forum where the Commission and NCAs meet to carry out two formal tasks: allocating cases among the NCAs (coordination of enforcement) and ensuring the coherent application of the rules (coordination of results). These two kinds of coordination are necessary because Regulation 1/2003 did not establish a system whereby the decision of one NCA binds others. To use the Commission’s jargon, there is a system of parallel competences. This means that, in theory, an agreement could be reviewed independently by more than one NCA and each could reach a different result. To avoid this outcome, which would frustrate the aims of the reform programme, the Commission expects that cases can be allocated via the ECN and has published a prescriptive notice to regulate case allocation, and there are safeguards to ensure rules are applied consistently.

Coordination of enforcement is provided for in the Notice on Cooperation within the Network of Competition Authorities. The basic principle is that each case should be taken up by either a single NCA, or several NCAs acting jointly, or the Commission. This leads to one decision per case and avoids inconsistent outcomes. The reason why a system of exclusive competences was not established is probably because Member States wanted to remain free to apply competition law independently of other NCAs. This is confirmed by the political declaration establishing the ECN, where, while Member States agree to cooperate with other NCAs and the Commission on the basis of equality, respect and solidarity, they also declare the independence of each NCA.

ECN will not operate to allocate cases, but to reallocate them, because a competition case will first be taken up by one NCA whose first duty is to notify the Commission and other NCAs that it has commenced an investigation. Only at that moment might a case be reallocated, and this can occur for two reasons: 1) the NCA itself seeks reallocation; 2) another NCA or the Commission might request that it address the case in question. In order to decide which NCA should act, it is determined which NCA is well placed on the basis of three criteria: 1) the effects of the infringement in question occur in its territory; 2) it is capable of issuing an appropriate remedy; 3) it is able to obtain the relevant information. Cooperation among NCAs continues once the case has been allocated in that information that NCAs have about the undertakings under investigation may be exchanged.

In contrast to the informal network structure put into place to ensure coordination of enforcement, the process for coordination of outcomes is hierarchical, because while there is little to ensure cooperation among the NCAs, the Commission controls the decision-making practice of each NCA: 1) an NCA may not reach a decision that is contrary to a Commission decision; 2) before adopting a decision, the NCA must send a draft to the Commission, and at this stage the Commission may make comments or take the drastic step of removing the case from the NCA and initiate proceedings itself.

The system affords the possibility of coordination to a much greater extent than the powers available to federal competition authorities in the US. This could be necessary given the relative inexperience of certain MS with competition law, although it does undermine the Commissions claim in the White Paper on Modernisation that there is a culture of competition in the EU. Commissions determination to control the results that NCAs reach stands in contrast to the assertion in the White Paper on Modernisation that there is abundant case law, clearly established basic principles and well-defined details.

More general reflection is warranted about the functioning of the ECN, and that is to consider what assumptions underlie networks and how far the Commission has designed a system with the potential to contribute to delivering effective enforcement. On one level, it has been argued that a well functioning network requires three conditions: mutual trust and cooperation; professionalism; and a common regulatory philosophy. Judged against these standards, the ECN is not perfect. While there is some degree of trust, the Commissions right to veto NCAs by taking a case away from them and the ability of one NCA to institute independent proceedings should it disagree with another NCA point to a lack of complete trust among the members of the ECN. There is a good degree of cooperation, but the Commission seems to retain its role as principal. All NCAs are increasingly professionalised, although Regulation 1/2003 does not require that an NCA should be independent of government control, which could weaken the role of the network.

Commission believes that there is a common regulatory philosophy after forty years of centralised competition law enforcement; however, as we said before, this statement is not consistent with its re-interpretation of Article 81(3). While the conditions for a successful network might not be perfect, it may be suggested that the presence of the ECN itself will make each NCA accountable to the others and eager to ensure the success of the network and the effective enforcement of competition law. Thus the network might strengthen itself as the members have an incentive to maintain their reputation in the eyes of their colleagues. Cooperation and a common regulatory philosophy can emerge through the working of the network. One flaw in the Commissions design for the ECN is its excessive zeal in holding NCAs to account, which may lead to too much homogeneity in the performance of NCAs. This criticism is based on the fact that the ECN gives the Commission hard law powers to control the NCAs when a less rigid scheme of accountability would be preferable, so as to allow a degree of regulatory diversity. It has been suggested, for example, that the ECN could function as a forum for comparing and evaluating the performance of the NCAs and that this would allow each NCA to have greater autonomy while creating a system where, incrementally, the methods of enforcement can converge by the dissemination of best practices. Moreover, diversity may be necessary as conditions of each market vary, requiring diverse regulatory efforts. In the latter case, it is worth remembering that one argument for Regulation 1/2003 is that NCAs have a better understanding of local markets.

Private enforcement: modernisation envisages an increased role for damages claims by parties suffering from anticompetitive conduct. While the European Courts proclaimed that Articles 81(1) and 82 have direct effect and granted actionable rights as early as 1974, to date there has been little recourse to the courts. A major study in 2004 suggested that there was total underdevelopment of damages actions for breaches of competition law, with approximately sixty cases since 1962.

This is a paltry record if compared with the United States where private actions outnumber public enforcement by a ratio of ten to one, and where some commentators suggest that there is under-enforcement in spite of these larger numbers.

There are two ways to explore the protective scope of competition law statutes. The first begins by suggesting that private litigation has a dual function: it protects the plaintiff and it deters further anticompetitive conduct. This is supported by the fact that the plaintiff must prove both that the defendants act restricts competition, and that he has suffered a personal loss. Competition law does not protect an individual, but the market. Private litigation then should be allowed when the plaintiff’s action helps to deter anticompetitive behaviour. The second way to justify a right to damages is to explore what classes of person EU competition law protects. On this basis, all consumers should be entitled to claim because EU competition law protects consumer welfare. This was made quite clear in the Courts explanation of the harm caused by a cartel: participation by an undertaking in anti-competitive practices and agreements constitutes an economic infringement designed to maximise its profits, generally by an intentional limitation of supply, an artificial division of the market and an artificial increase in prices. The effect of such agreements or of such practices is to restrict free competition and to prevent the attainment of the common market, in particular by hindering intra Community trade. Such harmful effects are passed directly on to consumers in terms of increased prices and reduced diversity of supply. Where an anti competitive practice or agreement is adopted in the cement sector, the entire construction and housing sector, and the real-estate market, suffer such effects.

Court argues that the individuals harmed by a cartel are all those who purchased cement, and all consumers further down the line that suffer as a result of the higher prices in the industry. It follows from this that consumers should have a right to damages because they are the direct beneficiaries of Article 81. Until recently, the Italian courts had refused to recognise the consumers right to secure damages, but in a path-breaking judgment Italy's highest court has now recognised that competition law safeguards consumer interests.

Damages for consumers can be justified on two alternative grounds: 1) they have a subjective right which is within the protective scope of Article 81; 2) their lawsuits deter unlawful agreements. The same two justifications can be invoked to establish that competitors have a right to seek damages.

Court of Justice has not examined what the protective scope of the competition laws is in the manner suggested above, and has said that other parties are also able to claim damages for infringements of Article 81. In a recent judgment the Court proclaimed: “any individual can claim compensation for the harm suffered where there is a causal relationship between the harm and an agreement of practice prohibited under Article 81 EC”. With this conclusion the Court appears to suggest that there is no need to ask who is protected by Article 81, because anyone whose loss is caused by the breach of Article 81 can claim damages.

It seems that giving parties to a contract a right to damages creates a range of special responsibilities on the potential defendant that are quite alien to the nature of Article 81 and closer to those we find in Article 82 but more extensive: the manufacturer has a duty to negotiate with care.

Competition law can be used as a strategy to harm competitors. This is the converse of the theory that the Court and Commission have embraced whereby the use of the legal process can be an abuse of dominance. Many parties to anticompetitive agreements have used the courts to secure a declaration that the contract is void as a way of escaping liability for breach of contract, a practice known as a Euro defence. If it is inappropriate to use competition law to secure an avoidance of contractual liability, it is even more inappropriate to allow a claim in damages.

Few judgments of the ECJ suggest that the right to damages for breaches of EU competition law has a private and a public dimension but that the two are indissoluble: the individual who has suffered loss is allowed to sue only because his claim safeguards the market by increasing the deterrent effect of the competition law.

It is not controversial that consumers should be entitled to secure damages, but it is more difficult to decide which consumers should have a right to damages.

Claimants can be divided into two groups: those whose claim is a follow on action after a competition authority has made an infringement decision and who thereby use the factual findings of the authority to help their claims, and stand alone claims by parties who identify a breach of competition law without a prior finding by a competition authority. The Commission is eager to encourage both, and has recently identified some of the major hurdles in a Green Paper.

Follow on claimants have a somewhat easier route to claim. In some jurisdictions (the UK and Germany) the national court is bound by the findings of a competition authority. In Germany, a national court hearing a follow on damages claim is bound by decisions of the EU Commission, the Bundeskartellamt and even of the competition authorities of other Member States. Follow-on lawsuits can make a significant dent in the profits of a undertaking embroiled in a cartel. The plaintiff in a stand alone action instead bears the burden of identifying the breach. In order to promote stand alone actions, someone would need to facilitate access to information held by the parties and by competition authorities.

The majority of damages actions against undertakings guilty of the more serious violations of competition law (cartels and abuses of dominance) are likely to be follow on lawsuits. That is, parties will wait for a competition authority to make a finding and then use this as the basis for a claim in damages. This means that private enforcement is not an alternative to public enforcement but merely a way of compensating those who suffer harm.

Another perspective from which to examine the relationship between private enforcement and modernisation is to consider how court proceedings interact with those of competition authorities. First, it seems that courts are not bound by commitment decisions that the Commission enters into under Article 9 of Regulation 1/2003. This recreates the same problem of uncertainty that existed with comfort letters. Second, courts are not bound by leniency schemes. So a party that settles with the Commission may still face private lawsuits. This problem is particularly poignant because it creates a risk that facilitating private litigation diminishes the incentive for parties to make leniency application. The Commission is investigating how to reconcile leniency programmes with damages claims.

Coordination between courts and the Commission is established to ensure consistent enforcement. Unlike the ECN, where the Commission possesses considerable powers to prevent inconsistent decisions, the independence of the courts prevents comparably aggressive checks on national courts. Regulation 1/2003 provides for three forms of cooperation.

First, the court may seek some assistance from the Commission; second, the court must transmit a copy of its judgment to the Commission; third, the Commission may act as amicus curiae to provide its opinion to the court. The last is the closest the Commission can get to influencing the national court, and there may be a risk of less confident courts following the Commission’s opinions. These forms of control suggest, that a more subtle network is in place among the national courts: on the one hand, their autonomy allows courts to explore different solutions to comparable problems, and on the other, courts will be referred to judgments of foreign courts and this will facilitate an exchange of ideas which is not as likely under the ECN with the Commissions more hands-on control to ensure uniformity.

Competition policy has been described as the EU first supranational policy because the Commission operates as an autonomous agency, free from interference from Member States, the Council or the European Parliament. Regulation 17/62 gave the Commission more powers than the Member States foresaw and it allowed the Commission to design a competition policy for the EU largely free from adverse judicial scrutiny, the ECJ backing most of the Commission’s interventions. After 1985 competition enforcement grew in volume and in diversity and the success of competition law led to calls for reform. The Commission wished for modernisation, ostensibly because of an overload of cases, but more probably in order to redirect its enforcement policy away from scrutinising notified agreements and towards regulating cartels. Certain Member States were concerned about the politicisation of decision making, and decentralised enforcement was seen as a means of resolving this criticism, by placing independent NCAs at the front line of competition enforcement. The procedural change comes together with a substantive change for competition law, at two levels: the priorities for enforcement have changed, and the substantive interpretation of the law is narrowed down, recourse to public policy considerations being replaced by an emphasis on effects on consumer welfare.

This substantive policy change is reinforced by the provisions of Regulation 1/2003 that strive to compel NCAs and national courts to apply Articles 81 and 82 in a harmonised manner and to the exclusion of national competition law. Whether or not enforcement is more efficient, control over the enforcement of competition law by the newly galvanised NCAs and courts is considerable. The network of NCAs seems to be a forum to facilitate the Commissions policy, by ensuring that there is only one authority in charge of any case and allowing the Commission the final word on any anticipated ruling of NCAs. While the institutional resettlement appears to decentralise enforcement, it merely decentralises the operational aspect of enforcement, leaving the policy aspect to the Commission. If we recall, looking back over the previous chapters on the substantive law, that the Commission is increasingly keen to view competition law as a means to achieve consumer welfare through competitive markets, then the effect of the kind of decentralisation we witness is to displace national economic policies in favour of a neoliberal, pro consumer economic policy favoured by the Commission. This might be challenged by the growth of national competition cultures that safeguard a wider range of interests, and by private litigation which, in the aftermath of Courage v. Crehan, supports the launching of lawsuits that undermine the pro consumer bias of modern EU competition law.

Chapter 11: Liberalization of markets

11.1 Liberalization of markets in the EU

In the 1980s, states began to relinquish control in a number of economic sectors that had been under their ownership since at least the post war era (telecommunications, energy, transport, postal services). For economic reasons and/or political considerations competition was excluded and in many countries the industries operated as a single, vertically integrated, state-owned monopoly. However, state monopolies were called into question by three considerations: economic, technological and political. Today, the provision of public services does not allow one to make a prima facie case for excluding competition. Instead the traditional approach to public life, based on stewardship and public duty, has been replaced by a market-oriented approach to the delivery of public goods and services. One useful way to describe the changed relationship between the state and newly liberalised industries is to deploy the concept of the regulatory state. That is, the relationship between the state and industry is no longer characterized by state ownership, support and control but by the following four features: privatisation, liberalization, re regulation, and the use of independent bodies to carry out regulatory tasks in the public interest.

Role of EU competition law in shaping the regulatory state. First, we explain how liberalization in the EU occurred: what economic, political and institutional configurations shaped this development.

Liberalization occurs by identifying which of the several components of the state monopoly can afford to have more than one supplier, and abolishing the monopolists exclusive rights in these segments. On this premise, there is no reason why many undertaking cannot compete to sell more attractive and useful handsets to make calls, or provide cheaper services. Conversely, some other market segments would not work better with competition. The local loop is one such market. The first step for liberalisation is identifying markets that would benefit from the entry of more players and abolishing state-granted monopoly rights in these segments, tolerating monopoly in sectors where competition would not make a difference or would be undesirable. Merely creating competitive markets does not guarantee competition. Specialised agency regulates prices and access to essential facilities in the telecommunications sector and detailed rules apply to ensure access in certain airline markets. Those special rules are risk based: the remedy is imposed upon proof of market power and risk of harm, not upon proof of misconduct.

The introduction of competition requires the creation of new markets by stripping down state monopolies. Markets need three kinds of legal support: property rights, governance structures and rules of exchange. Property rights in this context determine whether the state can continue to own certain enterprises, and whether a monopoly must surrender some of its property rights and give access to competitors. Governance structures are the rules that allow for competition, and rules of exchange determine the terms under which undertaking can trade with each other and with the consumer. Judged against these prerequisites, the legal tools that the Community has at its disposal are incomplete. Community cannot designate property rights widely. This is due to Article 295 EC which provides that the Treaty shall in no way prejudice the rules in Member States governing the system of property ownership. Therefore state ownership is lawful. Article 295 means that the Community cannot compel Member States to privatise an industry.

However, it can challenge the exercise of property rights when this goes against the Treaty, so, property owners may have to share essential goods or services. The Community has the power to alter governance structures and so liberalize markets. Two articles are of particular importance in giving the Community the legislative power to create markets.

Article 86 EC: 1) In the case of public undertakings and undertakings to which Member States grant special or exclusive rights, Member States shall neither enact nor maintain in force any measure contrary to the rules contained in this Treaty, in particular to those rules provided for in Article 12 and Articles 81 to 89; 2) Undertakings entrusted with the operation of services of general economic interest or having the character of a revenue producing monopoly shall be subject to the rules contained in this Treaty, in particular to the rules on competition, in so far as the application of such rules does not obstruct the performance, in law or in fact, of the particular tasks assigned to them. The development of trade must not be affected to such an extent as would be contrary to the interests of the Community; 3) The Commission shall ensure the application of the provisions of this Article and shall, where necessary, address appropriate directives or decisions to Member States.

This provision is ambiguous: Article 86(1) may be said to oblige Member States to ensure that state monopolies are subject to competition law it is a liberalization provision; on the other hand it provides that state monopolies are permitted, moreover when read with Article 86(2) the conclusion is that state monopolies are not subject to competition law if this would prevent the provision of services of general economic interest. Given that historically the basis for having state control over network industries was precisely to guarantee the provision of a service to all citizens in the public interest, Article 86 can be read to justify the continued existence of state monopolies. The ambiguous drafting reflects a political compromise between Member States in favour of continued public sector presence and the more liberal states. However Article 86(3) gives the Commission the power to legislate, and with appropriate judicial support (which arrived in the early 1990s), this can provide a rout to inject competition in sectors shielded by national law. This provision played a particularly important role in kick starting liberalisation of the telecommunications sector and in cajoling Member States to agree to liberalisation of the electricity sector.

The second tool is Article 95, which has been the legal basis for the vast majority of the Community's liberalisation initiatives. Legislation under Article 95 is designed to improve the conditions for the establishment and functioning of the internal market. It does so by harmonising national laws. Under this provision the Community can write laws that allow, say, a British courier company to deliver post to Spain. This is achieved by harmonising the laws about which types of postal services are excluded from competition, thereby banning state monopolies outside those markets. Harmonisation is justified by finding evidence that diversity of national regulation hampers the functioning of the internal market. This allows for an indirect challenge to national laws: they are not attacked for their protectionism, but because diversity frustrates the single market.

Of these two legal grounds for liberalization legislation, Article 86(3) is less politically correct because the Commission enacts the directives independently of other Community organs. In contrast, Article 95 legislation requires approval of legislation by the European Parliament and a qualified majority of the Council. The conventional distinction between Articles 86(3) and 95 is that the former is a liberalisation instrument, and the latter a harmonisation instrument. Such a neat relationship is not accurate.

First, the Court suggests that there is some overlap, so that the Commission would be able to use Article 95 for the purposes of liberalisation; second, the early telecommunications directives show that some harmonisation was carried out with Article 86(3). Article 86(3) was not used to dismantle exclusive rights in sectors other than telecommunications. As we suggest below, Article 86(3) directives are better seen as a tool that the Commission can use to force liberalisation strategically when political agreement under Article 95 is stalled by Member States.

Two doctrines developed by the ECJ have been an important complement to the legislative powers in the Treaty because the Court provided a legal basis to challenge anticompetitive state regulation, which allowed the Commission and Council to legislate to open markets to competition. Court rulings have also led to individual traders challenging anticompetitive state legislation that hampered their economic opportunities. The Court’s case law contributes to the construction of markets by facilitating positive integration or by achieving negative integration.

Article 10 EC obliges Member States to abstain from measures that could jeopardise the attainment of the objectives of the Treaty. The Court has decided that the obligation in Article 10 may be read jointly with Articles 81 and 82 and so prohibit any action by a Member State, which encourages undertakings to infringe the competition rules. This interpretation has been particularly useful in challenging state meddling in the air transport sector. One enlightening case is Ahmed Saeed. The defendants, two travel agents based in Germany, obtained airline tickets from airlines or travel agents established in another state. Although the starting point for the journey mentioned in those tickets was situated outside Germany, passengers who bought tickets from the defendants boarded their flight at a German airport where the scheduled flight made a stopover. In spite of the longer flight, these tickets cost less than tickets for flights departing from Germany. The travel agents’ antics were challenged by the association for the campaign against unfair competition, who argued that by selling cheap tickets the defendants contravened German law, which prohibits the sale of tickets at prices which are not approved by the competent federal minister. It transpired that air tariffs for flights from Germany were agreed among airlines flying from German airports and the government then approved those fees, which made tickets expensive. In other words, it was a government-sanctioned cartel. The Court of Justice held that the tariff agreements were in breach of Article 81, and that the German government was also acting unlawfully (in breach of Article 10 read jointly with Article 81). The Court ordered that the state must refrain from taking any measure, which might be construed as encouraging airlines to conclude tariff agreements contrary to the Treaty. The effect of this finding is significant: not only can travel agents now buy cheap tickets from abroad as the defendants had done, but as the government is no longer able to protect the airlines cartel, this can be challenged by a competition authority, which will cause prices to fall as airlines are forced to compete against each other for customers. This is an example of using EU competition law to achieve negative integration, as the judgment necessarily reverberated in every Member State that supported airline cartels: all became subject to legal challenge and, as we shall see below, the Commission seized on this to press for liberalisation.

The Article 10 doctrine is based on the need to ensure the effectiveness of competition law: the competition rules prohibiting anticompetitive conduct would be emasculated if undertakings could lobby national governments and ask for legislation that legitimises cartels. But the doctrine has two limits. First, it only applies if one can establish a link between state action and an agreement or an abuse of a dominant position. Either state law requires undertakings to breach Articles 81 and 82, or it encourages anticompetitive behaviour, or it allow undertaking to enter into a collusive agreement and then ratifies their decisions. On the contrary, if a national law has the effect of stifling price competition this is insufficient to invoke the rule. There must be a breach of competition law by the undertakings, which is causally connected to the action of the state. Second, the doctrine is inapplicable if the state allows a group of undertakings to enter into certain agreements, for example lawyers agreeing to fix fees for legal services, but the state then actively supervises the fees that have been agreed to make sure that they are compatible with the public interest. In this case the doctrine does not apply because the fees are subject to effective supervision by the state according to clearly articulated public interest considerations. When public interest considerations are inherent in the procedure that restricts competition, competition law does not apply.

The Commission’s liberalisation strategy was legitimated and strengthened by a radical set of judgments in the early 1990s interpreting Article 86(1) where the Court fashioned standards to test the legality of state monopolies. The Court regularly says that the creation of a monopoly by the grant of an exclusive right is not contrary to Article 86(1). It also attacks state monopolies when national legislation undermines the effectiveness of Articles 81 and 82. One does not have to search for an infringement by the undertaking, but only for whether the creation of a monopoly has or may have anticompetitive effects. The Court has identified three ways in which the grant of special or exclusive rights undermines competition. First, the grant of an exclusive right is unlawful when the beneficiary is not efficient. In the leading case, Hofner, Germany had granted a monopoly in the market for recruitment services, banning private recruitment agencies. This had been done in the public interest, to ensure a well-functioning employment market where recruitment agencies were widely available. It transpired that the state services were inadequate. In these circumstances, the state could be in breach for its regulatory or decisional intervention. That is, there is no infringement merely because the undertaking is managed badly, but there is an infringement if the state grants the monopoly to a undertaking that it knows is poorly managed or under resourced, or if the state fails to take steps once it sees that the undertaking is inept. Under this rule the state breaches Article 86(1) only if the provision of the service by the state monopoly is manifestly inadequate, providing the state a margin of appreciation as it carries out its regulatory tasks. The German government was bound to take these criteria into account to test whether the exclusive right was justifiable or if the market for employment services should be opened to competition. This doctrine embodies a powerful message from the Court in favour of liberalisation when state intervention leads to an inefficient allocation of resources. But the case law does not demand liberalisation. Rather the state that chooses, for public policy reasons, to grant an exclusive right is under an obligation to monitor the performance of undertakings to which it grants these rights on a regular basis, for the service may be satisfactory at its inception but become unable to meet demand as circumstances change.

If the state monopoly is inefficient the state may cure the inefficiency (by increasing its finances through taxation) but liberalisation is not compulsory, although it may become necessary if the cost of financing the service through taxation is politically unpalatable.

Second, the case law prohibits the grant of exclusive rights when this creates a conflict of interest and makes it likely that the undertaking will use these rights to exclude rivals. In RTT v. GB INNO, Belgium had granted the company holding a monopoly over the operation of the public telecommunications network (RTT) the sole power to lay down standards for telephone equipment and to check whether suppliers of equipment met those standards. RTT was also active in the market for telephone equipment. The upshot was that RTT was regulating market access of its competitors. The Court held that this would undermine equality of opportunity between economic operators, placing the undertaking in charge of regulating access at an obvious advantage over its competitors as it could raise entry barriers to competitors. Using a similar approach in ERT, the Court reviewed the Greek radio and television monopoly. The monopolist enjoyed exclusive rights over both the transmission of its own programmes and the retransmission of third party programmes, including those produced in other Member States. The Court held that Article 86(1) prohibits the grant of such exclusive rights where they are liable to create a situation in which the undertaking is led to infringe Article 82 as a result of a discriminatory broadcasting policy in favour of its own programmes. Generalised, these two cases suggest that vertical integration in network industries is problematic. It can allow the owner or controller of the network to exclude competitors, or at least raise their costs to the advantage of its upstream goods or services. Referring back to the diagram on page 443, the Courts case law suggests that there should be unbundling between the monopoly and competitive segments.

Third, the Court held that a Member State would infringe Article 86(1) by granting unnecessarily wide monopoly rights to one undertaking. In the leading case, a Belgian law granted a monopoly over a wide range of postal services to the Regie des Postes. Mr Corbeau flouted the law by offering an express postal service in the city of Liege, in competition with the state monopoly. When criminal charges were brought against him for breaching the Belgian postal monopoly, he challenged it under Article 86. The Court of Justice condemned the existence of the overly extensive monopoly rights granted to the Regie des Postes, but added that the grant of an extensive monopoly over several service markets to one undertaking could be tolerated if this was necessary to guarantee the provision of a universal postal service. If Belgium wished to reserve the (profitable) market for express delivery of mail to the same undertaking that also operates the (unprofitable) nationwide postal service, then the grant of both monopoly rights could be justified because operating the profitable sector helps to pay for the provision of a public service. However, if this justification is absent, then excessive monopoly rights are conferred unlawfully. This ruling builds on the RTT and ERT cases. There the Court condemned vertical integration, the grant of exclusive rights up- and downstream, because the undertaking was in a position to raise rivals costs. In Corbeau the Court considered the denial of market access that results when the state grants monopoly rights across too many markets, suffocating competition in potentially competitive markets.

Flipside of these rulings is twofold. First, the grant of a monopoly in one market is subject to an efficiency benchmark: Member States must ensure that the privileged undertaking is able to meet demand. Second, the grant of a monopoly to a undertaking present in more than one market is unlawful in two circumstances: when it creates temptations for the undertaking to use its market power to exclude others in related markets, and when granting a portfolio of monopolies to one operator is not necessary to finance the provision of a public service. These doctrines have the following effects. First, they legitimate liberalisation, by showing that Member State laws granting exclusive rights are incompatible with the principles of an open market economy. Second, they establish principles that find their way into the legislation. Third, they create incentives for private litigants and the Commission to pursue individual Member States, which can serve as a catalyst towards EU wide liberalisation.

Liberalization is accompanied by the creation of independent National Regulatory Authorities (NRAs) to monitor newly created markets. National Regulatory Authorities from a political science perspective, there are three reasons why governments delegate regulation to an independent authority: 1) the agency can specialise in the sector and gain expertise, making for better regulation; 2) governments can shift the blame to the agency when the liberalised market does not work well; 3) creating the agency demonstrates a credible commitment to efficient markets. Similar reasons motivated the creation of National Competition Authorities, but an unexpected effect was that many NCAs grew in prestige, distanced themselves from political control and increased their powers, well beyond those which states anticipated. It is likely that these effects will also occur for NRAs. First, the liberalization directives prescribe that NRAs must be independent of the undertakings they regulate. Second, the directives entrust specific tasks to the NRA. These two elements make it less possible for government to circumscribe the NRSs powers.

The regulatory framework empowers NRAs to impose obligations on certain market players, with considerable guidance from the Commission. The most burdensome obligations can be imposed on operators that have significant market power. The regulatory process begins with the Commission publishing a recommendation on relevant markets it believes may warrant NRA intervention, and guidelines for the assessment of market power. NRAs then take these two soft law instruments into account and review national markets to determine whether there are any instances where one or more undertakings have significant market power. The concept of significant market power (SMP) is analogous to the concept of dominance, and the procedure described resembles the start of an Article 82 inquiry, whereby markets are defined and dominance analysed. If the NRA finds that there is no SMP, then it must remove any regulatory obligation, while if it finds that certain undertakings hold SMP, then it is empowered to regulate the markets. This process can be characterized as a risk-based assessment whereby a finding of SMP triggers the application of the law, not unlike the procedure we have seen in the context of vertical restraints where the lack of market power allows for the application of a Block Exemption, or in the context of merger cases where thresholds are used to signal the potential presence of anticompetitive risk. The NRA’s draft analyses of market definition and market power are subject to review and veto by the Commission, while the NRA has greater independence in designing the appropriate remedy. The Commission has identified eighteen markets where regulation might be warranted, but two examples in the mobile phone sector will suffice to offer a practical illustration of how the process works in practice.

The first is access and call origination. Assume that undertaking wants to offer mobile phone services. It might want to do so by using the upstream network of another existing mobile phone operator (undertaking A). There is a risk to competition if the upstream network owner does not allow new entrants access, because it can reserve the downstream market to itself, or it can operate a price squeeze to make entry by undertaking B less profitable. If the NRA finds that there is SMP on the upstream market, it may impose one or more remedies on undertaking a drawn from the Access Directive.

Universal Service Directive allows the NRA to impose remedies on retail markets only upon proof of SMP and when wholesale regulations would not serve to promote competition.

But SSR is not limited to players that have substantial market power. There are also obligations that the NRA may impose upon all market operators without proof of SMP so as to guarantee that the market is more competitive. These include the obligation on all network operators to negotiate interconnection so that they can accept incoming calls from other networks. This ensures that all operators can deliver the calls originating in their network to any other. The policy underpinning these obligations is to ensure that there is a complete network, which benefits users and also each operator: as the value of a network increases the more persons one can communicate with through that network. Another example of SSR that applies across the industry is number portability. The NRA must ensure that users who switch from one service provider to another can keep the existing number. This is an effective means of facilitating consumer choice, which could not be achieved by competition law.

In regulating electronic communications, the NRA acts like a competition authority, and it is difficult to draw clear boundaries between regulatory and antitrust functions. The major distinction between SSR and competition should be this. While competition policy tells undertakings what not to do, regulation involves telling them what they are to do. This distinction breaks down because the Commission has used competition law in a sector-specific manner, to guide undertakings to reduce excessive prices and to grant access to essential facilities. Accordingly, it is legitimate to ask whether sector-specific regulation is necessary, given that competition law seems sufficiently broad in scope to achieve regulatory functions.

There are a number of ways to defend the role of SSR when it overlaps with competition law. First, we can condemn the Commission for corrupting competition law and extending it illegitimately to carry out the task of sector specific regulation. Second, an NRA has a comparative advantage over competition authorities, based on expertise and resources, making the NRA a better institution for regulating price and determining access. A competition authority intervenes in markets episodically; it lacks the resources for day-to-day management. It also lacks adequate remedial tools, when access remedies are imposed the Commission often subcontracts supervision to third parties. NRAs are better equipped to review systemic market failures. But this is an argument in favour of empowering the NRA to apply competition law and it does not make a case for SSR. Third, according to the Commission, there is a need for risk based regulation. That is, rather than wait for a dominant undertaking to refuse to supply, it is necessary to establish obligations to avoid further deterioration of the market. But, this justification is weak in that often Article 82 is applied prudentially before the abuse has caused harm, and conversely SSR remedies are likely to be imposed when the NRA has some experience that a undertaking with SMP is acting anticompetitively.

A fourth justification is that SSR can be more aggressive than competition law. The obligation not to discriminate is probably wider under SSR than under Article 82, and SSR applies to regulate price, something that general competition law eschews. The obligation to provide access is also more extensive under SSR than under Article 82. In the field of electronic communications NRAs may compel access when refusal would either hinder the emergence of a sustainable competitive market at the retail level, or would not be in the user’s best interests. NRA can force the grant of access rights to as many downstream competitors as it wants until it considers that the retail market is ‘sustainably competitive’, a phrase that eschews precise definitions, giving the regulator considerable flexibility. Moreover, the NRA has a second ground upon which it can require access: the interest of users, which seems to give it even more power in imposing an access remedy. The flexibility afforded to the regulator poses a risk that decisions will focus on short-term consumer interests at the expense of long-term interests: granting access too frequently risks denting incentives of new entrants on upstream markets.

These four justifications suggest that there is no neat borderline that divides the task of SSR and that of competition law: both address similar market failures with similar rules.

The Commission also takes the view that SSR should be phased out in the long term as markets become workably competitive, so that only competition law will apply. The legislative mandate for the NRAs is threefold. They must apply the law in order to promote competition, contribute to the development of the internal market, and promote the interests of citizens in the European Union. They may also use SSR to promote cultural and linguistic diversity and media pluralism. These goals suggest a long term future for SSR. Moreover, the economic and non-economic goals indicate that the regulation of this industry cannot be carried out solely by reference to modified competition law rules. SSR could be best aligned with industrial policy rather than competition policy.

NRAs operate in a complex environment, where three possible conflicts may arise. Again, we illustrate these with reference to electronic communications. The first is a ‘horizontal’ conflict between the decision of an NRA and principles of national competition law. The second is a ‘vertical’ conflict where the approach of the NRA may be inconsistent with the Community’s regulatory framework. The third is a ‘diagonal’ conflict between one set of supranational rules (EC competition law) and one set of national rules (SSR). Beginning with the vertical conflict: the Commission is able to exercise control over NRAs in a number of respects provided for in Article 7 of the Framework Directive. In brief, each NRA is called upon to make three determinations: define relevant markets, determine if a undertaking has SMP, and impose remedies. The NRA must communicate any draft measure (market definition, determination of SMP, or proposed remedies) to the Commission and Member States who have one month to comment. The Commission may consider that the definition of markets outside those identified by the Commission or the determination of SMP creates a barrier to the single market or is otherwise incompatible with the aims of the law, in which case it has two additional months to analyse the draft determination further and may veto the measures. There are no veto powers regarding the remedies that the NRA selects. Instead, a coordination mechanism has been put into place so that Member States and the Commission agree on suitable remedies. Article 7 procedures are carried out by DG Competition and DG Information Society, which enables the Commission to pool its expertise in market analysis and the development of the markets. The control mechanisms are not exactly the same as those established under Regulation 1/2003.

In the latter National Competition Authorities are free to decide what investigations to carry out, but their final decision is subject to veto as the Commission can review the draft decisions. However, the objectives of the review processes are the same: to guarantee consistency and provide legal certainty.

The Commission has yet to apply its veto power to draft competition decisions notified under Regulation 1/2003 and has exercised its veto power in the electronic communications field only four times to date. However, potentially more significant than the exercise of veto power are the Commission’s informal prompts to Member States to take certain action. Moreover, the network of NRAs is likely to furnish further peer pressure on recalcitrant NRAs, so that informal means are likely to be more useful in avoiding vertical conflicts. As far as horizontal conflicts go, the Community has left Member States to manage these. In the field of electronic communications Member States must ensure that competition authorities and NRAs consult each other, cooperate and share relevant information. In the postal sector in contrast the directive provides that the NRA ‘may’ be charged with the application of competition law in this sector Single agency in charge might be better able to manage SSR and competition rules by selecting the most appropriate remedy. A related horizontal conflict is between the NRA and a national court. A plaintiff seeking a remedy (access) can have recourse to the NRA, NCA or court simultaneously to try and obtain a favourable solution. Moreover, as the plaintiff can use EC and national competition law, this gives rise to diagonal conflicts. And to complete the picture, an additional diagonal conflict may arise when the Commission wishes to apply competition law in a market where the NRA has already taken action.

The general rule in diagonal conflicts is this: unless the directives on which SSR is based provide that competition law does not apply when SSR does, then it is possible to apply competition law even if the market has already been regulated under SSR. This approach was taken in Deutsche Telekom. In this case the Commission found that Deutsche Telekom (DT) had abused its dominant position by deploying a ‘price squeeze’. DT held a dominant position in the upstream market for the local loop, and was active in the downstream market for broadband internet connections. Would-be competitors sought access to the local loop, but DT charged them a wholesale price that was higher than the retail price that DT set for its broadband services. In this setting, competitors would be unable to mount any meaningful challenge since they could not offer end users a broadband connection cheaper than DT. It did not matter that the German NRA had already regulated both wholesale and retail prices: DT was not required to set prices fixed by the NRA and was still capable of selecting price levels that would not have squeezed competitors from the market.

The effect of this approach to diagonal conflicts is that the Commission is able to supplement its supervisory powers. Under the electronic communications directives the Commission can monitor what the NRAs choose to do, and using EC competition law the Commission can take action when NRAs fail to do so.

The similarity between SSR and competition law has more to do with the fact that they address similar economic problems than with any transfer of principles developed in one field. Moreover, SSR has a broader economic mandate. More preoccupying phenomenon is the application of competition law in a sector-specific manner. This manifests itself in two ways: 1) Commission uses competition law to set out its legislative intentions and anticipates the liberalisation of markets; 2) competition law is used to supplement or correct SSR.

We illustrate this use of EU competition law with examples drawn from a range of industrial sectors that have undergone liberalisation in the sections that follow.

The remedies that the Commission obtained and imposed show the regulatory streak in this decision. First, DPAG entered into a voluntary undertaking to divest its parcel service business. This was in response to the Commissions Statement of Objections (the document issued to the parties when the Commission starts infringement proceedings) where it noted that structural separation would prevent cross-subsidisation. DPAGs undertaking is complex but worth summarising in detail. A new undertaking (Newco) would be created to carry on the parcel business. If Newco were to source any part of its activities from DPAG (using DPAGs sorting facilities for processing parcels) then it would pay for these at market prices. Moreover, DPAG undertook that any service that Newco sourced from it would also be available, at the same price, to any other parcel service undertaking. Finally DPAG undertook to provide the Commission with reports on prices and Newco’s costs and revenues and guaranteed to keep the accounts of DPAG and Newco separate to ensure full transparency. The effects of this remedy are to prevent cross subsidisation, to facilitate its detection, and to facilitate the entry of new competitors. Remedy goes beyond preventing the recurrence of the abuse, butt also establishes a means to monitor the prices charged by Newco and facilitates market access by ensuring that new entrants can have access to some elements of DPAGs infrastructure should these be seen as essential by Newco. This means that if any segment of the parcel delivery market has natural monopoly qualities, access is guaranteed to new entrants, ex ante.

The Commission also imposed a second set of regulatory remedies. DPAG must submit a statement of Newcos costs and revenues, and an itemised statement of any prices that Newco paid for services that it procured from DPAG. DPAG must also submit details of any rebate schemes that Newco concluded with its six largest mail order customers. These reporting obligations allow the Commission to ensure Newcos prices are not predatory and its rebates not loyalty inducing. These remedies are remarkable because no comparable obligations are imposed when undertakings commit these abuses in competitive markets. The Commission’s intervention is usually an order to cease and desist from the practice.

The remedies go further than what is normally achieved in an Article 82 case: they are designed to prevent future abuses. To facilitate entry by allowing access to DPAGs facilities even without a finding that the facility is essential; and to prevent cross subsidisation by structural separation. Moreover, the remedies also go beyond what had been achieved by the Postal Services Directive in 1997. The directive merely tried to prevent cross-subsidisation by requiring that the undertaking with a reserved sector must keep separate accounts for non-reserved services. The Commission’s decision seems to cure a regulatory failure that plagues many other sectors. The liberalisation directives seek to prevent cross-subsidisation by providing for accounting separation but the Commission has been unable to press Member States to agree to more radical structural measures that would make cross-subsidisation more difficult to practise and easier to detect. However, some have queried how far the structural separation in DPAG is effective given that Newco will still be owned by the shareholders that control DPAG, so the possibility of cross-subsidies remains and can only be eliminated if Newco is sold to an independent buyer.

The background to the decision demonstrates the ambiguities and inconsistencies in the Community’s approach to liberalisation. In spite of the powerful legal tools that the EC has to create markets, its efforts depend upon Member States agreeing to reforms. In this market, powerful domestic interests prevailed over the strategic use of competition law that UPS attempted. Nor is the tension merely one between a Commission bent on liberalizing markets and protectionist Member States. At the same time that the Commission was seeking to open up the market, it tolerated the growth of DPAG through merger: after the Commission approved the joint venture between DHL and DPAG, UPS launched an action against the Commission complaining that DPAG had financed its acquisition of DHL through its revenues in the reserved sector and that this was an unlawful use of its monopoly position in breach of Article 82. However, the CFI disagreed, ruling that so long as the revenues used for the joint venture did not come from abusive practices (excessive pricing) DPAG was free to use the funds as it wished. On a formal level, the CFI’s ruling is unproblematic in that, as we noted before, cross-subsidisation per se is not an abuse of a dominant position. Nevertheless, if the Community is seriously interested in enhancing competition in the postal services market, it should recognise that the effect of DPAGs acquisition with money raised from the universal services is as inefficient as its financing of predatory pricing campaigns. There is also some conflict between the Commission insisting on structural separation in DPAG while, in clearing the DHL joint venture, it was satisfied with an undertaking from DPAG that it would not use revenues from the reserved sector to finance the operational costs of DHL.

The Commissions tolerance of DPAG’s growth through mergers may be interpreted in two ways. One is to see it as part of the Commission’s industrial policy, whereby it supports the creation of global. Second reading is that tolerating DPAG’s actions can be a means to press for greater liberalization later: mergers allow former monopolies a means to adjust to a competitive market by diversifying their activities and once DPAG and other former postal monopolies become financially secure, the Commission can have more leverage in negotiating further liberalisation. With tolerating the creation of a highly concentrated market as a means of injecting greater competition seems to be a high risk strategy.

The application of competition law is sensitive to the relevant sectors in many ways: the Commission launches cases strategically and interprets the law expansively to achieve regulatory ends. In many cases the remedies are ‘agreed’ with the parties, not imposed by the Commission. This seems to be done in good faith. That is, the Commission sees a political failure and steps in to get the market working. In the mid-1990s a similar trend was noted in US antitrust laws. Commentators noted that the agencies were increasingly settling cases by which undertakings agreed to modify their conduct to avoid a trial. Commentators were worried that competition law was moving from a ‘law enforcement’ model to a regulatory model. In a law enforcement model, the focus is on identifying an infringement, while a regulatory model focuses on the remedy. A comparable pattern is exhibited in the examples we have considered above. According to American critics, this change in the way enforcers thought about antitrust problems was undesirable for the following reasons. First, the remedies that the antitrust agency can get with a settlement are much more extensive than those it could get through litigation. Second, a settlement means that the undertaking is under constant supervision by the agency because if the undertaking wants to alter its conduct given changes in the marketplace, it may need approval from the agency should the proposed conduct infringe the settlement.

Third, settlements avoid litigation and so pretermit the need to show that the behaviour was unlawful. The upshot of this is that settlements are more about what the government wants than what the law prohibits.

The Community's attempt to create a competitive, Europe-wide market for utilities has regularly been opposed by those concerned that efficiency is in conflict with the public service obligations that utility companies have had, in particular under certain national traditions. The EU has managed the relationship between the economic objectives of the internal market and the social objectives of public service provision in three ways: first, by stating that certain public services, in particular those closely associated with the welfare state (social security and insurance schemes), fall outside the scope of EU competition law; second, by tolerating certain anticompetitive arrangements when these are necessary to finance the provision of public services; third, by imposing minimum public service obligations to be provided in all Member States. The first two approaches allow Member States to implement national policies to safeguard public services, while the third approach establishes public service obligations at a European level, and it is notable for two related features: first, the belief that public services can normally be delivered through competitive markets, and so suspending competition should be exceptional; second, an emphasis on consumer choice as a means to monitor public service delivery.

The subject of competition law is the undertaking. In the jargon of the EC Treaty, a undertaking is known as an undertaking, and the Courts have defined it as any entity engaged in economic activity, regardless of the legal status of the entity or the way, which it is financed. Any activity consisting in offering goods and services on a given market is an economic activity. This means that a self employed person or a multinational corporation are both undertakings. Court has elected to exclude two types of activity from the scope of competition law. First, when the activity falls within the essential prerogatives of the state it is excluded. No matter how much sovereignty is relinquished as a result of continuing economic and political convergence in the EU, certain aspects of statehood are not subjected to EC law. We find these in the EC Treaty (Article 296 in relation to defence) and in the case law of the Court of Justice, which has extended this exclusion to activities deemed to be part of the essential function of the state, such as air navigation and certain antipollution services. The focus is on the activity, not the actor – thus a body may act as an undertaking for some functions but not others. For instance, an antipollution service exercises state-like powers when it provides antipollution services and collects charges for these, and probably also when it purchases equipment necessary to provide these services. But it would be treated as an undertaking when carrying out acts unrelated to its state-like powers, for example if it also provided consultancy services for private undertakings wishing to install antipollution devices in their factories.

Second, the Court has restricted the application of competition law to certain aspects of the welfare state, especially when compulsory social security schemes have been challenged. The dynamics of the litigation that has allowed the Court to exclude competition law from the welfare state are similar. Person is required to make payments towards a state run social security scheme. but finds that there are better insurance provisions in the private sector, so she stops making compulsory contributions and obtains the relevant insurance from the private sector. When her failure to pay is challenged, she relies on the EC Treaty to argue that the state-run scheme restricts competition, so she is able to disregard it and buy the relevant service in the private sector.

Faced with these disputes, national courts sought the advice of the ECJ whose answer is that if a national scheme operates on the basis of the principle of solidarity, then the administrative authority is not acting as an undertaking, so the competition rules are inapplicable, provided that compulsory membership in the scheme is necessary to ensure the schemes financial equilibrium. On the contrary, if the scheme is not solidarity based, then the operator is an undertaking and it can be challenged under the competition rules. Two early cases (Poucet and FFSA) set out the Courts interpretation of solidarity. In Poucet the Court analysed sickness and maternity insurance schemes for self-employed persons and a basic pension scheme for skilled workers, while in FFSA the Court analysed a supplementary old age insurance scheme for self employed farmers. The schemes in Poucet were not operated by undertakings because they were compulsory; the sickness and maternity insurance scheme conferred equal benefits irrespective of contributions, and contributions were proportionate to income. The old-age pension scheme was financed by current workers, pension rights were not linked to contributions and the schemes that were in financial difficulty were subsidised by those that had a surplus. Accordingly, there were several manifestations of solidarity. Between rich and poor workers, between high risk and low-risk workers, between today’s workers and those who had retired, and between profitable and non profitable schemes.

Solidarity was not present in FFSA: the scheme was optional, benefits depended on the contributions made by each individual and on the financial results reached by the body operating the scheme. While the scheme was designed to pursue a social objective and it had to operate on a non profit basis, these elements were insufficient to declare the operator not an undertaking, so the scheme was subject to competition law. The major difficulty with the Courts case law is the imprecise nature of solidarity, so that the Court looks for manifestations of solidarity and declares that the scheme falls outside the scope of EU competition law when there is a sufficient degree of solidarity. The risk is that state social security policy occasionally falls to be regulated by EU competition law.

A related criticism is that solidarity is inherent in many other so that the special exclusion for social security schemes based on the principle of solidarity is hard to justify. Another problem is that proof of solidarity is insufficient: the Court also demands that the alleged restriction of competition (compulsory affiliation) must be essential for the financial balance of the scheme. Court is exempting certain schemes rather than excluding the application of competition law, so that the analysis in the following section might be more appropriate.

Public services were usually provided by the state because it is difficult to tempt the private sector: the cost of providing a universal electricity or postal service to all citizens at affordable rates is usually higher than the profits that can be generated. Public services are financed in ways that restrict competition, in particularly by facilitating cross-subsidisation. A example is the Corbeau litigation: in order to finance a universal postal service for all Belgian citizens, the government had granted the Regie des Postes a monopoly over a number of other profitable postal services, in the belief that the revenue generated in the profitable sector would pay for the cost of running a universal service. This method of finance goes against the policy of liberalisation, since it reserves monopoly rights over potentially competitive markets. The Court has recognised that financing public services through restrictions of competition may be unavoidable, and it has interpreted Article 86(2) EC in an increasingly complex manner to accommodate this method of financing.

Article 86(2) provides a defence for undertakings and Member States whereby Treaty obligations are disapplied to allow the delivery of public services. In order for state laws to benefit from this exemption, the following must be shown: 1) that there is a service of general economic interest, the performance of which is entrusted to the undertaking in question; 2) that the anticompetitive effects are necessary to allow the operator to ensure the performance of the service of general economic interest under economically acceptable conditions; 3) that the service provider is able to offer the services in question efficiently.

The first criterion allows Member States to define what economic services they believe to be of general interest. That is, the service is of an economic nature, so that Article 86(2) does not apply when the activity in question is not that which can be carried out by an undertaking and the service in question is one which the state de emsto be of general interest. The right of Member States to nominate services is subject to two limitations: the first is if the nomination is manifestly erroneous, and the second is that the public service mission must be clearly defined and explicitly entrusted to the undertaking in question by the state. In addition to the services provided by network industries (water, electricity, postal services, telecommunications), the Court has recognized that there is a general interest when an airline serves routes that are not commercially viable but which provide connections to remote areas of a country, when mooring services for ships are made available around the clock to ensure safety, and where a waste management scheme is operated to address environmental risks. Second, the restriction of competition is subjected to a proportionality test. In the context of cross subsidization, the state can justify the grant of monopoly rights in a potentially competitive market (parcel services) when the profits earned by the undertaking in the reserved market facilitate its performance of the service of general economic interest. Without foreclosing profitable markets, private undertakings would cherry pick profitable services, leaving the unprofitable sector to be covered by the state or not provided at all. The effect of Article 86(2) is to allow states to attract private undertakings to offer public services and, in exchange for their agreement to offer these services, to gain certain advantages in the form of profits in some profitable markets reserved to them. The proportionality requirement has seen read increasingly leniently by the Court and the Commission. In the early years a restriction of competition was only tolerated when it was the only feasible way of performing the service of general economic interest. Now Courts grant the exemption without the need to show that there are no less restrictive means of achieving the state goal, provided that the restriction allows the undertaking to provide the relevant service under economically acceptable conditions. This change is attributed to the increasingly strict interpretation of Article 86(1), which has led the Court to expand the possibilities for states to defend the grant of exclusive rights.

The third requirement indicates that the derogation is granted conditionally, the undertaking must provide the services efficiently. If the undertaking is manifestly unable to do so, then the derogation will become invalid. This requirement was introduced by the Court in Ambulanz Glockner.

The efficiency condition to the Article 86(2) exemption can be compared to the approach that the ECJ has taken in situations where the Member State decides to finance the public service provider out of public funds. Normally, subsidies provided by states are regulated by the rules on state aid, but the Court of Justice has declared that subsidies which merely compensate a undertaking for providing a service of general interest do not constitute state aid. In reaching this conclusion the Court has set out four requirements: first that the public service obligation is clearly defined, second that the compensation is established objectively and transparently, third that it does not exceed the cost of providing the service plus a reasonable profit, and finally that ideally the provider should be selected through a competitive tendering procedure. The last condition is important because it means that the state must auction the right to provide the public service to the most efficient undertaking. But under Article 86(2), there is no need to show that the most efficient provider has been selected, only that the undertaking offering the service is able to do so efficiently; which means that the public service is not necessarily offered at its lowest cost when Article 86(2) is applied, so that an auction model should also be developed in this context.

Court has recognised the right of Member States to suspend the application of competition law to ensure the efficient provision of public services, the Council, in its liberalisation directives, has begun to establish EU wide public service obligations. This serves two distinct legitimising functions. First, it legitimises Community intervention by pointing to the necessity to safeguard the general interest at Community level. Second, it is used to show the sceptical European citizen how Community law impacts their daily life. The importance of legitimising public services is so significant that the directives liberalising postal services and energy begin by defining universal services

before dismantling restrictions of competition. While respectful of national concerns, EU public service law has two features that distinguish it from national public service law: 1) an emphasis on consumer interests; 2) preference for market solutions.

The European notion of public services embodies the following characteristics: 1) universality (that is, the service must be available to all consumers throughout the territory); 2) continuity; 3) quality; 4) affordability; 5) user and consumer protection. The liberalisation directives identify universal service obligations in each sector. In the field of electronic communications, European universal service obligations are most well defined, and consumers have a range of entitlements as well as means to enforce their rights. Each end user has a right to the following. Connection to the telephone network at a fixed location to make local, national and international telephone calls, use fax machines and access the Internet. Comprehensive telephone directory updated once yearly and a comprehensive telephone directory inquiry service; access to public pay telephones. The right to these services is universal, meaning not only across the territory, but that Member States must also ensure that disabled users have access to them, that tariffs are regulated so that all social groups are able to afford these rights , and that there is no other form of discrimination among users.

Public service obligations are monitored in two ways: 1) National Regulatory Authority has powers to check the quality and prices of the universal services, and set quality targets; the NRA may impose financial penalties on undertakings that fail to provide universal services satisfactorily and may even withdraw the licence to operate. NRA also protects users from arbitrary actions of their provider; 2) end users are afforded a range of means to protect their interests and rights. Service providers must publish information on prices, standard terms and conditions. Contract with a user must specify the service and quality level, types of maintenance offered, conditions for terminating and renewing the contract, compensation schemes if the quality levels are not met and procedures for dispute settlement. Consumers must receive information to control expenditure. Subscribers also have the right to cancel without a penalty if the supplier changes the terms of the contract, and must receive one months notice regarding any proposed contract modification. The emphasis on consumer rights is a distinctive feature of the European model of public services because it makes the end user a part of the regulatory infrastructure. Choices give suppliers incentives to provide improved services. This fits with the market oriented approach to universal services to which we now turn.

The second distinctive feature of European universal services law is the preference for market delivery. This is a significant departure from certain national systems where the states legitimacy is based at least in part on its ability to provide certain services personally. Public service provision must now be delivered by the market whenever possible. The clearest manifestation of this principle is in electronic communications. First, Member States must allow any undertaking to offer the universal services, and the recital to the directive goes so far as to suggest that the universal service obligations could be allocated to operators demonstrating the most cost effective means of delivering access and services. This means that there would be competition for entering the universal services market. Second, in terms of financing the universal service in cases where the cost exceeds profits, the directive provides that universal service providers may be compensated in a competitively neutral way. Member States have two choices: the undertakings receive compensation from the state in the form of subsidies, or other telecommunications companies pay a levy which goes to finance the universal service provider. No restrictions on competition are allowed to finance universal services in the electronic communication sector.

The powerful message of the liberalization directives is that public services should be provided, but through competitive markets. This vision is out of line with that of a number of commentators, who rely on Article 16 to put forward a less market-oriented view. Article 16 was inserted in the Treaty of Amsterdam amidst concerns about the Community's application of competition law to public services, with some Member States wishing for a means to exclude the application of competition law to public services. However, it is not immediately clear what this Article can deliver, as it gives the Commission no extra legislative powers, nor can it have direct effect, so the purported duty on the Community and Member States to take care that public service missions are accomplished seems unenforceable.

Article 16: “Without prejudice to Articles 73, 86 and 87, and given the place occupied by services of general economic interest in the shared values of the Union as well as their role in promoting social and territorial cohesion, the Community and the Member States, each within their respective powers and within the scope of application of this Treaty, shall take care that such services operate on the basis of principles and conditions which enable them to fulfill their missions.”

The creative legal routes devised by Commission and Court to create markets are remarkable but they have ultimately proven inadequate to persuade all Member States to follow the path to liberalization in a uniform, speedy manner. The Lisbon competitiveness agenda is threatened by the reluctance of Member States to achieve open markets and by increased protectionism. Judged from the perspectives adopted in this book, market creation may be assessed as a process where the relevant institutions (Council, Commission, European Courts and Member States) faced several moments of disagreement, which led to delicate and laborious negotiations in certain instances, or to unilateral action to accelerate liberalization. Interested actors tried to use certain institutions to achieve liberalized markets. The institutional disagreements have to do with the economic and political configurations in the institutions. Commission and the Court, have been more easily convinced that markets can function to deliver goods and services more efficiently than state monopolies.

Turning to market regulation, we have seen the emergence of a European regulatory state. An institutional perspective on the regulatory state suggests that decentralized regulation through National Regulatory Authorities, combined with cooperative networks and supervision by the Commission, is clumsy. A centralized, EUwide regulatory structure would have two benefits. First, the Community regulator can make decisions by considering the interests of the Community as a whole and steer the industry in a coordinated manner. Second, the Community regulator may be less prey to government capture. The functional need for European agencies is difficult to challenge, but an alternative suggestion is to leave national regulators with greater freedom to experiment with different regulatory models and use networks as a means of identifying best practices. The lack of comprehensive liberalization has led the Commission to use competition law in ways that turn it into a regulator, or a regulators regulator, substituting its judgment for the perceived inaction at national level. In the vast majority of the competition cases considered in this chapter, the Commission has deployed competition law in ways different from those in other sectors, motivated by the wish to create more competitive structures. The institutional gap (that is, the lack of a single regulator for each industry) is the result of political reluctance to apply market rules completely, and causes the Commission to act beyond its powers.

An interesting economic repercussion of liberalization is the increase in Europe directed investment of many undertakings, market structures shifting from monopoly to regional oligopolies. Structural effect suggests hat the idea that sector specific regulation can give way to competition is misconceived, as markets are likely to remain highly concentrated for the foreseeable future. One of the weaknesses of EU liberalization has been the emphasis on market access for competitors at the expense of creating incentives to introduce more investment in infrastructure, which could increase market opening further. This gap also militates in favour of retaining, and strengthening, sector-specific regulation. Finally, from the perspective of politics, the role of competition principles is affected by concerns relating to the protection of citizens rights to have access to certain public services.

The Community has provided a variety of responses to accommodate public services and competition, in a manner reminiscent of its approach to the relationship between public policy and Article 81. There are instances where competition law is deemed inapplicable so as not to interfere with national welfare systems, others where an exemption from competition law obligations is granted, provided public services are offered in an efficient manner, at times by defining public service obligations as a matter of Community law, and then assuming that competitive markets are the best way to ensure citizens rights. This last approach shows how Member States have succeeded in recasting national worries about public services as matters that fall to be considered at European level.