The Role of Efficiency Claims in Antitrust Proceedings 2012

The OECD Competition Committee discussed efficiency claims in October 2012. This document includes an executive summary and a detailed summary of the discussion and the documents from the meeting: a background note by Fiorenzo Bovenzi and Anna Pisarkiewicz from the OECD Secretariat, expert notes by Hans W. Friederiszick, Frederic Michael Scherer and Helen Jenkins as well as written submissions from Australia, Chile, Colombia, European Union, Germany, Indonesia, Japan, Korea, Mexico, New Zealand, Russia, South Africa, Sweden, Switzerland, Chinese Taipei, Turkey, United Kingdom, United States and BIAC.

Economic Evidence in Merger Analysis (2011) Remedies in Merger Cases (2011) Vertical Mergers (2007) Dynamic Efficiencies in Merger Analysis (2007) Competition Policy and Efficiency Claims in Horizontal Agreements (1995)

Even if efficiencies and efficiency claims have been vigorously discussed for decades, they have rarely turned out to be decisive in competition proceedings. Still, their role in competition law has recently gained greater prominence, as witnessed by a number of recent merger decisions in different jurisdictions and the fact that efficiency claims are also more often put forward in abuse of dominance or monopolisation cases. The discussion revealed that while competition authorities in most jurisdictions examine efficiency claims in merger cases, only few of them carry out ex post evaluations of merger decisions. Moreover, an increasing number of jurisdictions acknowledges the role and examines efficiency claims also in dominance cases, even though in these cases such claims have had little practical impact so far. Overall, there was some agreement that while competition analysis should duly take into account the assessment of efficiencies, there is still much to be resolved with respect to how such assessments should be carried out.

DAF/COMP(2012)23

5

By the Secretariat

Several key points emerge from the discussion at the roundtable, the background paper and the

(1) Welfare economics points to various sources of efficiency gains. The conflict between allocative

vs. productive efficiency and static vs. dynamic efficiency is addressed by the Williamson trade-

off, which in an economic downturn faces several constraints.

The sources of efficiency examined in economic welfare analysis are static (allocative, productive) or dynamic. The underlying rationale for mergers can be the possibility of achieving efficiency gains. Thus, most merger assessments will discuss productive and/or dynamic efficiency. Practices examined as abuse of a dominant position, while potentially exclusionary, may also have such efficiency benefits (for example, tying or bundling practices).

The tension underlying competition policy is effectively between allocative vs. productive efficiency and static vs. dynamic efficiency. The welfare standard in use in a specific competition regime also affects whether certain types of efficiency gains are more easily accepted in practice. Rivalry and competitive markets result in pricing closely linked to underlying costs (allocative efficiency), to the benefit of consumer welfare. Thus, the intervention by competition authorities is to ensure that allocative efficiency is achieved and that firms do not earn excessive returns through exclusionary practices or mergers that hamper rivalry. Still, some mergers may yield lower-cost or higher-quality outputs (productive or dynamic efficiency), to the benefit of total welfare. One of the difficulties in competition assessment is to weigh consumer and producer welfare in a balanced way, so that efficiency benefits can be accepted when they increase producer welfare.

This conflict was addressed by Oliver Williamson, who proposed a total welfare approach. He argued that, if significant efficiencies occur, gains from efficiency tend to outweigh losses in consumer welfare, which provides strong theoretical evidence in favour of mergers. Professor Scherer expressed doubts with regard to the current validity of the Williamson trade-off, which are motivated by the economic downturn, in particular the widespread unemployment and Keynesian liquidity trap. Under non-crisis conditions, presuming full employment, resources released through merger-based efficiencies enable social gains by releasing resources that are used in other economic sectors to provide goods and services that benefit consumers. However, with high levels of unemployment, those resources may leak and efficiencies may not provide a consumer benefit. Second, the Williamson trade-off assumes that Say's law operates. That is, in the first instance, because of the monopoly power achieved through the merger, the price is raised and the consumer surplus is converted into higher profits or producer surplus. Next, those profits are assumed to re-circulate into effective demand for additional investment goods or through incremental consumer demand when distributed to shareholders. But in a Keynesian liquidity

they choose not to invest. Finally, there is some concern regarding a transfer of what before a merger was consumers' surplus to producers surplus and ultimately to the merging company's

DAF/COMP(2012)23

6

shareholders. If one assumes after Alfred Marshall that the marginal utility of money diminishes as wealth increases, a transfer to predominantly wealthy shareholders constitutes a redistribution of income from average-wealth consumers to rich consumers with lower marginal utilities. This implies a welfare loss, hence the consumer welfare standard would need to be favoured over the total welfare standard.

The consumer welfare standard might also be better suited to the institutional setting within which competition policy operates. The argument in favour of consumer welfare is that under

equally represented in comparison imbalance. Further, consumer welfare can help achieve optimum or second-best equilibrium with regard to the most welfare-enhancing combination of firms in a merger.

(2) Even though not all jurisdictions have introduced efficiency claims, such claims have been

progressively playing a more significant role in competition analysis. Competition authorities

that allow for efficiency claims in their antitrust proceedings have developed various approaches

with regard to procedure and substantive provisions.

Although a substantial number of jurisdictions have explicitly recognised efficiency claims in mergers, the insertion of efficiency-related provisions is not yet a common practice. Even if some legal systems do not foresee an efficiency defence, efficiency considerations may play a practical role in assessing individual cases due to specific provisions inherent to national competition acts. For instance, in Germany there is some scope for the consideration of efficiencies in the context of the balancing clause or under the ministerial authorisation. Next, the SIEC test will further promote a more effects-based assessment of merger cases and more efficiency claims seem likely. Another reason why the legislator may implicitly renounce the introduction of efficiency defence is that only the most anti-competitive mergers are prohibited (e.g. in Switzerland).

In jurisdictions where efficiency claims are introduced, several issues emerge. First, with regard to procedure, there may be an asymmetry in the analysis of efficiencies and anti-competitive effects. An efficiency defence might be a part of a two-stage process, where first there is a finding that a merger is anti-competitive, and second it is examined whether it can be justified on efficiency grounds. In other words, efficiencies are considered as a counterbalance to anti-competitive effects. There is a risk here that efficiency analysis becomes unnecessary if there are no anti-competitive effects. In an efficiency rebuttal, instead, evidence on efficiency claims forms part of a holistic analysis of positive and negative effects of the merger under consideration.

Second, the merging parties upon which the burden of proving efficiencies is placed may have a difficult job in making their case. This originates from the difficulty to substantiate efficiency claims with clear-

The next concern regards the crediting of fixed-cost efficiencies in mergers. In evaluating dominance of firms many jurisdictions apply the LRAIC or the AAC test that incorporates fixed-cost elements. In mergers, evaluation of variable costs is favoured. Thus, it is necessary to examine retrospectively the fixed-cost efficiencies in the approved mergers. This would allow probing the reason why agencies do not accept that all costs are marginal in the long run and why they offer different treatment to fixed-cost and marginal-cost efficiencies.

Other concern regards the treatment of dynamic efficiencies. The literature provides many examples supporting the view that dynamic efficiencies have a considerably greater potential to

DAF/COMP(2012)23

7

benefit consumers than static efficiencies. In the US, for example, dynamic efficiencies have been progressively taken into account, as the Genzyme/Novazyme case exemplifies.

Another issue concerns the merger specificity requirement. When taken literally, it would prevent the acceptance of an efficiency defence in many cases. Theoretically, parties could nearly always enter into contractual relations to combine resources and extract synergies. In some jurisdictions (e.g. Australia) the competition authority does not consider whether a conduct is necessary to achieve the likely net public benefits, and consideration of any alternative solutions is relevant when those alternatives are likely in the foreseeable future, absent the proposed conduct. Hence, the conduct is authorised if: i) those other means are not likely to occur in the counterfactual because of lack of market incentives or other reasons; ii) the conduct would produce efficiencies as compared to the counterfactual.

Efficiencies can also be taken into account at different stages of procedure. In the United Kingdom some of the efficiency claims made with respect to mergers are accepted either as part of the analysis of competitive effects or at the remedies stage in the form of relevant consumer benefits. In a number of cases when efficiencies were clear-cut, they were given considerable weight as part of the remedies package.

Finally, in some cases efficiency considerations or other objective justifications have allowed the clearance of mergers to monopoly or near-monopoly. For instance, competition authority in Colombia agreed to a merger between six companies, which attained a market share of 85.7 per cent. The decision was based on: operational model proposed by the firms, consideration that prices were regulated by a public authority, and the pursuit of the national economic interest. In New Zealand the competition authority also agreed to a merger to monopoly based on productive efficiencies and efficiencies linked with economies of scale and cost savings.

(3) There is a variety of tools and techniques to quantitatively assess whether the claimed efficiencies

will materialise after the merger.

Competition authorities and merging parties recognise that efficiency claims have to be evidence-based and quantifiable. For competition authorities, the evaluation of anticipated efficiencies is a difficult task since the type of efficiencies claimed and the role of such claims in the assessment of different antitrust cases vary significantly. It is widely acknowledged that it is often difficult to obtain reliable and adequate data which is crucial to estimate sound economic models. Further, there is a risk that economists provide parties with different models that come to different conclusions. This, however, should not imply that efficiency considerations are not given appropriate weight in antitrust assessment. As competition authorities become more familiar with measuring and verifying efficiencies, they become more likely to give such claims more weight in their proceedings.

Contributions submitted to the Secretariat point to different techniques that may help understand the likely benefits of corporate arrangements. Professor Scherer indicates in his paper that impartial observers with substantial experience in the relevant industry are best suited to assess claims that mergers will yield significant efficiencies. Dr. Jenkins specifies that cross-sectional and panel modelling of performance analysis using techniques such as data envelopment analysis (DEA) can be helpful to identify and quantify efficiency gains in antitrust proceedings. It can handle multiple inputs and outputs that cannot be reduced to a single input or a single output measure. The DEA measures efficiency by reference to an efficiency frontier, which is constructed as linear combinations of efficient companies. The method assumes that two or more companies can be combined to form a virtual company, with composite costs and outputs. The

DAF/COMP(2012)23

8

actual companies are compared with these virtual companies. If another actual or virtual company or their combination achieves the same output as the actual company at a lower cost, the actual company is judged to be inefficient. The DEA also informs whether the claimed effect is merger specific. Consequently, the method gives the competition authority more comfort about the verifiability of the claimed efficiency.

Compensating Marginal Cost Reduction (CMCR) is another practical tool that might serve to assess efficiencies. It has already been applied in Sweden in the context of merger assessment. Its underlying idea is that information about current margins and diversion ratios is enough to calculate the marginal cost efficiencies required to offset the increase in market power resulting from a merger. The advantage of the CMCR is that it only requires data on pre-merger mark-ups, diversion ratios and the exact marginal-cost reductions. Hence, there is no need to estimate pass-through or cost reductions to consumers. Nor is there a need to estimate how non-merging competitors would react to price changes.

(4) Ex-post evaluation of efficiencies enables competition authorities to examine performance of mergers and to improve their antitrust analysis. Some of its results can be used by the merging

parties to validate their claims during an investigation.

Evaluation indicates under what conditions and in the presence of which variables efficiency

claims are likely to be credible and eventually occur in the market. Hence, it allows for a

quantitative assessment of claimed efficiencies. Benchmark tests are another practical outcome

of the evaluation exercise. For instance, in the US evaluation has allowed to elaborate a method

by which a quality metric can be devised and applied to assess ex-post effects of a merger in the

healthcare sector.

Contributions to the Roundtable show that ex-post evaluations of mergers provide ambiguous conclusions with regard to the achievement of claimed efficiencies. Submissions from Japan and the US show that mergers do not always improve performance of firms. The background paper analyses a wide range of studies and comes to the conclusion that this may be due to the fact that there are many case- and sector-specific factors which influence the outcome. The reliability and effects of the ex-post examination of consummated mergers depend on dataset and variables examined.

(5) Questions regarding efficiency claims are also pertinent in dominance cases, in particular in the

IT sector, as there is less guidance from legislation and case law than in mergers. Further

analysis on how a more integrated approach to efficiencies and anti-competitive effects can be

implemented in dominance cases is required.

While in many jurisdictions (Canada, EU, US) legal provisions on abuse of a dominant position make no explicit allowance for an argument based on efficiency gains, legislation in some countries (Mexico, Republic of South Africa, Turkey) clearly allows dominant firms to bring forward efficiency claims. Despite lack of an explicit provision, a possibility to justify potentially anti-competitive conduct on efficiency grounds has been recognised by the EU and the US courts, as well as in soft-law instruments of the European Commission. However, the review of Article 102 TFEU decisions shows that efficiency defences are not invoked in majority of cases. In their paper Friederiszick and Gratz assert that since 2009 in 42 per cent of cases efficiencies or another objective justifications were put forward, in particular in the IT sector. Given the growing number of cases in this sector, it is recognised that a sound regulatory environment is crucial to handle efficiency considerations appropriately.

DAF/COMP(2012)23

9

Moreover, business and antitrust perspectives on potentially anti-competitive conduct differ. Empirical analysis of motives behind low price strategies, undertaken by Friederiszick and Gratz, shows that pricing below average variable cost is a common practice. Business managers consider aggressive pricing to be less advisable for a leading company in a growing market, which is either due to the fact that such practice is regarded as less effective for larger firms or because of well-understood consequences of violating antitrust rules. With regard to procedure, efficiencies in dominance cases are considered either as a defence or as a factor in the overall analysis of the competitive impact of a conduct in question. The first scenario is a two-stage analysis, in which the existence of abuse is established in the first place. Second, it is assessed whether claimed efficiencies outweigh potential anti-competitive effects. The second approach implies that efficiencies are an integral factor of the overall assessment. Hence, they are more intertwined with anti-competitive concerns. As a consequence, the prohibition of the potentially abusive conduct is not applied when efficiencies outweigh anti-competitive effects. The background paper notes that practical difficulties may be the main reason why competition authorities are tempted to consider efficiencies as a defence rather than as one of the factors in an overall assessment of abuse. Still, in some jurisdictions (Turkey) a balancing test is carried out. Hence, further analysis on how a more integrated approach to efficiencies and anti-competitive effects can be implemented in dominance cases is required.

! ! "#$#%&#'!()!*+,(!! ! -./0/'12/!3144#/0.)!0.!51&40$6/07.!

MONOPOLIZATION AND ABUSE OF DOMINANCE:

WHY EUROPE IS DIFFERENT

By Eleanor M. Fox!

The U.S. law of monopolization and the EU law of abuse of dominance share some

common ground. The projects for convergence, however, have tended to obscure some basic

differences. Each set of laws grows from its own roots and lives in its own “house” of

institutions and value sets. Convergence is more apparent at the agency level, while

distinctiveness is unmistakable at the level of the highest court of each jurisdiction. More than

twenty-five years ago, there was a significant gap at the high court level between the U.S. and

the EU law governing dominance. Over this quarter century, EU law has moved toward more

appreciation of outcome-focused economics while preserving other Community perspectives,

values, and objectives. Nonetheless, perhaps surprisingly, the size of the gap remains

approximately the same, due to the significant movement of U.S. law in the direction of

nonintervention. This article identifies precisely the points of divergence, in the interests of

knowledge and awareness, while supporting the projects of convergence, in the interests of

nurturing a sympathy of systems.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Walter J. Derenberg Professor of Trade Regulation, New York University School of Law.

AUTHOR’S NOTE: I thank Mathieu Relange for his excellent research assistance and Damien

Gerard for his helpful comments.

2 !

KEY WORDS: antitrust, abuse of dominance, monopolization, convergence, divergence,

comparative competition law

I. INTRODUCTION

There is growing antitrust literature on convergence. Much of the literature argues that in this

complex world with multitudinous global transactions we need common standards of law and,

indeed, that common soft norms can become international standards that the rest of the world

should adopt. In practice, the law of the United States and of the European Union (EU), or some

combination of it, is taken to be the developed world model.

It has long been recognized that discovering common EU/U.S. standards in the law on

abuse of dominance and the law on monopolization would be a challenge. But the challenge has

been accepted. Thus, work proceeds in the International Competition Network on particular sets

of conduct that might amount to abuse of dominance or monopolization, such as exclusive

dealing and predatory pricing, in an effort to derive recommended principles for the world.

This article argues that the efforts at convergence through derived international standards

tends to obscure fundamental differences and may thereby detract from a deeper comparative

understanding of the law. The efforts presume an ease of and incentive toward horizontal

accommodation by each jurisdiction to the other free from loyalty to its own system, and thus

they discount the reality that each system answers to its own drummer. The EU law on

exclusionary restraints must be derived from EU Treaty law and case law thereunder, not the law

of the United States, even though it may well be informed by the law of the United States, and

vice versa. This article examines rooted differences and explores their implications for

convergence. This inquiry does not imply any questioning of the projects for convergence by the

3 !

International Competition Network and others. Quite to the contrary; the projects for

convergence play a vital role in maximizing the sympathy of systems in the world.

This article will 1) briefly reflect on the different context and heritage of EU competition

law and U.S. antitrust law, including institutional design and the relationship between the

competition authority and the highest court, 2) pair cases from each side of the ocean to elucidate

differences in application of abuse of dominance–monopolization law, and 3) examine the reach

of EU abuse law where U.S. monopolization law has no counterpart, namely, affirmative duties

combined with settlement commitments to make markets work better, and prohibitions of

anticompetitive state grants and measures. Finally, the article will place the divergences in

context.

In 1986 I wrote an article comparing U.S. monopolization opinions and European abuse

of dominance judgments at the time.1 The article identified various divergences. Now, more

than a quarter of a century later, the roots of the divergences remain, but EU abuse law has

moved in the direction of U.S. monopolization law in 1986 by reason of its appreciation of

economic effects, and U.S. monopolization law has moved toward greater abstention. Thus,

while each body of law has evolved along its own spiral, the size of the gap between the two

remains approximately the same.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!1 Eleanor Fox, Monopolization and Dominance in the United States and the European

Community—Efficiency, Opportunity, and Fairness, 61 NOTRE DAME L. REV. 981 (1986).

4 !

II. THE TWO DIFFERENT HOUSES IN WHICH THE COMPETITION LAWS ARE

SITUATED

The U.S. house of antitrust is, structurally, the less complicated one. I shall describe it

first.

Famously, the Sherman Antitrust Act was adopted in 1890. Section 2 states that no

person shall monopolize or attempt or conspire to monopolize. The statute is enforceable by the

Department of Justice, by injured private parties, who may be awarded treble damages, and by

the states of the United States for their residents. It is enforceable also by the Federal Trade

Commission under its authority to prevent unfair methods of competition. The substantive scope

of the law has expanded and contracted overtime. For many years it was interpreted to help the

underdog against great concentrations of power. Today, the law has been stripped of its equity

and social policy content and is enforced in the name of consumer welfare and efficiency. The

U.S. monopoly law never had state anticompetitive acts significantly within its sights.

U.S. monopolization cases are litigated in the courts, or in the Federal Trade Commission

and the courts. The two federal antitrust agencies are largely independent; they can take or

decide not to take action, almost always without interference from higher governmental

authorities. In interpreting the law, they must apply the holdings of the Supreme Court of the

United States. The Supreme Court has decided a number of monopolization cases in the recent

past and has provided a decisional framework that puts much stress on a ground rule of “letting

markets work” on their own and thus minimal government intervention. The Court has

telegraphed its concern that if dominant firms are encumbered with duties to competitors, the

imposition of these duties will undercut incentives to invest and invent. The contemporary

Supreme Court repeatedly expresses concern with false positives, protecting competitors from

5 !

competition, and chilling low-price competition, and it expresses the institutional concern that

antitrust rules will involve the courts in supervisory duties that strain their capabilities and

amount to regulation.

The U.S. antitrust laws were adopted in a period of industrial revolution when big

business was consolidating and a few famous tycoons (dubbed Robber Barons) were

accumulating wealth and power at the expense of the “little guy.” In the last thirty years, the

Supreme Court has delinked antitrust from these origins and history. The antitrust case law of

the 21st century generally assumes that markets are vibrant and that if the little guys do not

succeed it is because they are inefficient; they must try harder. The law is not concerned with

fairness or level playing fields.

In the European Union, the competition law dates back to 1957 with the signing of the

Treaty Establishing the European Economic Community. The provision against abuse of

dominance is now numbered Article 102 of the Treaty on the Functioning of the European Union

(TFEU), a successor treaty. The Treaty Establishing the European Economic Community was

designed to establish one common market in order to anchor peace in Europe. In many ways the

Treaty tries to establish a level playing field, meaning that firms should be able to compete on

their merits and not be fenced out by power, privilege, or favoritism. The competition provisions

embody this principle. At the time the Treaty was adopted, most of the Member States (six at the

start) harbored privileged state-owned enterprises, gave preferences to their own nationals, and

maintained high trade barriers to protect “their” firms. The Treaty was designed to tear down

and keep down the barriers and preferences. This is still a central motivation of the Treaty. EU

law today is very much linked to the economic problems that gave birth to the European

Community and that still have relevance.

6 !

The competition law of the Treaty prohibits abuse of dominance and anticompetitive

agreements. It does so in many more words than does U.S. law, more along lines of civil than

common law. The very language of the Treaty in its prohibition of abuse of dominance2 includes

language of fairness and appears to express both equity and efficiency motivations.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!2 TFEU Article 102 provides:

Any abuse by one or more undertakings of a dominant position within the internal market

or in a substantial part of it shall be prohibited as incompatible with the internal market in

so far as it may affect trade between Member States.

Such abuse may, in particular, consist in:

(a) directly or indirectly imposing unfair purchase or selling prices or other unfair trading

conditions;

(b) limiting production, markets or technical development to the prejudice of consumers;

(c) applying dissimilar conditions to equivalent transactions with other trading parties,

thereby placing them at a competitive disadvantage;

(d) 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.

7 !

The Treaty also includes provisions that anticipate and counter abuses of dominance

facilitated by a Member State. First, TFEU Article 106(1) provides that, as to public

undertakings and those to which exclusive privileges are granted, Member States shall “neither

enact nor maintain in force any measure contrary to the rules contained in the [EU] Treaties, in

particular [the antidiscrimination and competition provisions].” As a practical matter this means

that, in many cases, state-granted exclusive rights cannot be enforced to preempt a competitive

market. Second, Article 4(3) of the companion Treaty on European Union imposes on the

Member States a duty of “sincere cooperation.” The Member States must “facilitate the

achievement of the Union’s tasks and refrain from any measure that that could jeopardize the

attainment of the Union’s objectives.” This means that Member State laws that enlist firms to

behave in prohibited anticompetitive ways must be “disapplied.”

This holistic conception strikes a different balance and creates a different relationship

between antitrust, regulation, public policy, and the state in the EU than in the United States.

In Europe, the competition authority is the Directorate General for Competition of the

European Commission, which is supervised by the commissioner who is entrusted with the

competition portfolio. The directorate typically undertakes market studies as well as

enforcement of antitrust (abuse of dominance and agreements), state aids, and merger law. It

confers with the European Parliament, and it responds to and incorporates the initiatives of the

European Commission as articulated by its president. It typically studies sectors in order to

identify market problems in areas of priority for the EU, as energy now is, and it follows up with

cases to help open up and integrate the market, as it has done and is doing in energy. The

European Commission is currently focusing also on telecommunications, postal services, the

knowledge-intensive (including digital) sector, and the pharmaceutical sector. It takes as a

8 !

priority facilitating connections to network industries across the EU. Not surprisingly, the

annual report of the European Commission on competition policy is quite different from the

annual report of the U.S. Department of Justice on antitrust. The former reflects the holistic

policy of the EU. The latter is first and foremost focused on the size of the fines and number and

length of the jail terms the Antitrust Division of the Department of Justice has imposed on

members of cartels by settlement agreement or through the courts.

EU competition law is enforced also by the Member States. Moreover, private parties

have a right under European law to compensation for any violation of the Treaty that harms

them; but they must sue under the procedural vehicles of the Member States, most of which

embody few incentives and some significant disincentives to private actions. A proposed

European framework directive would set minimum standards for the Member States and increase

the viability of private actions.

The European Court of Justice has developed a large body of case law on abuse of

dominance. The law establishes that dominant firms have a special responsibility not to erect or

maintain barriers that frustrate the access of nondominant firms to markets. Some recent

judgments reveal a concern for competitiveness, efficiency, and consumer welfare,3 but all

within the framework of general EU policy. Even the usage of the words “consumer welfare”

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!3 Case C-209/10, Post Danmark A/S v Konkurrencerådet, judgment of Mar. 27, 2012 (not yet

officially reported).

9 !

and the phrase “to protect consumers” do not correspond with the usage of the words in the U.S.

cases.4

In the EU the investigated party may settle the matter, without admitting liability, by

offering commitments that meet the European Commission’s concerns. While negotiated

outcomes are also available in the United States,5 U.S. enforcement of section 2 of the Sherman

Act and the somewhat more flexible Federal Trade Commission counterpart are not

overshadowed by settlements; in Europe commitment decisions are now the usual way of

resolving Article 102 violations alleged by the European Commission. Thus, a substantial

number of Article 102 outcomes derive from commitment decisions, resulting in a soft

regulatory law.

Who has control over the agenda for competition law and policy? Is there a coherent

competition law and policy for the United States? For the EU?

In the United States, “control” over the national antitrust agenda is loose. There are

multiple sources of antitrust enforcement. The Department of Justice can sue in the courts and

write amicus briefs in private cases. The Federal Trade Commission can bring proceedings,

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!4 See Case C-280/08P, Deutsche Telekom AG v. Comm’n, 2010 E.C.R. I-9555, paras. 179–182

(enforcement against a price squeeze may cause retail prices to rise to alleviate the squeeze, but

this is for the greater good of the functioning of the market to the benefit of all players).

5 See Press Release, Federal Trade Comm’n, Google Agrees to Change its Business Practices to

Resolve FTC Competition Concerns in the Markets for Devices Like Smart Phones, Games and

Tablets, and in Online Search (Jan. 3, 2013), available at

http://www.ftc.gov/opa/2013/01/google.shtm.

10 !

which may be appealed to the courts. But most antitrust cases in the United States are brought

by private parties. Indeed private cases, almost entirely, have shaped the Supreme Court

antitrust docket for many years. Antitrust policy is shaped by the agencies, but a preponderant

share of U.S. antitrust activity relates to enforcement of the law, not studies and advocacy. U.S.

antitrust agency policy prioritizes cartel cases, health care cases, and other restraints in the

pharmaceutical sector. Concerns and studies (especially consumer concerns identified by the

Federal Trade Commission) may result in policy statements and theories for antitrust lawsuits

and, in some cases, proposals for legislation. Upon the advent of a new presidential

administration, U.S. antitrust policy tends to involve a rhetorical resetting of the stage, either

promising more aggressive enforcement or vowing more trust in markets.

In Europe, control of the competition agenda is tighter. Common policy for the EU

(broad policy, of which competition is but one piece) is articulated at the European Commission

level with significant input from the Member States. In antitrust adjudication, there are more

direct lines to the top than in the United States—for example, from the Commission through the

General Court to the Court of Justice, and from national courts (which are bound by Commission

interpretations absent conflicting direction from above) directly to the Court of Justice. There is

no “bubbling up” of issues in lower courts, conflicts of circuits, and eventual, if ever, high court

resolution, as in the United States.

III. CASE PAIRS OF OPPOSITES

The most startling literal differences between U.S. and EU law on abuse of dominance

have been well recorded. They are most obvious in refusal to deal and margin-squeeze cases, but

11 !

are not limited to these cases. The contemporary U.S. Supreme Court opinions exude a

philosophy or perspective that permeates U.S. antitrust analysis.

The Supreme Court’s opinion in Verizon v. Trinko6 set the blue print for this perspective.

In Trinko, Bell Atlantic (later acquired by Verizon), was the incumbent telephone service

provider in the northeastern United States. Bell Atlantic owned the elements of the local loop—

bottleneck elements connecting long distance service to the local market. For many years, it

operated under conditions of legal monopoly in local service areas. Technology changed,

enabling local service competition. Congress passed the 1996 Telecommunications Act, easing

the way for new entry into local service and requiring incumbents to give their rivals

nondiscriminatory access to the local loop. Bell Atlantic nominally gave its new rivals access to

the local loop and undertook to serve them as required by the new Act, but it adopted a simple

plan to dissuade its customers from migrating to the feistier new rivals; it disrupted the rivals’

service. The rivals complained to the Federal Communications Commission, which agreed with

the complaints, penalized Bell Atlantic, and ordered Bell Atlantic to compensate the abused

rivals. Customers of the rivals sued, claiming that Bell Atlantic’s disruptive conduct was also an

antitrust violation, which it might well have been because the 1996 Act expressly does not

preempt the antitrust laws.

The case came to the Supreme Court on Bell Atlantic’s motion to dismiss the pleadings.

The Court first characterized the exclusionary practice as a refusal to deal (although Bell Atlantic

had never refused to deal), teeing up a rhetoric that would change American “monopolization”

discourse. The Court said that firms must have freedom not to deal, lest they be handicapped by

“sharing” duties which would undermine their incentives to invest and invent. While there may

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!6 Verizon Commc’ns v. Law Offices of Curtis V. Trinko, 540 U.S. 398 (2004).

12 !

be exceptional circumstances that justify a duty, the Court rejected the argument that the rival

telecoms had a right to fair and nondiscriminatory connective service to the local loop. It

declared that the Supreme Court has never recognized an essential facility doctrine, and even if

there is one it did not cover this case because there had been some dealing and a regulatory

agency had power to order dealing.

Extolling the virtues of freedom not to deal even for a dominant firm, the Trinko Court

stated that antitrust does not impose affirmative duties just because “some other approach might

yield greater competition.”7

In Telekomunikacja Polska, on virtually the same facts as Trinko, the European

Commission held that the strategy by the incumbent to limit access to its telephone network

distorted competition and violated TFEU Article 102.8 Under the Polish Telecommunications

Law, which was based on the EU regulatory framework, the Polish telecommmunications

operator was under the obligation to allow remunerated network access to alternative operators

willing to offer broadband Internet access to end-users; thus, to the new competitors on the retail

market. However, Telekomunikacja Polska used delaying tactics in negotiations with the new

rivals, failed to provide them accurate information or provided inaccurate information, and

limited their access to the subscriber lines. The Polish National Regulatory Authority found that

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!7 Id. at 415–16.

8 Case COMP/39.525, Telekomunikacja Polska, 2011 O.J. (C 324) 7. In the European cases the

regulatory authority is at Member State rather than European level. This might have led the

European authorities either to give more deference to the regulation, on subsidiarity grounds, or

to give less deference on policy coherence grounds. It chose the latter.

13 !

Telekomunikacja Polska’s conduct violated its Polish statutory obligation. The European

Commission also opened proceedings. It found that Telekomunikacja Polska’s practices were

part of a strategy to prevent or at least delay the entry of competitors into the retail market and

that they constituted an abuse of dominance within TFEU Article 102.

A few years after Trinko, the U.S. Supreme Court decided Pacific Bell Telephone Co. v.

linkLine Communications, Inc.9 Pacific Bell was the incumbent telephone service provider in a

West Coast region of the United States. It provided local telephone service and, as the historical

incumbent, owned the elements of the local loop. It also supplied digital subscriber line (DSL)

service—for fast computer access through phone lines—to Internet service providers (ISPs) at

wholesale and sold DSL service at retail. Under Federal Communications Commission

regulations, Pacific Bell had the duty to supply service at wholesale to DSL providers. During

some periods it charged its DSL retail customers less than it charged the ISPs at wholesale. The

price of wholesale service was regulated; Pacific Bell proposed the rate and the Federal

Communications Commission approved it, as the Commission must do for all filed rates unless

they are “unjust and unreasonable.”

The ISPs sued for a price squeeze, unlawful under section 2 of the Sherman Act. Pacific

Bell moved for judgment on the pleadings, arguing that, after Trinko, it had no antitrust duty to

deal and therefore no duty to avoid a price squeeze. The lower courts declined to dismiss the

case, acknowledging that Pacific Bell had no antitrust duty to deal but holding that Trinko was

not otherwise relevant because it was not a price squeeze case. The Supreme Court reversed. It

held that Trinko was controlling. In Trinko, it said, plaintiffs claimed that Bell Atlantic did not

give them good enough service; here plaintiffs claimed that the incumbent did not give them a

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!9 555 U.S. 438 (2009).

14 !

good enough price. It is all the same. Since Pacific Bell had no antitrust duty to deal at all, it

had no duty to sell at a price advantageous to the rivals. Accordingly, charging a high wholesale

price was not illegal. Furthermore, the retail price was not a predatory price (below cost en route

to monopoly).10 Therefore the retail price was not illegal. If both prices are legal, said the Court,

their sum cannot be illegal.

Virtually the same facts arose in Europe in Deutsche Telekom.11 The German regulator

regulated Deutsche Telekom’s prices, but Deutsche Telekom had sufficient scope to cause an

increase in its retail prices by applying to the regulator for the right to charge higher prices and

obtaining authorization to do so. Deutsche Telekom was not exonerated by the fact that the

German regulator encouraged the wholesale-retail price relationship that imposed the margin

squeeze and opined that the squeezed competitors should simply cross-subsidize to survive.12

The European Court of Justice held the margin squeeze illegal.

Deutsche Telekom was followed by TeliaSonera,13 in which a Swedish court made a

preliminary reference to the European Court of Justice seeking to be advised of the governing !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!10 Even if Pacific Bell had charged a predatory price, the Court was doubtful that the plaintiffs

could win their case. 555 U.S. at 456–57 (“Even if the amended complaint is further amended to

add a Brooke Group [predatory pricing] claim, it may not survive a motion to dismiss. For if

[Pacific Bell] can bankrupt the plaintiffs by refusing to deal altogether, the plaintiffs must

demonstrate why the law prevents [Pacific Bell] from putting them out of business by pricing

them out of the market.”).

11 Case C-280/08P, Deutsche Telekom AG v. Comm’n, 2010 E.C.R. I-9555.

12 Id. paras. 84–86.

13 Case C-52/09, Konkurrensverket v. TeliaSonera Sverige AB, 2011 E.C.R. 527.

15 !

principles of the law on margin squeeze. The Court advised as follows, invoking the oft-cited

duty of special responsibility of a dominant firm under EU law and liberally quoting from

Deutsche Telekom:

The function of those [competition] rules is precisely to prevent competition from being

distorted to the detriment of the public interest, individual undertakings and consumers,

thereby ensuring the well-being of the European Union . . . .14

* * *

Accordingly, Article 102 TFEU must be interpreted as referring not only to practices

which may cause damage to consumers directly . . . but also to those which are detrimental

to them through their impact on competition . . . . [I]t remains the case that, in accordance

with settled case-law, an undertaking which holds a dominant position has a special

responsibility not to allow its conduct to impair genuine undistorted competition in the

internal market . . . .15

* * *

It must moreover be made clear that since the unfairness, within the meaning of Article

102 TFEU, of such a pricing practice is linked to the very existence of the margin squeeze

and not to its precise spread, it is in no way necessary to establish that the wholesale prices

for [advanced DSL] input services to operators or the retail prices for broadband

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!14 Id. para. 22.

15 Id. para. 24.

16 !

connection services to end users are in themselves abusive on account of their excessive or

predatory nature . . . .16

Thus, we find directly conflicting analysis and directly conflicting holdings.

Fidelity rebate cases are lurking in the background as another potential subject for

comparison and contrast. In Tomra,17 the Court of Justice of the EU held that individualized,

targeted fidelity rebates by a dominant firm going back to day one on all purchases and tailored

to all or most of a customer’s requirements are likely to strengthen the firm’s dominant position

by distorting competition in violation of Article 102. The Court said:

“[I]t is . . . unnecessary to undertake an analy[sis] of the actual effects of the rebates on

competition, given that, for the purposes of establishing an infringement of Article 102

TFEU, it is sufficient to demonstrate that the conduct is capable of having an effect on

competition . . . .”18

The Court of Justice rejected Tomra’s defenses that it never priced below cost, that it did

not sacrifice profits, that its rebate plan had not harmed competition, and that competitors had a

sufficient open market to realize scale economics. The Court said that the “effective price for

the last units” was “very low because of the ‘suction effect’.”19 The “loyalty mechanism” was

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!16 Id. para. 34 (emphasis added).

17 Case C-549/10P, Tomra v. Comm’n, April 19, 2012 (not yet published).

18 Id. para. 79. See also id. para. 68.

19 Id. para. 78.

17 !

therefore able to “drive out [Tomra’s] competitors by means of the suction to itself of the

contestable part of the demand.”20

As to predatory pricing, the U.S. Supreme Court in Brooke Group21 articulated a

demanding rule requiring the plaintiff to prove not only below cost pricing but also a probable

recoupment scenario, meaning that a court will analyze the structure of the market and

incentives of the players to determine if the predator will probably be able to raise prices above

prepredation prices high enough and long enough to at least recoup its losses in the below-cost

sales. The European Court of Justice specifically considered and rejected the Brooke Group

requirement of probable recoupment.22

The U.S. Supreme Court has not yet spoken on fidelity rebates. Currently the circuits are

split between a rule of law based on exclusionary foreclosure subject to justification23 and a rule

that would protect the freedom of low pricing unless the price is predatory.24 Trinko and

linkLine may be the handwriting on the wall.

The recent Google investigations on both sides of the Atlantic provide another pair in

point, although not documented in judicial decisions. Google occupies about seventy percent of

the computer search market in the United States and more than eighty percent in Europe.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!20 Id. para. 79.

21 Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209 (1993).

22 Case C-62/86, AKZO Chemie BV v. Comm’n, 1991 E.C.R. I-3359; Case C-202/07P,

France Télécom SA v. Comm’n (Wanadoo), 2009 E.C.R. I-2369.

23 See ZF Meritor, LLC v. Eaton Corp., 696 F.3d 254 (3rd Cir. 2012), cert. denied, 133 S. Ct.

2025 (2013).

24 See Cascade Health Solutions v. PeaceHealth, 515 F.3d 883 (9th Cir. 2008).

18 !

Competitors, including Microsoft, complained that Google presents search results on vertical

searches (such as maps, restaurants and travel) that give Google services a privileged position

thereby diverting business from its competitors. The U.S. Federal Trade Commission conducted

an investigation for nearly two years and ultimately closed the investigation on grounds that the

evidence “did not justify legal action.”25 The European Commission opened an investigation,

found (preliminarily) serious violations, and is, as of December 2013, considering whether to

accept settlement commitments from Google.26

Why would it be so difficult to prove a monopolization violation against Google’s

preferences to itself in the United States and not so difficult to prove an abuse violation in the

EU? The following factors are relevant. U.S. law is demanding. It requires proof that Google

has monopoly power. It also requires a legal conclusion that Google had a duty of fair dealing

with its rivals—a hard burden after Trinko and linkLine. If Google had a duty, it would require

proof that Google used its monopoly power to obtain additional power by reason of its

preferences to itself which diverted business from its rivals in each separate market, such as

maps, travel services, and restaurant guides. It would require a showing of harm to consumers.

And, finally, a plaintiff would have to blunt the assertions (and fairly widespread default

assumptions) that 1) Google is the epitome of an inventive firm (it invented its search engine and

presumptively has the right to use it for its own advantage); 2) intervention by a court will chill

Google’s incentives to invent, and a rule of law imposing a duty on Google will in general chill

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!25 Press Release, Federal Trade Comm’n, supra note 5.

26 See Joaquin Almunia, Vice President and Commissioner for Competition, Speech at European

Parliament Hearing: The Google Antitrust Case: What Is at Stake? (Oct. 1, 2013), available at

http://europa.eu/rapid/press-release_SPEECH-13-768_en.htm.

19 !

firms’ incentives to invent; and 3) more pragmatically and immediately, any court order

requiring Google to deal equally with rivals would in effect be invoking the essential facilities

doctrine (which the U.S. does not have, at least not for this circumstance) and would risk

involving the court in duties of supervision beyond its capabilities, as cautioned against in

Trinko and linkLine.

In the United States, each of these elements presents a high hurdle for a plaintiff. As to

each, the bar is lower under European abuse of dominance law. With respect to the conduct

element itself (assuming sufficient proof of dominance and acknowledging no Trinko problem

regarding duty to deal), it is possible that the Commission would satisfy European Court of

Justice standards of prima facie proof by showing that Google’s preferences to itself diverted

from its rivals substantial business that they would otherwise have won “on the merits,” thus

distorting competition.

As appears from the cases, the EU perspective on abuse of dominance at the Court of

Justice level stresses the process of competition, seeking to enable all market actors to compete

on their merits, particularly efficient and potentially efficient competitors. The U.S. law of

monopolization at the Supreme Court level stresses the costs of antitrust intervention, tending

toward per se legality in a number of situations and otherwise imposing considerable burdens on

plaintiffs to show how the particular conduct will increase market power and harm consumers

and that the finding of a violation would not compromise low prices and incentives to innovate.

20 !

IV. OTHER CASES AND INITIATIVES FOR WHICH THE UNITED STATES HAS NO

COUNTERPART

While the cases of opposites are startling, initiatives reflecting interventions in Europe

where the U.S. antitrust law simply has no counterpart are at least equally informative of basic

differences in the regimes. Two lines of authority are relevant. The first we might call

affirmative antitrust; a dominant firm may have an affirmative obligation to make a market work

better. In the second, a Member State has adopted measures that harm the market, and the

Member State’s measures violate the competition principles of the Treaties. In some cases the

Member State must disapply its own law. In other cases the grantee of exclusive rights by the

state may be stripped of the right to exercise them.

While in the “pairs of opposites” cases, U.S. advocates may claim, rightly or wrongly,

that the EU precedents coddle competitors and handicap efficiency, in some “no counterpart”

cases EU law may have the high road if judged by the standard of efficiency. U.S. law simply

does not reach so far. At least this is so regarding certain state anticompetitive acts.

A. Affirmative obligations: Making markets better

The European Commission, concerned with underperformance in the European energy

sector, undertook a sector inquiry. It determined that there was insufficient competition,

including insufficient entry, and that a major part of this problem was that both the leading

German gas supplier, E.ON, and the leading French gas supplier, GDF, had reserved their gas

transport capacity to themselves—to their own transmission subsidiaries—for a term of many

years. As a result competitors could not access pipelines to reach customers. After the inquiry,

the Commission proposed to improve the EU legislative and regulatory framework for energy

markets. It also initiated several proceedings to open the gas market. The cases alleged abuses of

21 !

dominance and failure to meet essential facility obligations by conduct including “capacity

hoarding” and “strategic underinvestment” in new capacity; that is, failing to invest in capacity

needed by the market even though the investment would have been profitable.27 The cases were

concluded by commitment decisions of the gas suppliers to open their capacity by reducing their

reservations to themselves.28

The energy sector commitment cases are an example of integral European policy

culminating in obligations of the leading market players to take action to improve market

conditions. Earlier examples include telecommunications policy, including liberalization

measures and follow-on exemption decisions. Exemption decisions allowed joint ventures of

leading telecommunications companies from different Member States (such as France and

Germany) to proceed with their planned alliance on condition that the partners agreed to

liberalize their national markets, even though it was hard to detect any negative impact of the

alliance on competition.29

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!27 See Ricardo Cardoso, Sandra Kijewski, Oliver Koch, Patrick Lindberg & Károly Nagy, The

Commission’s GDF and E.ON Gas Decisions Concerning Long-Term Capacity Bookings—Use

of Own Infrastructure as Possible Abuse under Article 102 TFEU, COMPETITION POL’Y NEWSL.,

Competition Directorate of European Commission, no. 3, 2010, at 8.

28 See Case COMP/39.402, RWE, 2009 O.J. (C 133) 10; Case COMP/39.316, Gaz de France,

2010 O.J. (C 57) 13; Case COMP/39.317, E.ON, 2010 O.J. (C 278) 9; and Case COMP/39.315,

ENI, 2010 O.J. (C 352) 8.

29 See European Commission, XXVI Report on Competition Policy 1996, pp. 36-38 (1997).

22 !

The theme of affirmative obligations also permeates judgments of the European Court of

Justice in cases of dominant firms and their special responsibilities. In AstraZeneca,30 a firm

dominant in antiulcer medicine with the product Losec, was facing the expiration of its patent

and thus the competition of generics. Among other things, AstraZeneca changed the form of

Losec from capsules to tablets with multiple microgranules, and it applied to deregister the

marketing authorizations for Losec capsules in Denmark, Sweden, and Norway. If Losec

capsules were deregistered, generic companies could no longer rely on the registrations for proof

of safety and would have to incur significant expenses to get authorization; moreover, parallel

importers were likely to lose their parallel import licenses. The Court found that the request for

deregistration constituted an abuse of dominance. So holding, based on a long string of citations,

the Court reiterated the test for abuse: The dominant firm violates the law where its conduct, by

methods other than competition on the merits, “has the effect of hindering the maintenance of the

degree of competition still existing in the market or the growth of that competition.”31

AstraZeneca argued that it had a right to request withdrawal of an authorization that it was no

longer using and noted that it was costly to maintain registrations, but, the Court held, the fact

that AstraZeneca desired the deregistration to hinder the competition of parallel importers and

generics trumped any such right.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!30 Case C-457/10P, AstraZeneca AB v. Comm’n, available at

http://curia.europa.eu/juris/document/document.jsf?text=&docid=131490&pageIndex=0&doclan

g=EN&mode=req&dir=&occ=first&part=1&cid=431737

31 Id. para. 74 (emphasis added).

23 !

B. Invalidating state measures

The Treaty on European Union (TEU) and the TFEU (together, the Treaties) have two

sets of provisions that invalidate certain state anticompetitive measures under the competition

laws. The first regards grants of exclusive rights. The second regards anticompetitive legislation

that undermines the competition rules. The U.S. has no counterparts.

1. TFEU ARTICLE 106(1) The TFEU creates a category of public enterprises and

those to which Member States have granted exclusive rights. As to these firms or entities, the

Treaty imposes an obligation: Member States must not enact or maintain “any measure contrary

to the rules contained in the Treaties, in particular [the antidiscrimination and competition rules].

This obligation, contained in Article 106(1), is read in conjunction with the competition rules in

the TFEU, usually Article 102, which prohibits abuse of a dominant position. This combination

has a powerful effect. The State may not grant exclusive rights that produce an abuse of a

dominant position, and, as per Article 102, a dominant position can be abused by a contraction of

production (Article 102(b)), excessively high or low prices (Article 102(a)), or unreasonable

foreclosures (case law in general).

A number of cases illustrate the breadth of the EU law. Postal services are prominent

among them. In Corbeau,32 the European Court of Justice held that a mail delivery service with

an exclusive license could not legally prevent the entry of a private express delivery service

except to the extent that exclusivity was necessary to achieve a public mission. In Slovenská

po!ta,33 the Slovakian post office had a monopoly over traditional mail delivery. Private

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!32 Case C-320/91, Régie des Postes v. Corbeau, 1993 E.C.R. I-2533.

33 Case COMP/39.562, Slovenská po!ta, 2008 O.J. (C 322) 10, appeal pending, Case T-556/08,

Slovenská po!ta v. Comm’n.

24 !

enterprises developed an adjacent new market—hybrid mail services—that involved businesses’

sending invoices by Internet to local mail delivery offices, which would print and deliver the

invoices locally. To take advantage of this new market, the Slovakian post office adopted a

measure to extend its monopoly to the adjacent market, thus displacing the entrepreneurs who

had developed hybrid mail delivery. The adoption of this measure without justification was an

abuse of dominance.

A German public agency with the exclusive license to provide executive recruitment

services infringed the competition law by the mere exercise of its right to keep out competition,

because the agency was clearly not in a position to satisfy the demand for such services.34 In

another example, Port of Genoa, Italy was held to have infringed the law by adopting national

legislation granting a firm engaged in port services an exclusive right to organize dock work at

the port. The statute had created incentives for inefficient and anticompetitive behavior, and

Italy infringed the competition law.35

TFEU 106(1) has stunningly broad implications and may, along with Article 4(3) of

the TEU, constitute a long reach of European competition law to control anticompetitive state

measures.

2. TEU ARTICLE 4(3) The TEU is, in effect, the overarching institutional and

constitutional treaty. Article 4(3) of the TEU states that the Member States have a duty of

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!34 Case C-41/90, Höfner & Elser v. Macrotron GmbH, 1991 E.C.R. I-1979.

35 Case C-179/90, Merci Convenzionali Porto di Genova ApA v. Siderurgica Gabrielli SpA,

1991 E.C.R. I-5889, para. 19. Accord, Case C-18/88, RTT v. GB-Inno-BM, 1991 E.C.R. I-5941;

Case C-462/99, Connect Austria, 2003 E.C.R. I-5197.

25 !

“sincere cooperation” to carry out the tasks that flow from the Treaties. TEU Article 4(3)

provides that the States must “facilitate the achievement of the Union’s tasks and refrain from

any measure which could jeopardise the attainment of the Union’s objectives.” TEU Article 4(3)

is read in conjunction with the antitrust articles of the TFEU. This combination results in the

obligation of Member States not to enact or apply laws that would undermine the competition

rules, in particular by inducing or requiring market actors to act in violation of TFEU Articles

101 or 102.

Although TEU Article 4(3) has not been frequently invoked, its usage is instructive.

Perhaps the best example is the Italian matches case, Consorzio Industrie Fiammiferi.36 Italy

had a law organizing a match cartel. It required all producers of matches in Italy to join a

consortium. A minister was required to set the price for matches, and the consortium of Italian

producers was required to allocate quotas to all sellers in Italy. The process was to be overseen

by government officials. The consortium allocated quotas so as to substantially exclude German

and Swedish producers. The Italian Antitrust Authority brought proceedings and found antitrust

violations by both Italy and the Italian producers. As to Italy's responsibility, the European

Court of Justice said:

[A]lthough Articles [101 TFEU (anticompetitive agreements) and 102 TFEU (abuse of

dominance)] are, in themselves, concerned solely with the conduct of undertakings and

not with laws or regulations emanating from Member States, those articles, read in

conjunction with Article [4 TEU], which lays down a duty to cooperate, none the less

require the Member States not to introduce or maintain in force measures, even of a

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!36 Case C-198/01, Consorzio Industrie Fiammiferi v. Autorità Garante della Concorrenza e del

Mercato (Italian matches), 2003 E.C.R. I-8055.

26 !

legislative or regulatory nature, which may render ineffective the competition rules

applicable to undertakings.

The Court has held in particular that Articles [TEU 4 and TFEU] 101 are infringed “…

where a Member State requires or favours the adoption of agreements, decisions or

concerted practices contrary to Article [TFEU 101] or reinforces their effects, or where it

divests its own rules of the character of legislation by delegating to private economic

operators responsibility for taking decisions affecting the economic sphere…”

* * *

The duty to disapply national legislation which contravenes Community law applies not

only to national courts but also to all organs of the State, including administrative

authorities, which entails, if the circumstances so require, the obligation to take all

appropriate measures to enable Community law to be fully applied.37

The judgment of the European Court of Justice in Italian matches38 states a rule

directly contrary to the one applied by the U.S. Supreme Court in the celebrated case of Parker

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!37 Id. paras. 45-6 and 49 (emphasis added). See also id. para. 50 ("[The competition] rules would

be rendered less effective if, in the course of an investigation under [TFEU 101] into the conduct

of undertakings, the authority were not able to declare a national measure contrary to the

combined provisions of Articles [TEU 4] and [TFEU 101] and if, consequently, it failed to

disapply it.”).

38 The principle is limited by the EU version of the U.S. state action doctrine. See Joined Cases

C-94/04 & C-202/04, Cipolla v. Fazari and Macrino v. Meloni, 2006 E.C.R. I-11421. Italy had

adopted lawyer fee schedules based on a draft submitted by the lawyers’ association. The Court

27 !

v. Brown, where the U.S. Court refused to enjoin the operation of a California statute organizing

a public-private raisin cartel. If the case had arisen in Europe, the Italian court would have been

obliged to “disapply” the cartel-enabling law.39

If the statute were similarly anticompetitive but did not commandeer private firms to

carry out the cartel, the EU analysis would move seamlessly from competition law to internal

market trade (free movement) law.40 If Italy had a law requiring retailers of burial caskets to be

funeral directors licensed by Italy (a law not explainable except by protectionism) and some

modicum of inter-Member State commerce were involved,41 Italy would have violated the

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!declined to find a violation of TEU 4(3) and TFEU 101 for lawyer price-fixing because the fee

schedule had to be approved by the Minister of Justice before it could enter into effect.

Therefore the fee schedule was not deprived of its character as legislation by delegation to

private operators. The Court observed, however, that the Italian rules (requiring high fees) could

make it harder for out-of-state lawyers to contest the Italian market, and accordingly it held that

the Italian court was obliged to consider whether the state-sanctioned fee schedule violated EU

free movement rules.

39 See Eleanor Fox, What If Parker v. Brown Were Italian?, in 2003 Fordham Corp. L. Inst.,

International Antitrust Law & Policy (B. Hawk ed., 2004). In Parker v. Brown, the state won;

under EU law, the state would have lost. The Italian statute was disapplied rather than

preempted because the European Court has no power to void a Member State law.

40 See Cipolla, 2006 E.C.R. I-11421.

41 In the United States there is a conflict of circuits on the constitutionality of just such state laws

on the sale of caskets. Compare St. Joseph Abbey v. Castille, – F.3d – (5th Cir. 2013)

28 !

Treaties’ articles on free movement of goods, free movement of workers, and freedom of

establishment, but not competition law.

The U.S. has no counterpart to TFEU 106 or TEU 4(3). Perhaps the United States is

not so clearly in need of a counterpart because there are fewer—albeit still abundant—instances

of excessively anticompetitive state measures obviously not justified by any public interest,

along with a greater chance that the political process will work to limit such restraints. But

probably few other competition jurisdictions in the world share these characteristics. Perhaps as

a by-product of the U.S. gap, the subject of anticompetitive state acts is not on the agenda of

U.S./EU convergence and is not normally on the template of technical assistance providers and

other advisors (surely not on the syllabus of most Americans) of what a good competition law

should contain.

V. DIVERGENCES AND ASYMMETRIES IN CONTEXT

Notably, this article is not about symmetries and correspondences. A significant portion

of U.S. and EU law on abuse of dominance and monopolization corresponds. In particular,

monopolistic conduct prohibited by section 2 of the Sherman Act is likely to constitute an abuse

of dominance under TFEU Article 102, although not vice versa. A striking resemblance would

be expected, because the law purports to treat the same subject.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!(unconstitutional; no rational basis), with Powers v. Harris, 379 F.3d 1208 (10th Cir. 2004)

(constitutional; protectionism is a rational basis).

29 !

The convergences are greatest at the agency level, not the court level. Agency policy and

formulations of laws are very much influenced by technicians, especially antitrust economists the

most prominent of whom share a relatively similar training and culture across the world.42

So, too, divergences would be expected, because each set of laws grows from its own

roots. The divergences are most pronounced among the highest courts of the land. The

Competition Directorate in Europe faces a highest court that applies the Treaties’ values of

openness and access, often engages in formal legalistic analysis, and sometimes applies rules of

fairness. Accordingly, the Court’s pronouncements tend to proscribe more than the Competition

Directorate’s guidance would. The U.S. agencies face a highest court that tends to apply values

of trust in the market and deep respect for the business judgment of even dominant firms,

expecting thereby to maximize innovation, efficiency, and a notion of freedom. The Supreme

Court’s holdings proscribe less conduct than most current U.S. agency officials deem

anticompetitive.43 Thus we see more administrative-level convergence and more judicial-level

divergence.44

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!42 The homogenization of antitrust-economics and analysis may be only half good. The currently

dominant approach tends to reflect a political economy theory of nonintervention. Its default

assumptions do not describe the monopolized economies of poor developing countries. It is

likely, in most cases, to advantage those who are already advantaged.

43 There is, consequently, debate within the Federal Trade Commission as to whether and how

section 5 of the Federal Trade Commission Act should be used to fill the gap. See, e.g., In re

Robert Bosch GmbH, Docket No. C-4377, FTC File No. 121 0081,

http://www.ftc.gov/opa/2012/11/bosch.shtm (Commissioners’ separate statements); Joshua

Wright, FTC Commissioner, Recalibrating Section 5: A Response to the CPI Symposium,

30 !

Despite the subtle and not so subtle divergences, there are many on-going projects for

convergence—nudging the law (or statements of it by the agencies as applied to various conduct)

to “be” the same thing in all jurisdictions as applied to the same facts. The premise is that trade

and commerce are international and competition law is national. A crazy quilt of a hundred

different rules applied to the same transaction or conduct puts huge costs on the transactions and

the firms involved in them and creates huge inefficiencies for which both the transacting

businesses and the consuming public pay. Further, it creates hostilities between clashing

jurisdictions. If we can find, nudge, or push commonality, or even the appearance of it, we can

smooth trade and political interactions between our two jurisdictions and the world.

There is a tension between the ideals of a smoothly flowing world of commerce

following harmonious rules of law and the contextual needs and individuality of systems. But,

more important to this article, there is a reality of not insignificant divergences, based not just on

different words in decisions and judgments but different roots, histories, statutes (and Treaties),

patterns of growth and adjustment, and meanings. To appreciate and capitalize on the

commonalities is noble and constructive and can indeed help the world go 'round more smoothly.

But to “imagine away” the differences is not constructive. It is important that distinctions are

appreciated, by parties to transactions, by policymakers, by students and teachers, and by the

people in general, for purposes of information, awareness, sympathetic understanding,

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!Competition Policy International Antitrust Chronicle Nov. 2013 (2); also at

http://ftc.gov/speeches/wright.shtm.

44 See T.K.H. Cheng, Convergence and its Discontents: A Reconsideration of the Merits of

Convergence of Global Competition Law, 12 CHI. J. INT’L L. 433 (2012).

31 !

consideration of possibilities for improvement, and validation and legitimacy in any given

culture.

This article has been a small attempt to appreciate the differences.